python3 AND python2 packages.

tablib/packages/markup.py

@@ -68,7 +68,7 @@ class element:
         """Append the actual tags to content."""
 
         out = "<%s" % tag
-        for key, value in kwargs.items( ):
+        for key, value in kwargs.iteritems( ):
             if value is not None:               # when value is None that means stuff like <... checked>
                 key = key.strip('_')            # strip this so class_ will mean class, etc.
                 if key == 'http_equiv':         # special cases, maybe change _ to - overall?
@@ -156,17 +156,17 @@ class page:
 
         if mode == 'strict_html' or mode == 'html':
             self.onetags = valid_onetags
-            self.onetags += list(map( str.lower, self.onetags ))
+            self.onetags += map( string.lower, self.onetags )
             self.twotags = valid_twotags
-            self.twotags += list(map( str.lower, self.twotags ))
+            self.twotags += map( string.lower, self.twotags )
             self.deptags = deprecated_onetags + deprecated_twotags
-            self.deptags += list(map( str.lower, self.deptags ))
+            self.deptags += map( string.lower, self.deptags )
             self.mode = 'strict_html'
         elif mode == 'loose_html':
             self.onetags = valid_onetags + deprecated_onetags 
-            self.onetags += list(map( str.lower, self.onetags ))
+            self.onetags += map( string.lower, self.onetags )
             self.twotags = valid_twotags + deprecated_twotags
-            self.twotags += list(map( str.lower, self.twotags ))
+            self.twotags += map( string.lower, self.twotags )
             self.mode = mode
         elif mode == 'xml':
             if onetags and twotags:
@@ -183,7 +183,7 @@ class page:
 
     def __getattr__( self, attr ):
         if attr.startswith("__") and attr.endswith("__"):
-            raise AttributeError(attr)
+            raise AttributeError, attr
         return element( attr, case=self.case, parent=self )
 
     def __str__( self ):
@@ -307,7 +307,7 @@ class page:
         """This convenience function is only useful for html.
         It adds css stylesheet(s) to the document via the <link> element."""
       
-        if isinstance( filelist, str ):
+        if isinstance( filelist, basestring ):
             self.link( href=filelist, rel='stylesheet', type='text/css', media='all' )
         else:
             for file in filelist:
@@ -319,20 +319,20 @@ class page:
         a dictionary of the form { 'name':'content' }."""
 
         if isinstance( mydict, dict ):
-            for name, content in mydict.items( ):
+            for name, content in mydict.iteritems( ):
                 self.meta( name=name, content=content )
         else:
-            raise TypeError("Metainfo should be called with a dictionary argument of name:content pairs.")
+            raise TypeError, "Metainfo should be called with a dictionary argument of name:content pairs."
 
     def scripts( self, mydict ):
         """Only useful in html, mydict is dictionary of src:type pairs will
         be rendered as <script type='text/type' src=src></script>"""
 
         if isinstance( mydict, dict ):
-            for src, type in mydict.items( ):
+            for src, type in mydict.iteritems( ):
                 self.script( '', src=src, type='text/%s' % type )
         else:
-            raise TypeError("Script should be given a dictionary of src:type pairs.")
+            raise TypeError, "Script should be given a dictionary of src:type pairs."
 
 
 class _oneliner:
@@ -345,7 +345,7 @@ class _oneliner:
     
     def __getattr__( self, attr ):
         if attr.startswith("__") and attr.endswith("__"):
-            raise AttributeError(attr)
+            raise AttributeError, attr
         return element( attr, case=self.case, parent=None )
 
 oneliner = _oneliner( case='lower' )
@@ -359,14 +359,14 @@ def _argsdicts( args, mydict ):
     elif len( args ) == 1:
         args = _totuple( args[0] )
     else:
-        raise Exception("We should have never gotten here.")
+        raise Exception, "We should have never gotten here."
 
-    mykeys = list(mydict.keys( ))
-    myvalues = list(map( _totuple, list(mydict.values( )) ))
+    mykeys = mydict.keys( )
+    myvalues = map( _totuple, mydict.values( ) )
 
-    maxlength = max( list(map( len, [ args ] + myvalues )) )
+    maxlength = max( map( len, [ args ] + myvalues ) )
 
-    for i in range( maxlength ):
+    for i in xrange( maxlength ):
         thisdict = { }
         for key, value in zip( mykeys, myvalues ):
             try:
@@ -383,7 +383,7 @@ def _argsdicts( args, mydict ):
 def _totuple( x ):
     """Utility stuff to convert string, int, float, None or anything to a usable tuple."""
 
-    if isinstance( x, str ):
+    if isinstance( x, basestring ):
         out = x,
     elif isinstance( x, ( int, float ) ):
         out = str( x ),
@@ -397,7 +397,7 @@ def _totuple( x ):
 def escape( text, newline=False ):
     """Escape special html characters."""
 
-    if isinstance( text, str ):
+    if isinstance( text, basestring ):
         if '&' in text:
             text = text.replace( '&', '&amp;' )
         if '>' in text:
@@ -419,7 +419,7 @@ _escape = escape
 def unescape( text ):
     """Inverse of escape."""
     
-    if isinstance( text, str ):
+    if isinstance( text, basestring ):
         if '&amp;' in text:
             text = text.replace( '&amp;', '&' )
         if '&gt;' in text:
@@ -481,4 +481,4 @@ class CustomizationError( MarkupError ):
         self.message = "If you customize the allowed elements, you must define both types 'onetags' and 'twotags'."
 
 if __name__ == '__main__':
-    print(__doc__)
+    print __doc__

tablib/packages/markup3.py

@@ -0,0 +1,484 @@
+# This code is in the public domain, it comes
+# with absolutely no warranty and you can do
+# absolutely whatever you want with it.
+
+__date__ = '17 May 2007'
+__version__ = '1.7'
+__doc__= """
+This is markup.py - a Python module that attempts to
+make it easier to generate HTML/XML from a Python program
+in an intuitive, lightweight, customizable and pythonic way.
+
+The code is in the public domain.
+
+Version: %s as of %s.
+
+Documentation and further info is at http://markup.sourceforge.net/
+
+Please send bug reports, feature requests, enhancement
+ideas or questions to nogradi at gmail dot com.
+
+Installation: drop markup.py somewhere into your Python path.
+""" % ( __version__, __date__ )
+
+import string
+
+class element:
+    """This class handles the addition of a new element."""
+
+    def __init__( self, tag, case='lower', parent=None ):
+        self.parent = parent
+
+        if case == 'lower':
+            self.tag = tag.lower( )
+        else:
+            self.tag = tag.upper( )
+    
+    def __call__( self, *args, **kwargs ):
+        if len( args ) > 1:
+            raise ArgumentError( self.tag )
+
+        # if class_ was defined in parent it should be added to every element
+        if self.parent is not None and self.parent.class_ is not None:
+            if 'class_' not in kwargs:
+                kwargs['class_'] = self.parent.class_
+            
+        if self.parent is None and len( args ) == 1:
+            x = [ self.render( self.tag, False, myarg, mydict ) for myarg, mydict in _argsdicts( args, kwargs ) ]
+            return '\n'.join( x )
+        elif self.parent is None and len( args ) == 0:
+            x = [ self.render( self.tag, True, myarg, mydict ) for myarg, mydict in _argsdicts( args, kwargs ) ]
+            return '\n'.join( x )
+            
+        if self.tag in self.parent.twotags:
+            for myarg, mydict in _argsdicts( args, kwargs ):
+                self.render( self.tag, False, myarg, mydict )
+        elif self.tag in self.parent.onetags:
+            if len( args ) == 0:
+                for myarg, mydict in _argsdicts( args, kwargs ):
+                    self.render( self.tag, True, myarg, mydict )    # here myarg is always None, because len( args ) = 0
+            else:
+                raise ClosingError( self.tag )
+        elif self.parent.mode == 'strict_html' and self.tag in self.parent.deptags:
+            raise DeprecationError( self.tag )
+        else:
+            raise InvalidElementError( self.tag, self.parent.mode )
+    
+    def render( self, tag, single, between, kwargs ):
+        """Append the actual tags to content."""
+
+        out = "<%s" % tag
+        for key, value in kwargs.items( ):
+            if value is not None:               # when value is None that means stuff like <... checked>
+                key = key.strip('_')            # strip this so class_ will mean class, etc.
+                if key == 'http_equiv':         # special cases, maybe change _ to - overall?
+                    key = 'http-equiv'
+                elif key == 'accept_charset':
+                    key = 'accept-charset'
+                out = "%s %s=\"%s\"" % ( out, key, escape( value ) )
+            else:
+                out = "%s %s" % ( out, key )
+        if between is not None:
+            out = "%s>%s</%s>" % ( out, between, tag )
+        else:
+            if single:
+                out = "%s />" % out
+            else:
+                out = "%s>" % out
+        if self.parent is not None:
+            self.parent.content.append( out )
+        else:
+            return out
+    
+    def close( self ):
+        """Append a closing tag unless element has only opening tag."""
+
+        if self.tag in self.parent.twotags:
+            self.parent.content.append( "</%s>" % self.tag )
+        elif self.tag in self.parent.onetags:
+            raise ClosingError( self.tag )
+        elif self.parent.mode == 'strict_html' and self.tag in self.parent.deptags:
+            raise DeprecationError( self.tag )
+
+    def open( self, **kwargs ):
+        """Append an opening tag."""
+
+        if self.tag in self.parent.twotags or self.tag in self.parent.onetags:
+            self.render( self.tag, False, None, kwargs )
+        elif self.mode == 'strict_html' and self.tag in self.parent.deptags:
+            raise DeprecationError( self.tag )
+
+class page:
+    """This is our main class representing a document. Elements are added
+    as attributes of an instance of this class."""
+
+    def __init__( self, mode='strict_html', case='lower', onetags=None, twotags=None, separator='\n', class_=None ):
+        """Stuff that effects the whole document.
+
+        mode -- 'strict_html'   for HTML 4.01 (default)
+                'html'          alias for 'strict_html'
+                'loose_html'    to allow some deprecated elements
+                'xml'           to allow arbitrary elements
+
+        case -- 'lower'         element names will be printed in lower case (default)
+                'upper'         they will be printed in upper case
+
+        onetags --              list or tuple of valid elements with opening tags only
+        twotags --              list or tuple of valid elements with both opening and closing tags
+                                these two keyword arguments may be used to select
+                                the set of valid elements in 'xml' mode
+                                invalid elements will raise appropriate exceptions
+        
+        separator --            string to place between added elements, defaults to newline
+        
+        class_ --               a class that will be added to every element if defined"""
+        
+        valid_onetags = [ "AREA", "BASE", "BR", "COL", "FRAME", "HR", "IMG", "INPUT", "LINK", "META", "PARAM" ]
+        valid_twotags = [ "A", "ABBR", "ACRONYM", "ADDRESS", "B", "BDO", "BIG", "BLOCKQUOTE", "BODY", "BUTTON",
+                "CAPTION", "CITE", "CODE", "COLGROUP", "DD", "DEL", "DFN", "DIV", "DL", "DT", "EM", "FIELDSET",
+                "FORM", "FRAMESET", "H1", "H2", "H3", "H4", "H5", "H6", "HEAD", "HTML", "I", "IFRAME", "INS",
+                "KBD", "LABEL", "LEGEND", "LI", "MAP", "NOFRAMES", "NOSCRIPT", "OBJECT", "OL", "OPTGROUP",
+                "OPTION", "P", "PRE", "Q", "SAMP", "SCRIPT", "SELECT", "SMALL", "SPAN", "STRONG", "STYLE",
+                "SUB", "SUP", "TABLE", "TBODY", "TD", "TEXTAREA", "TFOOT", "TH", "THEAD", "TITLE", "TR",
+                "TT", "UL", "VAR" ]
+        deprecated_onetags = [ "BASEFONT", "ISINDEX" ]
+        deprecated_twotags = [ "APPLET", "CENTER", "DIR", "FONT", "MENU", "S", "STRIKE", "U" ]
+
+        self.header = [ ]
+        self.content = [ ]
+        self.footer = [ ]
+        self.case = case
+        self.separator = separator
+
+        # init( ) sets it to True so we know that </body></html> has to be printed at the end
+        self._full = False
+        self.class_= class_
+
+        if mode == 'strict_html' or mode == 'html':
+            self.onetags = valid_onetags
+            self.onetags += list(map( str.lower, self.onetags ))
+            self.twotags = valid_twotags
+            self.twotags += list(map( str.lower, self.twotags ))
+            self.deptags = deprecated_onetags + deprecated_twotags
+            self.deptags += list(map( str.lower, self.deptags ))
+            self.mode = 'strict_html'
+        elif mode == 'loose_html':
+            self.onetags = valid_onetags + deprecated_onetags 
+            self.onetags += list(map( str.lower, self.onetags ))
+            self.twotags = valid_twotags + deprecated_twotags
+            self.twotags += list(map( str.lower, self.twotags ))
+            self.mode = mode
+        elif mode == 'xml':
+            if onetags and twotags:
+                self.onetags = onetags
+                self.twotags = twotags
+            elif ( onetags and not twotags ) or ( twotags and not onetags ):
+                raise CustomizationError( )
+            else:
+                self.onetags = russell( )
+                self.twotags = russell( )
+            self.mode = mode
+        else:
+            raise ModeError( mode )
+
+    def __getattr__( self, attr ):
+        if attr.startswith("__") and attr.endswith("__"):
+            raise AttributeError(attr)
+        return element( attr, case=self.case, parent=self )
+
+    def __str__( self ):
+        
+        if self._full and ( self.mode == 'strict_html' or self.mode == 'loose_html' ):
+            end = [ '</body>', '</html>' ]
+        else:
+            end = [ ]
+        
+        return self.separator.join( self.header + self.content + self.footer + end )
+
+    def __call__( self, escape=False ):
+        """Return the document as a string.
+
+        escape --   False   print normally
+                    True    replace < and > by &lt; and &gt;
+                            the default escape sequences in most browsers"""
+
+        if escape:
+            return _escape( self.__str__( ) )
+        else:
+            return self.__str__( )
+
+    def add( self, text ):
+        """This is an alias to addcontent."""
+        self.addcontent( text )
+
+    def addfooter( self, text ):
+        """Add some text to the bottom of the document"""
+        self.footer.append( text )
+
+    def addheader( self, text ):
+        """Add some text to the top of the document"""
+        self.header.append( text )
+
+    def addcontent( self, text ):
+        """Add some text to the main part of the document"""
+        self.content.append( text )
+
+
+    def init( self, lang='en', css=None, metainfo=None, title=None, header=None,
+              footer=None, charset=None, encoding=None, doctype=None, bodyattrs=None, script=None ):
+        """This method is used for complete documents with appropriate
+        doctype, encoding, title, etc information. For an HTML/XML snippet
+        omit this method.
+
+        lang --     language, usually a two character string, will appear
+                    as <html lang='en'> in html mode (ignored in xml mode)
+        
+        css --      Cascading Style Sheet filename as a string or a list of
+                    strings for multiple css files (ignored in xml mode)
+
+        metainfo -- a dictionary in the form { 'name':'content' } to be inserted
+                    into meta element(s) as <meta name='name' content='content'>
+                    (ignored in xml mode)
+
+        bodyattrs --a dictionary in the form { 'key':'value', ... } which will be added
+                    as attributes of the <body> element as <body key='value' ... >
+                    (ignored in xml mode)
+
+        script --   dictionary containing src:type pairs, <script type='text/type' src=src></script>
+
+        title --    the title of the document as a string to be inserted into
+                    a title element as <title>my title</title> (ignored in xml mode)
+
+        header --   some text to be inserted right after the <body> element
+                    (ignored in xml mode)
+
+        footer --   some text to be inserted right before the </body> element
+                    (ignored in xml mode)
+
+        charset --  a string defining the character set, will be inserted into a
+                    <meta http-equiv='Content-Type' content='text/html; charset=myset'>
+                    element (ignored in xml mode)
+
+        encoding -- a string defining the encoding, will be put into to first line of
+                    the document as <?xml version='1.0' encoding='myencoding' ?> in
+                    xml mode (ignored in html mode)
+
+        doctype --  the document type string, defaults to
+                    <!DOCTYPE HTML PUBLIC '-//W3C//DTD HTML 4.01 Transitional//EN'>
+                    in html mode (ignored in xml mode)"""
+
+        self._full = True
+
+        if self.mode == 'strict_html' or self.mode == 'loose_html':
+            if doctype is None:
+                doctype = "<!DOCTYPE HTML PUBLIC '-//W3C//DTD HTML 4.01 Transitional//EN'>"
+            self.header.append( doctype )
+            self.html( lang=lang )
+            self.head( )
+            if charset is not None:
+                self.meta( http_equiv='Content-Type', content="text/html; charset=%s" % charset )
+            if metainfo is not None:
+                self.metainfo( metainfo )
+            if css is not None:
+                self.css( css )
+            if title is not None:
+                self.title( title )
+            if script is not None:
+                self.scripts( script )
+            self.head.close()
+            if bodyattrs is not None:
+                self.body( **bodyattrs )
+            else:
+                self.body( )
+            if header is not None:
+                self.content.append( header )
+            if footer is not None:
+                self.footer.append( footer )
+
+        elif self.mode == 'xml':
+            if doctype is None:
+                if encoding is not None:
+                    doctype = "<?xml version='1.0' encoding='%s' ?>" % encoding
+                else:
+                    doctype = "<?xml version='1.0' ?>"
+            self.header.append( doctype )
+
+    def css( self, filelist ):
+        """This convenience function is only useful for html.
+        It adds css stylesheet(s) to the document via the <link> element."""
+      
+        if isinstance( filelist, str ):
+            self.link( href=filelist, rel='stylesheet', type='text/css', media='all' )
+        else:
+            for file in filelist:
+                self.link( href=file, rel='stylesheet', type='text/css', media='all' )
+
+    def metainfo( self, mydict ):
+        """This convenience function is only useful for html.
+        It adds meta information via the <meta> element, the argument is
+        a dictionary of the form { 'name':'content' }."""
+
+        if isinstance( mydict, dict ):
+            for name, content in mydict.items( ):
+                self.meta( name=name, content=content )
+        else:
+            raise TypeError("Metainfo should be called with a dictionary argument of name:content pairs.")
+
+    def scripts( self, mydict ):
+        """Only useful in html, mydict is dictionary of src:type pairs will
+        be rendered as <script type='text/type' src=src></script>"""
+
+        if isinstance( mydict, dict ):
+            for src, type in mydict.items( ):
+                self.script( '', src=src, type='text/%s' % type )
+        else:
+            raise TypeError("Script should be given a dictionary of src:type pairs.")
+
+
+class _oneliner:
+    """An instance of oneliner returns a string corresponding to one element.
+    This class can be used to write 'oneliners' that return a string
+    immediately so there is no need to instantiate the page class."""
+    
+    def __init__( self, case='lower' ):
+        self.case = case
+    
+    def __getattr__( self, attr ):
+        if attr.startswith("__") and attr.endswith("__"):
+            raise AttributeError(attr)
+        return element( attr, case=self.case, parent=None )
+
+oneliner = _oneliner( case='lower' )
+upper_oneliner = _oneliner( case='upper' )
+
+def _argsdicts( args, mydict ):
+    """A utility generator that pads argument list and dictionary values, will only be called with len( args ) = 0, 1."""
+    
+    if len( args ) == 0:
+        args = None, 
+    elif len( args ) == 1:
+        args = _totuple( args[0] )
+    else:
+        raise Exception("We should have never gotten here.")
+
+    mykeys = list(mydict.keys( ))
+    myvalues = list(map( _totuple, list(mydict.values( )) ))
+
+    maxlength = max( list(map( len, [ args ] + myvalues )) )
+
+    for i in range( maxlength ):
+        thisdict = { }
+        for key, value in zip( mykeys, myvalues ):
+            try:
+                thisdict[ key ] = value[i]
+            except IndexError:
+                thisdict[ key ] = value[-1]
+        try:
+            thisarg = args[i]
+        except IndexError:
+            thisarg = args[-1]
+
+        yield thisarg, thisdict
+
+def _totuple( x ):
+    """Utility stuff to convert string, int, float, None or anything to a usable tuple."""
+
+    if isinstance( x, str ):
+        out = x,
+    elif isinstance( x, ( int, float ) ):
+        out = str( x ),
+    elif x is None:
+        out = None,
+    else:
+        out = tuple( x )
+
+    return out
+
+def escape( text, newline=False ):
+    """Escape special html characters."""
+
+    if isinstance( text, str ):
+        if '&' in text:
+            text = text.replace( '&', '&amp;' )
+        if '>' in text:
+            text = text.replace( '>', '&gt;' )
+        if '<' in text:
+            text = text.replace( '<', '&lt;' )
+        if '\"' in text:
+            text = text.replace( '\"', '&quot;' )
+        if '\'' in text:
+            text = text.replace( '\'', '&quot;' )
+        if newline:
+            if '\n' in text:
+                text = text.replace( '\n', '<br>' )
+
+    return text
+
+_escape = escape
+
+def unescape( text ):
+    """Inverse of escape."""
+    
+    if isinstance( text, str ):
+        if '&amp;' in text:
+            text = text.replace( '&amp;', '&' )
+        if '&gt;' in text:
+            text = text.replace( '&gt;', '>' )
+        if '&lt;' in text:
+            text = text.replace( '&lt;', '<' )
+        if '&quot;' in text:
+            text = text.replace( '&quot;', '\"' )
+
+    return text
+
+class dummy:
+    """A dummy class for attaching attributes."""
+    pass
+
+doctype = dummy( )
+doctype.frameset = "<!DOCTYPE HTML PUBLIC '-//W3C//DTD HTML 4.01 Frameset//EN' 'http://www.w3.org/TR/html4/frameset.dtd'>"
+doctype.strict = "<!DOCTYPE HTML PUBLIC '-//W3C//DTD HTML 4.01//EN' 'http://www.w3.org/TR/html4/strict.dtd'>"
+doctype.loose = "<!DOCTYPE HTML PUBLIC '-//W3C//DTD HTML 4.01 Transitional//EN' 'http://www.w3.org/TR/html4/loose.dtd'>"
+
+class russell:
+    """A dummy class that contains anything."""
+
+    def __contains__( self, item ):
+        return True
+
+
+class MarkupError( Exception ):
+    """All our exceptions subclass this."""
+    def __str__( self ):
+        return self.message
+
+class ClosingError( MarkupError ):
+    def __init__( self, tag ):
+        self.message = "The element '%s' does not accept non-keyword arguments (has no closing tag)." % tag
+
+class OpeningError( MarkupError ):
+    def __init__( self, tag ):
+        self.message = "The element '%s' can not be opened." % tag
+
+class ArgumentError( MarkupError ):
+    def __init__( self, tag ):
+        self.message = "The element '%s' was called with more than one non-keyword argument." % tag
+
+class InvalidElementError( MarkupError ):
+    def __init__( self, tag, mode ):
+        self.message = "The element '%s' is not valid for your mode '%s'." % ( tag, mode )
+
+class DeprecationError( MarkupError ):
+    def __init__( self, tag ):
+        self.message = "The element '%s' is deprecated, instantiate markup.page with mode='loose_html' to allow it." % tag
+
+class ModeError( MarkupError ):
+    def __init__( self, mode ):
+        self.message = "Mode '%s' is invalid, possible values: strict_html, loose_html, xml." % mode
+
+class CustomizationError( MarkupError ):
+    def __init__( self ):
+        self.message = "If you customize the allowed elements, you must define both types 'onetags' and 'twotags'."
+
+if __name__ == '__main__':
+    print(__doc__)

tablib/packages/xlwt/Bitmap.py

@@ -1,258 +1,262 @@
-#  Portions are Copyright (C) 2005 Roman V. Kiseliov
-#  Portions are Copyright (c) 2004 Evgeny Filatov <fufff@users.sourceforge.net>
-#  Portions are Copyright (c) 2002-2004 John McNamara (Perl Spreadsheet::WriteExcel)
-
-from .BIFFRecords import BiffRecord
-from struct import *
-
-
-def _size_col(sheet, col):
-    return sheet.col_width(col)
-
-
-def _size_row(sheet, row):
-    return sheet.row_height(row)
-
-
-def _position_image(sheet, row_start, col_start, x1, y1, width, height):
-    """Calculate the vertices that define the position of the image as required by
-    the OBJ record.
-
-             +------------+------------+
-             |     A      |      B     |
-       +-----+------------+------------+
-       |     |(x1,y1)     |            |
-       |  1  |(A1)._______|______      |
-       |     |    |              |     |
-       |     |    |              |     |
-       +-----+----|    BITMAP    |-----+
-       |     |    |              |     |
-       |  2  |    |______________.     |
-       |     |            |        (B2)|
-       |     |            |     (x2,y2)|
-       +---- +------------+------------+
-
-    Example of a bitmap that covers some of the area from cell A1 to cell B2.
-
-    Based on the width and height of the bitmap we need to calculate 8 vars:
-        col_start, row_start, col_end, row_end, x1, y1, x2, y2.
-    The width and height of the cells are also variable and have to be taken into
-    account.
-    The values of col_start and row_start are passed in from the calling
-    function. The values of col_end and row_end are calculated by subtracting
-    the width and height of the bitmap from the width and height of the
-    underlying cells.
-    The vertices are expressed as a percentage of the underlying cell width as
-    follows (rhs values are in pixels):
-
-           x1 = X / W *1024
-           y1 = Y / H *256
-           x2 = (X-1) / W *1024
-           y2 = (Y-1) / H *256
-
-           Where:  X is distance from the left side of the underlying cell
-                   Y is distance from the top of the underlying cell
-                   W is the width of the cell
-                   H is the height of the cell
-
-    Note: the SDK incorrectly states that the height should be expressed as a
-    percentage of 1024.
-
-    col_start  - Col containing upper left corner of object
-    row_start  - Row containing top left corner of object
-    x1  - Distance to left side of object
-    y1  - Distance to top of object
-    width  - Width of image frame
-    height  - Height of image frame
-
-    """
-    # Adjust start column for offsets that are greater than the col width
-    while x1 >= _size_col(sheet, col_start):
-        x1 -= _size_col(sheet, col_start)
-        col_start += 1
-    # Adjust start row for offsets that are greater than the row height
-    while y1 >= _size_row(sheet, row_start):
-        y1 -= _size_row(sheet, row_start)
-        row_start += 1
-    # Initialise end cell to the same as the start cell
-    row_end = row_start   # Row containing bottom right corner of object
-    col_end = col_start   # Col containing lower right corner of object
-    width = width + x1 - 1
-    height = height + y1 - 1
-    # Subtract the underlying cell widths to find the end cell of the image
-    while (width >= _size_col(sheet, col_end)):
-        width -= _size_col(sheet, col_end)
-        col_end += 1
-    # Subtract the underlying cell heights to find the end cell of the image
-    while (height >= _size_row(sheet, row_end)):
-        height -= _size_row(sheet, row_end)
-        row_end += 1
-    # Bitmap isn't allowed to start or finish in a hidden cell, i.e. a cell
-    # with zero height or width.
-    if ((_size_col(sheet, col_start) == 0) or (_size_col(sheet, col_end) == 0)
-            or (_size_row(sheet, row_start) == 0) or (_size_row(sheet, row_end) == 0)):
-        return
-    # Convert the pixel values to the percentage value expected by Excel
-    x1 = int(float(x1) / _size_col(sheet, col_start) * 1024)
-    y1 = int(float(y1) / _size_row(sheet, row_start) * 256)
-    # Distance to right side of object
-    x2 = int(float(width) / _size_col(sheet, col_end) * 1024)
-    # Distance to bottom of object
-    y2 = int(float(height) / _size_row(sheet, row_end) * 256)
-    return (col_start, x1, row_start, y1, col_end, x2, row_end, y2)
-
-
-class ObjBmpRecord(BiffRecord):
-    _REC_ID = 0x005D    # Record identifier
-
-    def __init__(self, row, col, sheet, im_data_bmp, x, y, scale_x, scale_y):
-        # Scale the frame of the image.
-        width = im_data_bmp.width * scale_x
-        height = im_data_bmp.height * scale_y
-
-        # Calculate the vertices of the image and write the OBJ record
-        coordinates = _position_image(sheet, row, col, x, y, width, height)
-        # print coordinates
-        col_start, x1, row_start, y1, col_end, x2, row_end, y2 = coordinates
-
-        """Store the OBJ record that precedes an IMDATA record. This could be generalise
-        to support other Excel objects.
-
-        """
-        cObj = 0x0001      # Count of objects in file (set to 1)
-        OT = 0x0008        # Object type. 8 = Picture
-        id = 0x0001        # Object ID
-        grbit = 0x0614     # Option flags
-        colL = col_start    # Col containing upper left corner of object
-        dxL = x1            # Distance from left side of cell
-        rwT = row_start     # Row containing top left corner of object
-        dyT = y1            # Distance from top of cell
-        colR = col_end      # Col containing lower right corner of object
-        dxR = x2            # Distance from right of cell
-        rwB = row_end       # Row containing bottom right corner of object
-        dyB = y2            # Distance from bottom of cell
-        cbMacro = 0x0000    # Length of FMLA structure
-        Reserved1 = 0x0000  # Reserved
-        Reserved2 = 0x0000  # Reserved
-        icvBack = 0x09      # Background colour
-        icvFore = 0x09      # Foreground colour
-        fls = 0x00          # Fill pattern
-        fAuto = 0x00        # Automatic fill
-        icv = 0x08          # Line colour
-        lns = 0xff          # Line style
-        lnw = 0x01          # Line weight
-        fAutoB = 0x00       # Automatic border
-        frs = 0x0000        # Frame style
-        cf = 0x0009         # Image format, 9 = bitmap
-        Reserved3 = 0x0000  # Reserved
-        cbPictFmla = 0x0000 # Length of FMLA structure
-        Reserved4 = 0x0000  # Reserved
-        grbit2 = 0x0001     # Option flags
-        Reserved5 = 0x0000  # Reserved
-
-        data = pack("<L", cObj)
-        data += pack("<H", OT)
-        data += pack("<H", id)
-        data += pack("<H", grbit)
-        data += pack("<H", colL)
-        data += pack("<H", dxL)
-        data += pack("<H", rwT)
-        data += pack("<H", dyT)
-        data += pack("<H", colR)
-        data += pack("<H", dxR)
-        data += pack("<H", rwB)
-        data += pack("<H", dyB)
-        data += pack("<H", cbMacro)
-        data += pack("<L", Reserved1)
-        data += pack("<H", Reserved2)
-        data += pack("<B", icvBack)
-        data += pack("<B", icvFore)
-        data += pack("<B", fls)
-        data += pack("<B", fAuto)
-        data += pack("<B", icv)
-        data += pack("<B", lns)
-        data += pack("<B", lnw)
-        data += pack("<B", fAutoB)
-        data += pack("<H", frs)
-        data += pack("<L", cf)
-        data += pack("<H", Reserved3)
-        data += pack("<H", cbPictFmla)
-        data += pack("<H", Reserved4)
-        data += pack("<H", grbit2)
-        data += pack("<L", Reserved5)
-
-        self._rec_data = data
-
-def _process_bitmap(bitmap):
-    """Convert a 24 bit bitmap into the modified internal format used by Windows.
-    This is described in BITMAPCOREHEADER and BITMAPCOREINFO structures in the
-    MSDN library.
-
-    """
-    # Open file and binmode the data in case the platform needs it.
-    fh = open(bitmap, 'rb')
-    try:
-        # Slurp the file into a string.
-        data = fh.read()
-    finally:
-        fh.close()
-    # Check that the file is big enough to be a bitmap.
-    if len(data) <= 0x36:
-        raise Exception("bitmap doesn't contain enough data.")
-    # The first 2 bytes are used to identify the bitmap.
-    if (data[:2] != b"BM"):
-        raise Exception("bitmap doesn't appear to to be a valid bitmap image.")
-    # Remove bitmap data: ID.
-    data = data[2:]
-    # Read and remove the bitmap size. This is more reliable than reading
-    # the data size at offset 0x22.
-    #
-    size = unpack("<L", data[:4])[0]
-    size -=  0x36   # Subtract size of bitmap header.
-    size +=  0x0C   # Add size of BIFF header.
-    data = data[4:]
-    # Remove bitmap data: reserved, offset, header length.
-    data = data[12:]
-    # Read and remove the bitmap width and height. Verify the sizes.
-    width, height = unpack("<LL", data[:8])
-    data = data[8:]
-    if (width > 0xFFFF):
-        raise Exception("bitmap: largest image width supported is 65k.")
-    if (height > 0xFFFF):
-        raise Exception("bitmap: largest image height supported is 65k.")
-    # Read and remove the bitmap planes and bpp data. Verify them.
-    planes, bitcount = unpack("<HH", data[:4])
-    data = data[4:]
-    if (bitcount != 24):
-        raise Exception("bitmap isn't a 24bit true color bitmap.")
-    if (planes != 1):
-        raise Exception("bitmap: only 1 plane supported in bitmap image.")
-    # Read and remove the bitmap compression. Verify compression.
-    compression = unpack("<L", data[:4])[0]
-    data = data[4:]
-    if (compression != 0):
-        raise Exception("bitmap: compression not supported in bitmap image.")
-    # Remove bitmap data: data size, hres, vres, colours, imp. colours.
-    data = data[20:]
-    # Add the BITMAPCOREHEADER data
-    header = pack("<LHHHH", 0x000c, width, height, 0x01, 0x18)
-    data = header + data
-    return (width, height, size, data)
-
-
-class ImDataBmpRecord(BiffRecord):
-    _REC_ID = 0x007F
-
-    def __init__(self, filename):
-        """Insert a 24bit bitmap image in a worksheet. The main record required is
-        IMDATA but it must be proceeded by a OBJ record to define its position.
-
-        """
-        BiffRecord.__init__(self)
-
-        self.width, self.height, self.size, data = _process_bitmap(filename)
-        # Write the IMDATA record to store the bitmap data
-        cf = 0x09
-        env = 0x01
-        lcb = self.size
-        self._rec_data = pack("<HHL", cf, env, lcb) + data
+# -*- coding: windows-1251 -*-
+
+#  Portions are Copyright (C) 2005 Roman V. Kiseliov
+#  Portions are Copyright (c) 2004 Evgeny Filatov <fufff@users.sourceforge.net>
+#  Portions are Copyright (c) 2002-2004 John McNamara (Perl Spreadsheet::WriteExcel)
+
+from BIFFRecords import BiffRecord
+from struct import *
+
+
+def _size_col(sheet, col):
+    return sheet.col_width(col)
+
+
+def _size_row(sheet, row):
+    return sheet.row_height(row)
+
+
+def _position_image(sheet, row_start, col_start, x1, y1, width, height):
+    """Calculate the vertices that define the position of the image as required by
+    the OBJ record.
+
+             +------------+------------+
+             |     A      |      B     |
+       +-----+------------+------------+
+       |     |(x1,y1)     |            |
+       |  1  |(A1)._______|______      |
+       |     |    |              |     |
+       |     |    |              |     |
+       +-----+----|    BITMAP    |-----+
+       |     |    |              |     |
+       |  2  |    |______________.     |
+       |     |            |        (B2)|
+       |     |            |     (x2,y2)|
+       +---- +------------+------------+
+
+    Example of a bitmap that covers some of the area from cell A1 to cell B2.
+
+    Based on the width and height of the bitmap we need to calculate 8 vars:
+        col_start, row_start, col_end, row_end, x1, y1, x2, y2.
+    The width and height of the cells are also variable and have to be taken into
+    account.
+    The values of col_start and row_start are passed in from the calling
+    function. The values of col_end and row_end are calculated by subtracting
+    the width and height of the bitmap from the width and height of the
+    underlying cells.
+    The vertices are expressed as a percentage of the underlying cell width as
+    follows (rhs values are in pixels):
+
+           x1 = X / W *1024
+           y1 = Y / H *256
+           x2 = (X-1) / W *1024
+           y2 = (Y-1) / H *256
+
+           Where:  X is distance from the left side of the underlying cell
+                   Y is distance from the top of the underlying cell
+                   W is the width of the cell
+                   H is the height of the cell
+
+    Note: the SDK incorrectly states that the height should be expressed as a
+    percentage of 1024.
+
+    col_start  - Col containing upper left corner of object
+    row_start  - Row containing top left corner of object
+    x1  - Distance to left side of object
+    y1  - Distance to top of object
+    width  - Width of image frame
+    height  - Height of image frame
+
+    """
+    # Adjust start column for offsets that are greater than the col width
+    while x1 >= _size_col(sheet, col_start):
+        x1 -= _size_col(sheet, col_start)
+        col_start += 1
+    # Adjust start row for offsets that are greater than the row height
+    while y1 >= _size_row(sheet, row_start):
+        y1 -= _size_row(sheet, row_start)
+        row_start += 1
+    # Initialise end cell to the same as the start cell
+    row_end = row_start   # Row containing bottom right corner of object
+    col_end = col_start   # Col containing lower right corner of object
+    width = width + x1 - 1
+    height = height + y1 - 1
+    # Subtract the underlying cell widths to find the end cell of the image
+    while (width >= _size_col(sheet, col_end)):
+        width -= _size_col(sheet, col_end)
+        col_end += 1
+    # Subtract the underlying cell heights to find the end cell of the image
+    while (height >= _size_row(sheet, row_end)):
+        height -= _size_row(sheet, row_end)
+        row_end += 1
+    # Bitmap isn't allowed to start or finish in a hidden cell, i.e. a cell
+    # with zero height or width.
+    if ((_size_col(sheet, col_start) == 0) or (_size_col(sheet, col_end) == 0)
+            or (_size_row(sheet, row_start) == 0) or (_size_row(sheet, row_end) == 0)):
+        return
+    # Convert the pixel values to the percentage value expected by Excel
+    x1 = int(float(x1) / _size_col(sheet, col_start) * 1024)
+    y1 = int(float(y1) / _size_row(sheet, row_start) * 256)
+    # Distance to right side of object
+    x2 = int(float(width) / _size_col(sheet, col_end) * 1024)
+    # Distance to bottom of object
+    y2 = int(float(height) / _size_row(sheet, row_end) * 256)
+    return (col_start, x1, row_start, y1, col_end, x2, row_end, y2)
+
+
+class ObjBmpRecord(BiffRecord):
+    _REC_ID = 0x005D    # Record identifier
+
+    def __init__(self, row, col, sheet, im_data_bmp, x, y, scale_x, scale_y):
+        # Scale the frame of the image.
+        width = im_data_bmp.width * scale_x
+        height = im_data_bmp.height * scale_y
+
+        # Calculate the vertices of the image and write the OBJ record
+        coordinates = _position_image(sheet, row, col, x, y, width, height)
+        # print coordinates
+        col_start, x1, row_start, y1, col_end, x2, row_end, y2 = coordinates
+
+        """Store the OBJ record that precedes an IMDATA record. This could be generalise
+        to support other Excel objects.
+
+        """
+        cObj = 0x0001      # Count of objects in file (set to 1)
+        OT = 0x0008        # Object type. 8 = Picture
+        id = 0x0001        # Object ID
+        grbit = 0x0614     # Option flags
+        colL = col_start    # Col containing upper left corner of object
+        dxL = x1            # Distance from left side of cell
+        rwT = row_start     # Row containing top left corner of object
+        dyT = y1            # Distance from top of cell
+        colR = col_end      # Col containing lower right corner of object
+        dxR = x2            # Distance from right of cell
+        rwB = row_end       # Row containing bottom right corner of object
+        dyB = y2            # Distance from bottom of cell
+        cbMacro = 0x0000    # Length of FMLA structure
+        Reserved1 = 0x0000  # Reserved
+        Reserved2 = 0x0000  # Reserved
+        icvBack = 0x09      # Background colour
+        icvFore = 0x09      # Foreground colour
+        fls = 0x00          # Fill pattern
+        fAuto = 0x00        # Automatic fill
+        icv = 0x08          # Line colour
+        lns = 0xff          # Line style
+        lnw = 0x01          # Line weight
+        fAutoB = 0x00       # Automatic border
+        frs = 0x0000        # Frame style
+        cf = 0x0009         # Image format, 9 = bitmap
+        Reserved3 = 0x0000  # Reserved
+        cbPictFmla = 0x0000 # Length of FMLA structure
+        Reserved4 = 0x0000  # Reserved
+        grbit2 = 0x0001     # Option flags
+        Reserved5 = 0x0000  # Reserved
+
+        data = pack("<L", cObj)
+        data += pack("<H", OT)
+        data += pack("<H", id)
+        data += pack("<H", grbit)
+        data += pack("<H", colL)
+        data += pack("<H", dxL)
+        data += pack("<H", rwT)
+        data += pack("<H", dyT)
+        data += pack("<H", colR)
+        data += pack("<H", dxR)
+        data += pack("<H", rwB)
+        data += pack("<H", dyB)
+        data += pack("<H", cbMacro)
+        data += pack("<L", Reserved1)
+        data += pack("<H", Reserved2)
+        data += pack("<B", icvBack)
+        data += pack("<B", icvFore)
+        data += pack("<B", fls)
+        data += pack("<B", fAuto)
+        data += pack("<B", icv)
+        data += pack("<B", lns)
+        data += pack("<B", lnw)
+        data += pack("<B", fAutoB)
+        data += pack("<H", frs)
+        data += pack("<L", cf)
+        data += pack("<H", Reserved3)
+        data += pack("<H", cbPictFmla)
+        data += pack("<H", Reserved4)
+        data += pack("<H", grbit2)
+        data += pack("<L", Reserved5)
+
+        self._rec_data = data
+
+def _process_bitmap(bitmap):
+    """Convert a 24 bit bitmap into the modified internal format used by Windows.
+    This is described in BITMAPCOREHEADER and BITMAPCOREINFO structures in the
+    MSDN library.
+
+    """
+    # Open file and binmode the data in case the platform needs it.
+    fh = file(bitmap, "rb")
+    try:
+        # Slurp the file into a string.
+        data = fh.read()
+    finally:
+        fh.close()
+    # Check that the file is big enough to be a bitmap.
+    if len(data) <= 0x36:
+        raise Exception("bitmap doesn't contain enough data.")
+    # The first 2 bytes are used to identify the bitmap.
+    if (data[:2] != "BM"):
+        raise Exception("bitmap doesn't appear to to be a valid bitmap image.")
+    # Remove bitmap data: ID.
+    data = data[2:]
+    # Read and remove the bitmap size. This is more reliable than reading
+    # the data size at offset 0x22.
+    #
+    size = unpack("<L", data[:4])[0]
+    size -=  0x36   # Subtract size of bitmap header.
+    size +=  0x0C   # Add size of BIFF header.
+    data = data[4:]
+    # Remove bitmap data: reserved, offset, header length.
+    data = data[12:]
+    # Read and remove the bitmap width and height. Verify the sizes.
+    width, height = unpack("<LL", data[:8])
+    data = data[8:]
+    if (width > 0xFFFF):
+        raise Exception("bitmap: largest image width supported is 65k.")
+    if (height > 0xFFFF):
+        raise Exception("bitmap: largest image height supported is 65k.")
+    # Read and remove the bitmap planes and bpp data. Verify them.
+    planes, bitcount = unpack("<HH", data[:4])
+    data = data[4:]
+    if (bitcount != 24):
+        raise Exception("bitmap isn't a 24bit true color bitmap.")
+    if (planes != 1):
+        raise Exception("bitmap: only 1 plane supported in bitmap image.")
+    # Read and remove the bitmap compression. Verify compression.
+    compression = unpack("<L", data[:4])[0]
+    data = data[4:]
+    if (compression != 0):
+        raise Exception("bitmap: compression not supported in bitmap image.")
+    # Remove bitmap data: data size, hres, vres, colours, imp. colours.
+    data = data[20:]
+    # Add the BITMAPCOREHEADER data
+    header = pack("<LHHHH", 0x000c, width, height, 0x01, 0x18)
+    data = header + data
+    return (width, height, size, data)
+
+
+class ImDataBmpRecord(BiffRecord):
+    _REC_ID = 0x007F
+
+    def __init__(self, filename):
+        """Insert a 24bit bitmap image in a worksheet. The main record required is
+        IMDATA but it must be proceeded by a OBJ record to define its position.
+
+        """
+        BiffRecord.__init__(self)
+
+        self.width, self.height, self.size, data = _process_bitmap(filename)
+        # Write the IMDATA record to store the bitmap data
+        cf = 0x09
+        env = 0x01
+        lcb = self.size
+        self._rec_data = pack("<HHL", cf, env, lcb) + data
+
+

tablib/packages/xlwt/Cell.py

@@ -1,233 +1,243 @@
-from struct import unpack, pack
-from . import BIFFRecords
-
-class StrCell(object):
-    __slots__ = ["rowx", "colx", "xf_idx", "sst_idx"]
-
-    def __init__(self, rowx, colx, xf_idx, sst_idx):
-        self.rowx = rowx
-        self.colx = colx
-        self.xf_idx = xf_idx
-        self.sst_idx = sst_idx
-
-    def get_biff_data(self):
-        return pack('<5HL', 0x00FD, 10, self.rowx, self.colx, self.xf_idx, self.sst_idx)
-
-class BlankCell(object):
-    __slots__ = ["rowx", "colx", "xf_idx"]
-
-    def __init__(self, rowx, colx, xf_idx):
-        self.rowx = rowx
-        self.colx = colx
-        self.xf_idx = xf_idx
-
-    def get_biff_data(self):
-        return pack('<5H', 0x0201, 6, self.rowx, self.colx, self.xf_idx)
-
-class MulBlankCell(object):
-    __slots__ = ["rowx", "colx1", "colx2", "xf_idx"]
-
-    def __init__(self, rowx, colx1, colx2, xf_idx):
-        self.rowx = rowx
-        self.colx1 = colx1
-        self.colx2 = colx2
-        self.xf_idx = xf_idx
-
-    def get_biff_data(self):
-        return BIFFRecords.MulBlankRecord(self.rowx,
-            self.colx1, self.colx2, self.xf_idx).get()
-
-class NumberCell(object):
-    __slots__ = ["rowx", "colx", "xf_idx", "number"]
-
-    def __init__(self, rowx, colx, xf_idx, number):
-        self.rowx = rowx
-        self.colx = colx
-        self.xf_idx = xf_idx
-        self.number = float(number)
-
-    def get_encoded_data(self):
-        rk_encoded = 0
-        num = self.number
-
-        # The four possible kinds of RK encoding are *not* mutually exclusive.
-        # The 30-bit integer variety picks up the most.
-        # In the code below, the four varieties are checked in descending order
-        # of bangs per buck, or not at all.
-        # SJM 2007-10-01
-
-        if -0x20000000 <= num < 0x20000000: # fits in 30-bit *signed* int
-            inum = int(num)
-            if inum == num: # survives round-trip
-                rk_encoded = 2 | (inum << 2)
-                return 1, rk_encoded
-
-        temp = num * 100
-
-        if -0x20000000 <= temp < 0x20000000:
-            # That was step 1: the coded value will fit in
-            # a 30-bit signed integer.
-            itemp = int(round(temp, 0))
-            # That was step 2: "itemp" is the best candidate coded value.
-            # Now for step 3: simulate the decoding,
-            # to check for round-trip correctness.
-            if itemp / 100.0 == num:
-                rk_encoded = 3 | (itemp << 2)
-                return 1, rk_encoded
-
-        if 0: # Cost of extra pack+unpack not justified by tiny yield.
-            packed = pack('<d', num)
-            w01, w23 = unpack('<2i', packed)
-            if not w01 and not(w23 & 3):
-                return 1, w23
-
-            packed100 = pack('<d', temp)
-            w01, w23 = unpack('<2i', packed100)
-            if not w01 and not(w23 & 3):
-                return 1, w23 | 1
-
-        return 0, pack('<5Hd', 0x0203, 14, self.rowx, self.colx, self.xf_idx, num)
-
-    def get_biff_data(self):
-        isRK, value = self.get_encoded_data()
-        if isRK:
-            return pack('<5Hi', 0x27E, 10, self.rowx, self.colx, self.xf_idx, value)
-        return value # NUMBER record already packed
-
-class BooleanCell(object):
-    __slots__ = ["rowx", "colx", "xf_idx", "number"]
-
-    def __init__(self, rowx, colx, xf_idx, number):
-        self.rowx = rowx
-        self.colx = colx
-        self.xf_idx = xf_idx
-        self.number = number
-
-    def get_biff_data(self):
-        return BIFFRecords.BoolErrRecord(self.rowx,
-            self.colx, self.xf_idx, self.number, 0).get()
-
-error_code_map = {
-    0x00:  0, # Intersection of two cell ranges is empty
-    0x07:  7, # Division by zero
-    0x0F: 15, # Wrong type of operand
-    0x17: 23, # Illegal or deleted cell reference
-    0x1D: 29, # Wrong function or range name
-    0x24: 36, # Value range overflow
-    0x2A: 42, # Argument or function not available
-    '#NULL!' :  0, # Intersection of two cell ranges is empty
-    '#DIV/0!':  7, # Division by zero
-    '#VALUE!': 36, # Wrong type of operand
-    '#REF!'  : 23, # Illegal or deleted cell reference
-    '#NAME?' : 29, # Wrong function or range name
-    '#NUM!'  : 36, # Value range overflow
-    '#N/A!'  : 42, # Argument or function not available
-}
-
-class ErrorCell(object):
-    __slots__ = ["rowx", "colx", "xf_idx", "number"]
-
-    def __init__(self, rowx, colx, xf_idx, error_string_or_code):
-        self.rowx = rowx
-        self.colx = colx
-        self.xf_idx = xf_idx
-        try:
-            self.number = error_code_map[error_string_or_code]
-        except KeyError:
-            raise Exception('Illegal error value (%r)' % error_string_or_code)
-
-    def get_biff_data(self):
-        return BIFFRecords.BoolErrRecord(self.rowx,
-            self.colx, self.xf_idx, self.number, 1).get()
-
-class FormulaCell(object):
-    __slots__ = ["rowx", "colx", "xf_idx", "frmla", "calc_flags"]
-
-    def __init__(self, rowx, colx, xf_idx, frmla, calc_flags=0):
-        self.rowx = rowx
-        self.colx = colx
-        self.xf_idx = xf_idx
-        self.frmla = frmla
-        self.calc_flags = calc_flags
-
-    def get_biff_data(self):
-        return BIFFRecords.FormulaRecord(self.rowx,
-            self.colx, self.xf_idx, self.frmla.rpn(), self.calc_flags).get()
-
-# module-level function for *internal* use by the Row module
-
-def _get_cells_biff_data_mul(rowx, cell_items):
-    # Return the BIFF data for all cell records in the row.
-    # Adjacent BLANK|RK records are combined into MUL(BLANK|RK) records.
-    pieces = []
-    nitems = len(cell_items)
-    i = 0
-    while i < nitems:
-        icolx, icell = cell_items[i]
-        if isinstance(icell, NumberCell):
-            isRK, value = icell.get_encoded_data()
-            if not isRK:
-                pieces.append(value) # pre-packed NUMBER record
-                i += 1
-                continue
-            muldata = [(value, icell.xf_idx)]
-            target = NumberCell
-        elif isinstance(icell, BlankCell):
-            muldata = [icell.xf_idx]
-            target = BlankCell
-        else:
-            pieces.append(icell.get_biff_data())
-            i += 1
-            continue
-        lastcolx = icolx
-        j = i
-
-        packed_record = b'' # (to_py3): 'b' binary data
-        for j in range(i+1, nitems):
-            jcolx, jcell = cell_items[j]
-            if jcolx != lastcolx + 1:
-                nexti = j
-                break
-            if not isinstance(jcell, target):
-                nexti = j
-                break
-            if target == NumberCell:
-                isRK, value = jcell.get_encoded_data()
-                if not isRK:
-                    packed_record = value
-                    nexti = j + 1
-                    break
-                muldata.append((value, jcell.xf_idx))
-            else:
-                muldata.append(jcell.xf_idx)
-            lastcolx = jcolx
-        else:
-            nexti = j + 1
-        if target == NumberCell:
-            if lastcolx == icolx:
-                # RK record
-                value, xf_idx = muldata[0]
-                pieces.append(pack('<5Hi', 0x027E, 10, rowx, icolx, xf_idx, value))
-            else:
-                # MULRK record
-                nc = lastcolx - icolx + 1
-                pieces.append(pack('<4H', 0x00BD, 6 * nc + 6, rowx, icolx))
-                # (to_py3): 'b' binary data
-                pieces.append(b''.join([pack('<Hi', xf_idx, value) for value, xf_idx in muldata]))
-                pieces.append(pack('<H', lastcolx))
-        else:
-            if lastcolx == icolx:
-                # BLANK record
-                xf_idx = muldata[0]
-                pieces.append(pack('<5H', 0x0201, 6, rowx, icolx, xf_idx))
-            else:
-                # MULBLANK record
-                nc = lastcolx - icolx + 1
-                pieces.append(pack('<4H', 0x00BE, 2 * nc + 6, rowx, icolx))
-                # (to_py3): 'b' binary data
-                pieces.append(b''.join([pack('<H', xf_idx) for xf_idx in muldata]))
-                pieces.append(pack('<H', lastcolx))
-        if packed_record:
-            pieces.append(packed_record)
-        i = nexti
-    return b''.join(pieces) # (to_py3): 'b' binary data
+# -*- coding: windows-1252 -*-
+
+from struct import unpack, pack
+import BIFFRecords
+
+class StrCell(object):
+    __slots__ = ["rowx", "colx", "xf_idx", "sst_idx"]
+
+    def __init__(self, rowx, colx, xf_idx, sst_idx):
+        self.rowx = rowx
+        self.colx = colx
+        self.xf_idx = xf_idx
+        self.sst_idx = sst_idx
+
+    def get_biff_data(self):
+        # return BIFFRecords.LabelSSTRecord(self.rowx, self.colx, self.xf_idx, self.sst_idx).get()
+        return pack('<5HL', 0x00FD, 10, self.rowx, self.colx, self.xf_idx, self.sst_idx)
+
+class BlankCell(object):
+    __slots__ = ["rowx", "colx", "xf_idx"]
+
+    def __init__(self, rowx, colx, xf_idx):
+        self.rowx = rowx
+        self.colx = colx
+        self.xf_idx = xf_idx
+
+    def get_biff_data(self):
+        # return BIFFRecords.BlankRecord(self.rowx, self.colx, self.xf_idx).get()
+        return pack('<5H', 0x0201, 6, self.rowx, self.colx, self.xf_idx)
+
+class MulBlankCell(object):
+    __slots__ = ["rowx", "colx1", "colx2", "xf_idx"]
+
+    def __init__(self, rowx, colx1, colx2, xf_idx):
+        self.rowx = rowx
+        self.colx1 = colx1
+        self.colx2 = colx2
+        self.xf_idx = xf_idx
+
+    def get_biff_data(self):
+        return BIFFRecords.MulBlankRecord(self.rowx,
+            self.colx1, self.colx2, self.xf_idx).get()
+
+class NumberCell(object):
+    __slots__ = ["rowx", "colx", "xf_idx", "number"]
+
+    def __init__(self, rowx, colx, xf_idx, number):
+        self.rowx = rowx
+        self.colx = colx
+        self.xf_idx = xf_idx
+        self.number = float(number)
+
+    def get_encoded_data(self):
+        rk_encoded = 0
+        num = self.number
+
+        # The four possible kinds of RK encoding are *not* mutually exclusive.
+        # The 30-bit integer variety picks up the most.
+        # In the code below, the four varieties are checked in descending order
+        # of bangs per buck, or not at all.
+        # SJM 2007-10-01
+
+        if -0x20000000 <= num < 0x20000000: # fits in 30-bit *signed* int
+            inum = int(num)
+            if inum == num: # survives round-trip
+                # print "30-bit integer RK", inum, hex(inum)
+                rk_encoded = 2 | (inum << 2)
+                return 1, rk_encoded
+
+        temp = num * 100
+
+        if -0x20000000 <= temp < 0x20000000:
+            # That was step 1: the coded value will fit in
+            # a 30-bit signed integer.
+            itemp = int(round(temp, 0))
+            # That was step 2: "itemp" is the best candidate coded value.
+            # Now for step 3: simulate the decoding,
+            # to check for round-trip correctness.
+            if itemp / 100.0 == num:
+                # print "30-bit integer RK*100", itemp, hex(itemp)
+                rk_encoded = 3 | (itemp << 2)
+                return 1, rk_encoded
+
+        if 0: # Cost of extra pack+unpack not justified by tiny yield.
+            packed = pack('<d', num)
+            w01, w23 = unpack('<2i', packed)
+            if not w01 and not(w23 & 3):
+                # 34 lsb are 0
+                # print "float RK", w23, hex(w23)
+                return 1, w23
+
+            packed100 = pack('<d', temp)
+            w01, w23 = unpack('<2i', packed100)
+            if not w01 and not(w23 & 3):
+                # 34 lsb are 0
+                # print "float RK*100", w23, hex(w23)
+                return 1, w23 | 1
+
+        #print "Number"
+        #print
+        return 0, pack('<5Hd', 0x0203, 14, self.rowx, self.colx, self.xf_idx, num)
+
+    def get_biff_data(self):
+        isRK, value = self.get_encoded_data()
+        if isRK:
+            return pack('<5Hi', 0x27E, 10, self.rowx, self.colx, self.xf_idx, value)
+        return value # NUMBER record already packed
+
+class BooleanCell(object):
+    __slots__ = ["rowx", "colx", "xf_idx", "number"]
+
+    def __init__(self, rowx, colx, xf_idx, number):
+        self.rowx = rowx
+        self.colx = colx
+        self.xf_idx = xf_idx
+        self.number = number
+
+    def get_biff_data(self):
+        return BIFFRecords.BoolErrRecord(self.rowx,
+            self.colx, self.xf_idx, self.number, 0).get()
+
+error_code_map = {
+    0x00:  0, # Intersection of two cell ranges is empty
+    0x07:  7, # Division by zero
+    0x0F: 15, # Wrong type of operand
+    0x17: 23, # Illegal or deleted cell reference
+    0x1D: 29, # Wrong function or range name
+    0x24: 36, # Value range overflow
+    0x2A: 42, # Argument or function not available
+    '#NULL!' :  0, # Intersection of two cell ranges is empty
+    '#DIV/0!':  7, # Division by zero
+    '#VALUE!': 36, # Wrong type of operand
+    '#REF!'  : 23, # Illegal or deleted cell reference
+    '#NAME?' : 29, # Wrong function or range name
+    '#NUM!'  : 36, # Value range overflow
+    '#N/A!'  : 42, # Argument or function not available
+}
+
+class ErrorCell(object):
+    __slots__ = ["rowx", "colx", "xf_idx", "number"]
+
+    def __init__(self, rowx, colx, xf_idx, error_string_or_code):
+        self.rowx = rowx
+        self.colx = colx
+        self.xf_idx = xf_idx
+        try:
+            self.number = error_code_map[error_string_or_code]
+        except KeyError:
+            raise Exception('Illegal error value (%r)' % error_string_or_code)
+
+    def get_biff_data(self):
+        return BIFFRecords.BoolErrRecord(self.rowx,
+            self.colx, self.xf_idx, self.number, 1).get()
+
+class FormulaCell(object):
+    __slots__ = ["rowx", "colx", "xf_idx", "frmla", "calc_flags"]
+
+    def __init__(self, rowx, colx, xf_idx, frmla, calc_flags=0):
+        self.rowx = rowx
+        self.colx = colx
+        self.xf_idx = xf_idx
+        self.frmla = frmla
+        self.calc_flags = calc_flags
+
+    def get_biff_data(self):
+        return BIFFRecords.FormulaRecord(self.rowx,
+            self.colx, self.xf_idx, self.frmla.rpn(), self.calc_flags).get()
+
+# module-level function for *internal* use by the Row module
+
+def _get_cells_biff_data_mul(rowx, cell_items):
+    # Return the BIFF data for all cell records in the row.
+    # Adjacent BLANK|RK records are combined into MUL(BLANK|RK) records.
+    pieces = []
+    nitems = len(cell_items)
+    i = 0
+    while i < nitems:
+        icolx, icell = cell_items[i]
+        if isinstance(icell, NumberCell):
+            isRK, value = icell.get_encoded_data()
+            if not isRK:
+                pieces.append(value) # pre-packed NUMBER record
+                i += 1
+                continue
+            muldata = [(value, icell.xf_idx)]
+            target = NumberCell
+        elif isinstance(icell, BlankCell):
+            muldata = [icell.xf_idx]
+            target = BlankCell
+        else:
+            pieces.append(icell.get_biff_data())
+            i += 1
+            continue
+        lastcolx = icolx
+        j = i
+        packed_record = ''
+        for j in xrange(i+1, nitems):
+            jcolx, jcell = cell_items[j]
+            if jcolx != lastcolx + 1:
+                nexti = j
+                break
+            if not isinstance(jcell, target):
+                nexti = j
+                break
+            if target == NumberCell:
+                isRK, value = jcell.get_encoded_data()
+                if not isRK:
+                    packed_record = value
+                    nexti = j + 1
+                    break
+                muldata.append((value, jcell.xf_idx))
+            else:
+                muldata.append(jcell.xf_idx)
+            lastcolx = jcolx
+        else:
+            nexti = j + 1
+        if target == NumberCell:
+            if lastcolx == icolx:
+                # RK record
+                value, xf_idx = muldata[0]
+                pieces.append(pack('<5Hi', 0x027E, 10, rowx, icolx, xf_idx, value))
+            else:
+                # MULRK record
+                nc = lastcolx - icolx + 1
+                pieces.append(pack('<4H', 0x00BD, 6 * nc + 6, rowx, icolx))
+                pieces.append(''.join([pack('<Hi', xf_idx, value) for value, xf_idx in muldata]))
+                pieces.append(pack('<H', lastcolx))
+        else:
+            if lastcolx == icolx:
+                # BLANK record
+                xf_idx = muldata[0]
+                pieces.append(pack('<5H', 0x0201, 6, rowx, icolx, xf_idx))
+            else:
+                # MULBLANK record
+                nc = lastcolx - icolx + 1
+                pieces.append(pack('<4H', 0x00BE, 2 * nc + 6, rowx, icolx))
+                pieces.append(''.join([pack('<H', xf_idx) for xf_idx in muldata]))
+                pieces.append(pack('<H', lastcolx))
+        if packed_record:
+            pieces.append(packed_record)
+        i = nexti
+    return ''.join(pieces)
+

tablib/packages/xlwt/Column.py

@@ -1,34 +1,34 @@
-# -*- coding: windows-1252 -*-
-
-from .BIFFRecords import ColInfoRecord
-
-class Column(object):
-    def __init__(self, colx, parent_sheet):
-        if not(isinstance(colx, int) and 0 <= colx <= 255):
-            raise ValueError("column index (%r) not an int in range(256)" % colx)
-        self._index = colx
-        self._parent = parent_sheet
-        self._parent_wb = parent_sheet.get_parent()
-        self._xf_index = 0x0F
-
-        self.width = 0x0B92
-        self.hidden = 0
-        self.level = 0
-        self.collapse = 0
-
-    def set_style(self, style):
-        self._xf_index = self._parent_wb.add_style(style)
-
-    def width_in_pixels(self):
-        # *** Approximation ****
-        return int(round(self.width * 0.0272 + 0.446, 0))
-
-    def get_biff_record(self):
-        options =  (self.hidden & 0x01) << 0
-        options |= (self.level & 0x07) << 8
-        options |= (self.collapse & 0x01) << 12
-
-        return ColInfoRecord(self._index, self._index, self.width, self._xf_index, options).get()
-
-
-
+# -*- coding: windows-1252 -*-
+
+from BIFFRecords import ColInfoRecord
+
+class Column(object):
+    def __init__(self, colx, parent_sheet):
+        if not(isinstance(colx, int) and 0 <= colx <= 255):
+            raise ValueError("column index (%r) not an int in range(256)" % colx)
+        self._index = colx
+        self._parent = parent_sheet
+        self._parent_wb = parent_sheet.get_parent()
+        self._xf_index = 0x0F
+
+        self.width = 0x0B92
+        self.hidden = 0
+        self.level = 0
+        self.collapse = 0
+
+    def set_style(self, style):
+        self._xf_index = self._parent_wb.add_style(style)
+
+    def width_in_pixels(self):
+        # *** Approximation ****
+        return int(round(self.width * 0.0272 + 0.446, 0))
+
+    def get_biff_record(self):
+        options =  (self.hidden & 0x01) << 0
+        options |= (self.level & 0x07) << 8
+        options |= (self.collapse & 0x01) << 12
+
+        return ColInfoRecord(self._index, self._index, self.width, self._xf_index, options).get()
+
+
+

tablib/packages/xlwt/ExcelFormula.py

@@ -1,41 +1,43 @@
-from . import ExcelFormulaParser, ExcelFormulaLexer
-import struct
-from .antlr import ANTLRException
-
-
-class Formula(object):
-    __slots__ = ["__init__",  "__s", "__parser", "__sheet_refs", "__xcall_refs"]
-
-
-    def __init__(self, s):
-        try:
-            self.__s = s
-            lexer = ExcelFormulaLexer.Lexer(s)
-            self.__parser = ExcelFormulaParser.Parser(lexer)
-            self.__parser.formula()
-            self.__sheet_refs = self.__parser.sheet_references
-            self.__xcall_refs = self.__parser.xcall_references
-        except ANTLRException as e:
-            # print e
-            raise ExcelFormulaParser.FormulaParseException("can't parse formula " + s)
-
-    def get_references(self):
-        return self.__sheet_refs, self.__xcall_refs
-
-    def patch_references(self, patches):
-        for offset, idx in patches:
-            self.__parser.rpn = self.__parser.rpn[:offset] + struct.pack('<H', idx) + self.__parser.rpn[offset+2:]
-
-    def text(self):
-        return self.__s
-
-    def rpn(self):
-        '''
-        Offset    Size    Contents
-        0         2       Size of the following formula data (sz)
-        2         sz      Formula data (RPN token array)
-        [2+sz]    var.    (optional) Additional data for specific tokens
-
-        '''
-        return struct.pack("<H", len(self.__parser.rpn)) + self.__parser.rpn
-
+# -*- coding: windows-1252 -*-
+
+import ExcelFormulaParser, ExcelFormulaLexer
+import struct
+from antlr import ANTLRException
+
+
+class Formula(object):
+    __slots__ = ["__init__",  "__s", "__parser", "__sheet_refs", "__xcall_refs"]
+
+
+    def __init__(self, s):
+        try:
+            self.__s = s
+            lexer = ExcelFormulaLexer.Lexer(s)
+            self.__parser = ExcelFormulaParser.Parser(lexer)
+            self.__parser.formula()
+            self.__sheet_refs = self.__parser.sheet_references
+            self.__xcall_refs = self.__parser.xcall_references
+        except ANTLRException, e:
+            # print e
+            raise ExcelFormulaParser.FormulaParseException, "can't parse formula " + s
+
+    def get_references(self):
+        return self.__sheet_refs, self.__xcall_refs
+
+    def patch_references(self, patches):
+        for offset, idx in patches:
+            self.__parser.rpn = self.__parser.rpn[:offset] + struct.pack('<H', idx) + self.__parser.rpn[offset+2:]
+
+    def text(self):
+        return self.__s
+
+    def rpn(self):
+        '''
+        Offset    Size    Contents
+        0         2       Size of the following formula data (sz)
+        2         sz      Formula data (RPN token array)
+        [2+sz]    var.    (optional) Additional data for specific tokens
+
+        '''
+        return struct.pack("<H", len(self.__parser.rpn)) + self.__parser.rpn
+

tablib/packages/xlwt/ExcelFormulaLexer.py

@@ -1,126 +1,128 @@
-import sys
-from .antlr import EOF, CommonToken as Tok, TokenStream, TokenStreamException
-import struct
-from . import ExcelFormulaParser
-from re import compile as recompile, match, LOCALE, UNICODE, IGNORECASE, VERBOSE
-
-
-int_const_pattern = r"\d+\b"
-flt_const_pattern = r"""
-    (?:
-        (?: \d* \. \d+ ) # .1 .12 .123 etc 9.1 etc 98.1 etc
-        |
-        (?: \d+ \. ) # 1. 12. 123. etc
-    )
-    # followed by optional exponent part
-    (?: [Ee] [+-]? \d+ ) ?
-    """
-str_const_pattern = r'"(?:[^"]|"")*"'
-#range2d_pattern   = recompile(r"\$?[A-I]?[A-Z]\$?\d+:\$?[A-I]?[A-Z]\$?\d+"
-ref2d_r1c1_pattern = r"[Rr]0*[1-9][0-9]*[Cc]0*[1-9][0-9]*"
-ref2d_pattern     = r"\$?[A-I]?[A-Z]\$?0*[1-9][0-9]*"
-true_pattern      = r"TRUE\b"
-false_pattern     = r"FALSE\b"
-if_pattern        = r"IF\b"
-choose_pattern    = r"CHOOSE\b"
-name_pattern      = r"\w[\.\w]*"
-quotename_pattern = r"'(?:[^']|'')*'" #### It's essential that this bracket be non-grouping.
-ne_pattern        = r"<>"
-ge_pattern        = r">="
-le_pattern        = r"<="
-
-pattern_type_tuples = (
-    (flt_const_pattern, ExcelFormulaParser.NUM_CONST),
-    (int_const_pattern, ExcelFormulaParser.INT_CONST),
-    (str_const_pattern, ExcelFormulaParser.STR_CONST),
-#    (range2d_pattern  , ExcelFormulaParser.RANGE2D),
-    (ref2d_r1c1_pattern, ExcelFormulaParser.REF2D_R1C1),
-    (ref2d_pattern    , ExcelFormulaParser.REF2D),
-    (true_pattern     , ExcelFormulaParser.TRUE_CONST),
-    (false_pattern    , ExcelFormulaParser.FALSE_CONST),
-    (if_pattern       , ExcelFormulaParser.FUNC_IF),
-    (choose_pattern   , ExcelFormulaParser.FUNC_CHOOSE),
-    (name_pattern     , ExcelFormulaParser.NAME),
-    (quotename_pattern, ExcelFormulaParser.QUOTENAME),
-    (ne_pattern,        ExcelFormulaParser.NE),
-    (ge_pattern,        ExcelFormulaParser.GE),
-    (le_pattern,        ExcelFormulaParser.LE),
-)
-
-_re = recompile(
-    '(' + ')|('.join([i[0] for i in pattern_type_tuples]) + ')',
-    VERBOSE+LOCALE+IGNORECASE)
-
-_toktype = [None] + [i[1] for i in pattern_type_tuples]
-# need dummy at start because re.MatchObject.lastindex counts from 1
-
-single_char_lookup = {
-    '=': ExcelFormulaParser.EQ,
-    '<': ExcelFormulaParser.LT,
-    '>': ExcelFormulaParser.GT,
-    '+': ExcelFormulaParser.ADD,
-    '-': ExcelFormulaParser.SUB,
-    '*': ExcelFormulaParser.MUL,
-    '/': ExcelFormulaParser.DIV,
-    ':': ExcelFormulaParser.COLON,
-    ';': ExcelFormulaParser.SEMICOLON,
-    ',': ExcelFormulaParser.COMMA,
-    '(': ExcelFormulaParser.LP,
-    ')': ExcelFormulaParser.RP,
-    '&': ExcelFormulaParser.CONCAT,
-    '%': ExcelFormulaParser.PERCENT,
-    '^': ExcelFormulaParser.POWER,
-    '!': ExcelFormulaParser.BANG,
-    }
-
-class Lexer(TokenStream):
-    def __init__(self, text):
-        self._text = text[:]
-        self._pos = 0
-        self._line = 0
-
-    def isEOF(self):
-        return len(self._text) <= self._pos
-
-    def curr_ch(self):
-        return self._text[self._pos]
-
-    def next_ch(self, n = 1):
-        self._pos += n
-
-    def is_whitespace(self):
-        return self.curr_ch() in " \t\n\r\f\v"
-
-    def match_pattern(self):
-        m = _re.match(self._text, self._pos)
-        if not m:
-            return None
-        self._pos = m.end(0)
-        return Tok(type = _toktype[m.lastindex], text = m.group(0), col = m.start(0) + 1)
-
-    def nextToken(self):
-        # skip whitespace
-        while not self.isEOF() and self.is_whitespace():
-            self.next_ch()
-        if self.isEOF():
-            return Tok(type = EOF)
-        # first, try to match token with 2 or more chars
-        t = self.match_pattern()
-        if t:
-            return t
-        # second, we want 1-char tokens
-        te = self.curr_ch()
-        try:
-            ty = single_char_lookup[te]
-        except KeyError:
-            raise TokenStreamException(
-                "Unexpected char %r in column %u." % (self.curr_ch(), self._pos))
-        self.next_ch()
-        return Tok(type=ty, text=te, col=self._pos)
-
-if __name__ == '__main__':
-    try:
-        for t in Lexer(""" 1.23 456 "abcd" R2C2 a1 iv65536 true false if choose a_name 'qname' <> >= <= """):
-            print(t)
-    except TokenStreamException as e:
-        print("error:", e)
+# -*- coding: windows-1252 -*-
+
+import sys
+from antlr import EOF, CommonToken as Tok, TokenStream, TokenStreamException
+import struct
+import ExcelFormulaParser
+from re import compile as recompile, match, LOCALE, UNICODE, IGNORECASE, VERBOSE
+
+
+int_const_pattern = r"\d+\b"
+flt_const_pattern = r"""
+    (?:
+        (?: \d* \. \d+ ) # .1 .12 .123 etc 9.1 etc 98.1 etc
+        |
+        (?: \d+ \. ) # 1. 12. 123. etc
+    )
+    # followed by optional exponent part
+    (?: [Ee] [+-]? \d+ ) ?
+    """
+str_const_pattern = r'"(?:[^"]|"")*"'
+#range2d_pattern   = recompile(r"\$?[A-I]?[A-Z]\$?\d+:\$?[A-I]?[A-Z]\$?\d+"
+ref2d_r1c1_pattern = r"[Rr]0*[1-9][0-9]*[Cc]0*[1-9][0-9]*"
+ref2d_pattern     = r"\$?[A-I]?[A-Z]\$?0*[1-9][0-9]*"
+true_pattern      = r"TRUE\b"
+false_pattern     = r"FALSE\b"
+if_pattern        = r"IF\b"
+choose_pattern    = r"CHOOSE\b"
+name_pattern      = r"\w[\.\w]*"
+quotename_pattern = r"'(?:[^']|'')*'" #### It's essential that this bracket be non-grouping.
+ne_pattern        = r"<>"
+ge_pattern        = r">="
+le_pattern        = r"<="
+
+pattern_type_tuples = (
+    (flt_const_pattern, ExcelFormulaParser.NUM_CONST),
+    (int_const_pattern, ExcelFormulaParser.INT_CONST),
+    (str_const_pattern, ExcelFormulaParser.STR_CONST),
+#    (range2d_pattern  , ExcelFormulaParser.RANGE2D),
+    (ref2d_r1c1_pattern, ExcelFormulaParser.REF2D_R1C1),
+    (ref2d_pattern    , ExcelFormulaParser.REF2D),
+    (true_pattern     , ExcelFormulaParser.TRUE_CONST),
+    (false_pattern    , ExcelFormulaParser.FALSE_CONST),
+    (if_pattern       , ExcelFormulaParser.FUNC_IF),
+    (choose_pattern   , ExcelFormulaParser.FUNC_CHOOSE),
+    (name_pattern     , ExcelFormulaParser.NAME),
+    (quotename_pattern, ExcelFormulaParser.QUOTENAME),
+    (ne_pattern,        ExcelFormulaParser.NE),
+    (ge_pattern,        ExcelFormulaParser.GE),
+    (le_pattern,        ExcelFormulaParser.LE),
+)
+
+_re = recompile(
+    '(' + ')|('.join([i[0] for i in pattern_type_tuples]) + ')',
+    VERBOSE+LOCALE+IGNORECASE)
+
+_toktype = [None] + [i[1] for i in pattern_type_tuples]
+# need dummy at start because re.MatchObject.lastindex counts from 1
+
+single_char_lookup = {
+    '=': ExcelFormulaParser.EQ,
+    '<': ExcelFormulaParser.LT,
+    '>': ExcelFormulaParser.GT,
+    '+': ExcelFormulaParser.ADD,
+    '-': ExcelFormulaParser.SUB,
+    '*': ExcelFormulaParser.MUL,
+    '/': ExcelFormulaParser.DIV,
+    ':': ExcelFormulaParser.COLON,
+    ';': ExcelFormulaParser.SEMICOLON,
+    ',': ExcelFormulaParser.COMMA,
+    '(': ExcelFormulaParser.LP,
+    ')': ExcelFormulaParser.RP,
+    '&': ExcelFormulaParser.CONCAT,
+    '%': ExcelFormulaParser.PERCENT,
+    '^': ExcelFormulaParser.POWER,
+    '!': ExcelFormulaParser.BANG,
+    }
+
+class Lexer(TokenStream):
+    def __init__(self, text):
+        self._text = text[:]
+        self._pos = 0
+        self._line = 0
+
+    def isEOF(self):
+        return len(self._text) <= self._pos
+
+    def curr_ch(self):
+        return self._text[self._pos]
+
+    def next_ch(self, n = 1):
+        self._pos += n
+
+    def is_whitespace(self):
+        return self.curr_ch() in " \t\n\r\f\v"
+
+    def match_pattern(self):
+        m = _re.match(self._text, self._pos)
+        if not m:
+            return None
+        self._pos = m.end(0)
+        return Tok(type = _toktype[m.lastindex], text = m.group(0), col = m.start(0) + 1)
+
+    def nextToken(self):
+        # skip whitespace
+        while not self.isEOF() and self.is_whitespace():
+            self.next_ch()
+        if self.isEOF():
+            return Tok(type = EOF)
+        # first, try to match token with 2 or more chars
+        t = self.match_pattern()
+        if t:
+            return t
+        # second, we want 1-char tokens
+        te = self.curr_ch()
+        try:
+            ty = single_char_lookup[te]
+        except KeyError:
+            raise TokenStreamException(
+                "Unexpected char %r in column %u." % (self.curr_ch(), self._pos))
+        self.next_ch()
+        return Tok(type=ty, text=te, col=self._pos)
+
+if __name__ == '__main__':
+    try:
+        for t in Lexer(""" 1.23 456 "abcd" R2C2 a1 iv65536 true false if choose a_name 'qname' <> >= <= """):
+            print t
+    except TokenStreamException, e:
+        print "error:", e

tablib/packages/xlwt/Formatting.py

@@ -1,261 +1,261 @@
-#!/usr/bin/env python
-'''
-The  XF  record is able to store explicit cell formatting attributes or the
-attributes  of  a cell style. Explicit formatting includes the reference to
-a  cell  style  XF  record. This allows to extend a defined cell style with
-some  explicit  attributes.  The  formatting  attributes  are  divided into
-6 groups:
-
-Group           Attributes
--------------------------------------
-Number format   Number format index (index to FORMAT record)
-Font            Font index (index to FONT record)
-Alignment       Horizontal and vertical alignment, text wrap, indentation,
-                orientation/rotation, text direction
-Border          Border line styles and colours
-Background      Background area style and colours
-Protection      Cell locked, formula hidden
-
-For  each  group  a flag in the cell XF record specifies whether to use the
-attributes  contained  in  that  XF  record  or  in  the  referenced  style
-XF  record. In style XF records, these flags specify whether the attributes
-will  overwrite  explicit  cell  formatting  when  the  style is applied to
-a  cell. Changing a cell style (without applying this style to a cell) will
-change  all  cells which already use that style and do not contain explicit
-cell  attributes for the changed style attributes. If a cell XF record does
-not  contain  explicit  attributes  in a group (if the attribute group flag
-is not set), it repeats the attributes of its style XF record.
-
-'''
-
-from . import BIFFRecords
-
-class Font(object):
-
-    ESCAPEMENT_NONE         = 0x00
-    ESCAPEMENT_SUPERSCRIPT  = 0x01
-    ESCAPEMENT_SUBSCRIPT    = 0x02
-
-    UNDERLINE_NONE          = 0x00
-    UNDERLINE_SINGLE        = 0x01
-    UNDERLINE_SINGLE_ACC    = 0x21
-    UNDERLINE_DOUBLE        = 0x02
-    UNDERLINE_DOUBLE_ACC    = 0x22
-
-    FAMILY_NONE         = 0x00
-    FAMILY_ROMAN        = 0x01
-    FAMILY_SWISS        = 0x02
-    FAMILY_MODERN       = 0x03
-    FAMILY_SCRIPT       = 0x04
-    FAMILY_DECORATIVE   = 0x05
-
-    CHARSET_ANSI_LATIN          = 0x00
-    CHARSET_SYS_DEFAULT         = 0x01
-    CHARSET_SYMBOL              = 0x02
-    CHARSET_APPLE_ROMAN         = 0x4D
-    CHARSET_ANSI_JAP_SHIFT_JIS  = 0x80
-    CHARSET_ANSI_KOR_HANGUL     = 0x81
-    CHARSET_ANSI_KOR_JOHAB      = 0x82
-    CHARSET_ANSI_CHINESE_GBK    = 0x86
-    CHARSET_ANSI_CHINESE_BIG5   = 0x88
-    CHARSET_ANSI_GREEK          = 0xA1
-    CHARSET_ANSI_TURKISH        = 0xA2
-    CHARSET_ANSI_VIETNAMESE     = 0xA3
-    CHARSET_ANSI_HEBREW         = 0xB1
-    CHARSET_ANSI_ARABIC         = 0xB2
-    CHARSET_ANSI_BALTIC         = 0xBA
-    CHARSET_ANSI_CYRILLIC       = 0xCC
-    CHARSET_ANSI_THAI           = 0xDE
-    CHARSET_ANSI_LATIN_II       = 0xEE
-    CHARSET_OEM_LATIN_I         = 0xFF
-
-    def __init__(self):
-        # twip = 1/20 of a point = 1/1440 of a inch
-        # usually resolution == 96 pixels per 1 inch
-        # (rarely 120 pixels per 1 inch or another one)
-
-        self.height = 0x00C8 # 200: this is font with height 10 points
-        self.italic = False
-        self.struck_out = False
-        self.outline = False
-        self.shadow = False
-        self.colour_index = 0x7FFF
-        self.bold = False
-        self._weight = 0x0190 # 0x02BC gives bold font
-        self.escapement = self.ESCAPEMENT_NONE
-        self.underline = self.UNDERLINE_NONE
-        self.family = self.FAMILY_NONE
-        self.charset = self.CHARSET_SYS_DEFAULT
-        self.name = b'Arial'
-
-    def get_biff_record(self):
-        height = self.height
-
-        options = 0x00
-        if self.bold:
-            options |= 0x01
-            self._weight = 0x02BC
-        if self.italic:
-            options |= 0x02
-        if self.underline != self.UNDERLINE_NONE:
-            options |= 0x04
-        if self.struck_out:
-            options |= 0x08
-        if self.outline:
-            options |= 0x010
-        if self.shadow:
-            options |= 0x020
-
-        colour_index = self.colour_index
-        weight = self._weight
-        escapement = self.escapement
-        underline = self.underline
-        family = self.family
-        charset = self.charset
-        name = self.name
-
-        return BIFFRecords.FontRecord(height, options, colour_index, weight, escapement,
-                    underline, family, charset,
-                    name)
-
-    def _search_key(self):
-        return (
-            self.height,
-            self.italic,
-            self.struck_out,
-            self.outline,
-            self.shadow,
-            self.colour_index,
-            self.bold,
-            self._weight,
-            self.escapement,
-            self.underline,
-            self.family,
-            self.charset,
-            self.name,
-            )
-
-class Alignment(object):
-    HORZ_GENERAL                = 0x00
-    HORZ_LEFT                   = 0x01
-    HORZ_CENTER                 = 0x02
-    HORZ_RIGHT                  = 0x03
-    HORZ_FILLED                 = 0x04
-    HORZ_JUSTIFIED              = 0x05 # BIFF4-BIFF8X
-    HORZ_CENTER_ACROSS_SEL      = 0x06 # Centred across selection (BIFF4-BIFF8X)
-    HORZ_DISTRIBUTED            = 0x07 # Distributed (BIFF8X)
-
-    VERT_TOP                    = 0x00
-    VERT_CENTER                 = 0x01
-    VERT_BOTTOM                 = 0x02
-    VERT_JUSTIFIED              = 0x03 # Justified (BIFF5-BIFF8X)
-    VERT_DISTRIBUTED            = 0x04 # Distributed (BIFF8X)
-
-    DIRECTION_GENERAL           = 0x00 # BIFF8X
-    DIRECTION_LR                = 0x01
-    DIRECTION_RL                = 0x02
-
-    ORIENTATION_NOT_ROTATED     = 0x00
-    ORIENTATION_STACKED         = 0x01
-    ORIENTATION_90_CC           = 0x02
-    ORIENTATION_90_CW           = 0x03
-
-    ROTATION_0_ANGLE            = 0x00
-    ROTATION_STACKED            = 0xFF
-
-    WRAP_AT_RIGHT               = 0x01
-    NOT_WRAP_AT_RIGHT           = 0x00
-
-    SHRINK_TO_FIT               = 0x01
-    NOT_SHRINK_TO_FIT           = 0x00
-
-    def __init__(self):
-        self.horz = self.HORZ_GENERAL
-        self.vert = self.VERT_BOTTOM
-        self.dire = self.DIRECTION_GENERAL
-        self.orie = self.ORIENTATION_NOT_ROTATED
-        self.rota = self.ROTATION_0_ANGLE
-        self.wrap = self.NOT_WRAP_AT_RIGHT
-        self.shri = self.NOT_SHRINK_TO_FIT
-        self.inde = 0
-        self.merg = 0
-
-    def _search_key(self):
-        return (
-            self.horz, self.vert, self.dire, self.orie, self.rota,
-            self.wrap, self.shri, self.inde, self.merg,
-            )
-
-class Borders(object):
-    NO_LINE = 0x00
-    THIN    = 0x01
-    MEDIUM  = 0x02
-    DASHED  = 0x03
-    DOTTED  = 0x04
-    THICK   = 0x05
-    DOUBLE  = 0x06
-    HAIR    = 0x07
-    #The following for BIFF8
-    MEDIUM_DASHED               = 0x08
-    THIN_DASH_DOTTED            = 0x09
-    MEDIUM_DASH_DOTTED          = 0x0A
-    THIN_DASH_DOT_DOTTED        = 0x0B
-    MEDIUM_DASH_DOT_DOTTED      = 0x0C
-    SLANTED_MEDIUM_DASH_DOTTED  = 0x0D
-
-    NEED_DIAG1      = 0x01
-    NEED_DIAG2      = 0x01
-    NO_NEED_DIAG1   = 0x00
-    NO_NEED_DIAG2   = 0x00
-
-    def __init__(self):
-        self.left   = self.NO_LINE
-        self.right  = self.NO_LINE
-        self.top    = self.NO_LINE
-        self.bottom = self.NO_LINE
-        self.diag   = self.NO_LINE
-
-        self.left_colour   = 0x40
-        self.right_colour  = 0x40
-        self.top_colour    = 0x40
-        self.bottom_colour = 0x40
-        self.diag_colour   = 0x40
-
-        self.need_diag1 = self.NO_NEED_DIAG1
-        self.need_diag2 = self.NO_NEED_DIAG2
-
-    def _search_key(self):
-        return (
-             self.left, self.right, self.top, self.bottom, self.diag,
-             self.left_colour, self.right_colour, self.top_colour,
-             self.bottom_colour, self.diag_colour,
-             self.need_diag1, self.need_diag2,
-            )
-
-class Pattern(object):
-    # patterns 0x00 - 0x12
-    NO_PATTERN      = 0x00
-    SOLID_PATTERN   = 0x01
-
-    def __init__(self):
-        self.pattern = self.NO_PATTERN
-        self.pattern_fore_colour = 0x40
-        self.pattern_back_colour = 0x41
-
-    def _search_key(self):
-        return (
-            self.pattern,
-            self.pattern_fore_colour,
-            self.pattern_back_colour,
-            )
-
-class Protection(object):
-    def __init__(self):
-        self.cell_locked = 1
-        self.formula_hidden = 0
-
-    def _search_key(self):
-        return (
-            self.cell_locked,
-            self.formula_hidden,
-            )
+#!/usr/bin/env python
+'''
+The  XF  record is able to store explicit cell formatting attributes or the
+attributes  of  a cell style. Explicit formatting includes the reference to
+a  cell  style  XF  record. This allows to extend a defined cell style with
+some  explicit  attributes.  The  formatting  attributes  are  divided into
+6 groups:
+
+Group           Attributes
+-------------------------------------
+Number format   Number format index (index to FORMAT record)
+Font            Font index (index to FONT record)
+Alignment       Horizontal and vertical alignment, text wrap, indentation,
+                orientation/rotation, text direction
+Border          Border line styles and colours
+Background      Background area style and colours
+Protection      Cell locked, formula hidden
+
+For  each  group  a flag in the cell XF record specifies whether to use the
+attributes  contained  in  that  XF  record  or  in  the  referenced  style
+XF  record. In style XF records, these flags specify whether the attributes
+will  overwrite  explicit  cell  formatting  when  the  style is applied to
+a  cell. Changing a cell style (without applying this style to a cell) will
+change  all  cells which already use that style and do not contain explicit
+cell  attributes for the changed style attributes. If a cell XF record does
+not  contain  explicit  attributes  in a group (if the attribute group flag
+is not set), it repeats the attributes of its style XF record.
+
+'''
+
+import BIFFRecords
+
+class Font(object):
+
+    ESCAPEMENT_NONE         = 0x00
+    ESCAPEMENT_SUPERSCRIPT  = 0x01
+    ESCAPEMENT_SUBSCRIPT    = 0x02
+
+    UNDERLINE_NONE          = 0x00
+    UNDERLINE_SINGLE        = 0x01
+    UNDERLINE_SINGLE_ACC    = 0x21
+    UNDERLINE_DOUBLE        = 0x02
+    UNDERLINE_DOUBLE_ACC    = 0x22
+
+    FAMILY_NONE         = 0x00
+    FAMILY_ROMAN        = 0x01
+    FAMILY_SWISS        = 0x02
+    FAMILY_MODERN       = 0x03
+    FAMILY_SCRIPT       = 0x04
+    FAMILY_DECORATIVE   = 0x05
+
+    CHARSET_ANSI_LATIN          = 0x00
+    CHARSET_SYS_DEFAULT         = 0x01
+    CHARSET_SYMBOL              = 0x02
+    CHARSET_APPLE_ROMAN         = 0x4D
+    CHARSET_ANSI_JAP_SHIFT_JIS  = 0x80
+    CHARSET_ANSI_KOR_HANGUL     = 0x81
+    CHARSET_ANSI_KOR_JOHAB      = 0x82
+    CHARSET_ANSI_CHINESE_GBK    = 0x86
+    CHARSET_ANSI_CHINESE_BIG5   = 0x88
+    CHARSET_ANSI_GREEK          = 0xA1
+    CHARSET_ANSI_TURKISH        = 0xA2
+    CHARSET_ANSI_VIETNAMESE     = 0xA3
+    CHARSET_ANSI_HEBREW         = 0xB1
+    CHARSET_ANSI_ARABIC         = 0xB2
+    CHARSET_ANSI_BALTIC         = 0xBA
+    CHARSET_ANSI_CYRILLIC       = 0xCC
+    CHARSET_ANSI_THAI           = 0xDE
+    CHARSET_ANSI_LATIN_II       = 0xEE
+    CHARSET_OEM_LATIN_I         = 0xFF
+
+    def __init__(self):
+        # twip = 1/20 of a point = 1/1440 of a inch
+        # usually resolution == 96 pixels per 1 inch
+        # (rarely 120 pixels per 1 inch or another one)
+
+        self.height = 0x00C8 # 200: this is font with height 10 points
+        self.italic = False
+        self.struck_out = False
+        self.outline = False
+        self.shadow = False
+        self.colour_index = 0x7FFF
+        self.bold = False
+        self._weight = 0x0190 # 0x02BC gives bold font
+        self.escapement = self.ESCAPEMENT_NONE
+        self.underline = self.UNDERLINE_NONE
+        self.family = self.FAMILY_NONE
+        self.charset = self.CHARSET_SYS_DEFAULT
+        self.name = 'Arial'
+
+    def get_biff_record(self):
+        height = self.height
+
+        options = 0x00
+        if self.bold:
+            options |= 0x01
+            self._weight = 0x02BC
+        if self.italic:
+            options |= 0x02
+        if self.underline != self.UNDERLINE_NONE:
+            options |= 0x04
+        if self.struck_out:
+            options |= 0x08
+        if self.outline:
+            options |= 0x010
+        if self.shadow:
+            options |= 0x020
+
+        colour_index = self.colour_index
+        weight = self._weight
+        escapement = self.escapement
+        underline = self.underline
+        family = self.family
+        charset = self.charset
+        name = self.name
+
+        return BIFFRecords.FontRecord(height, options, colour_index, weight, escapement,
+                    underline, family, charset,
+                    name)
+
+    def _search_key(self):
+        return (
+            self.height,
+            self.italic,
+            self.struck_out,
+            self.outline,
+            self.shadow,
+            self.colour_index,
+            self.bold,
+            self._weight,
+            self.escapement,
+            self.underline,
+            self.family,
+            self.charset,
+            self.name,
+            )
+
+class Alignment(object):
+    HORZ_GENERAL                = 0x00
+    HORZ_LEFT                   = 0x01
+    HORZ_CENTER                 = 0x02
+    HORZ_RIGHT                  = 0x03
+    HORZ_FILLED                 = 0x04
+    HORZ_JUSTIFIED              = 0x05 # BIFF4-BIFF8X
+    HORZ_CENTER_ACROSS_SEL      = 0x06 # Centred across selection (BIFF4-BIFF8X)
+    HORZ_DISTRIBUTED            = 0x07 # Distributed (BIFF8X)
+
+    VERT_TOP                    = 0x00
+    VERT_CENTER                 = 0x01
+    VERT_BOTTOM                 = 0x02
+    VERT_JUSTIFIED              = 0x03 # Justified (BIFF5-BIFF8X)
+    VERT_DISTRIBUTED            = 0x04 # Distributed (BIFF8X)
+
+    DIRECTION_GENERAL           = 0x00 # BIFF8X
+    DIRECTION_LR                = 0x01
+    DIRECTION_RL                = 0x02
+
+    ORIENTATION_NOT_ROTATED     = 0x00
+    ORIENTATION_STACKED         = 0x01
+    ORIENTATION_90_CC           = 0x02
+    ORIENTATION_90_CW           = 0x03
+
+    ROTATION_0_ANGLE            = 0x00
+    ROTATION_STACKED            = 0xFF
+
+    WRAP_AT_RIGHT               = 0x01
+    NOT_WRAP_AT_RIGHT           = 0x00
+
+    SHRINK_TO_FIT               = 0x01
+    NOT_SHRINK_TO_FIT           = 0x00
+
+    def __init__(self):
+        self.horz = self.HORZ_GENERAL
+        self.vert = self.VERT_BOTTOM
+        self.dire = self.DIRECTION_GENERAL
+        self.orie = self.ORIENTATION_NOT_ROTATED
+        self.rota = self.ROTATION_0_ANGLE
+        self.wrap = self.NOT_WRAP_AT_RIGHT
+        self.shri = self.NOT_SHRINK_TO_FIT
+        self.inde = 0
+        self.merg = 0
+
+    def _search_key(self):
+        return (
+            self.horz, self.vert, self.dire, self.orie, self.rota,
+            self.wrap, self.shri, self.inde, self.merg,
+            )
+
+class Borders(object):
+    NO_LINE = 0x00
+    THIN    = 0x01
+    MEDIUM  = 0x02
+    DASHED  = 0x03
+    DOTTED  = 0x04
+    THICK   = 0x05
+    DOUBLE  = 0x06
+    HAIR    = 0x07
+    #The following for BIFF8
+    MEDIUM_DASHED               = 0x08
+    THIN_DASH_DOTTED            = 0x09
+    MEDIUM_DASH_DOTTED          = 0x0A
+    THIN_DASH_DOT_DOTTED        = 0x0B
+    MEDIUM_DASH_DOT_DOTTED      = 0x0C
+    SLANTED_MEDIUM_DASH_DOTTED  = 0x0D
+
+    NEED_DIAG1      = 0x01
+    NEED_DIAG2      = 0x01
+    NO_NEED_DIAG1   = 0x00
+    NO_NEED_DIAG2   = 0x00
+
+    def __init__(self):
+        self.left   = self.NO_LINE
+        self.right  = self.NO_LINE
+        self.top    = self.NO_LINE
+        self.bottom = self.NO_LINE
+        self.diag   = self.NO_LINE
+
+        self.left_colour   = 0x40
+        self.right_colour  = 0x40
+        self.top_colour    = 0x40
+        self.bottom_colour = 0x40
+        self.diag_colour   = 0x40
+
+        self.need_diag1 = self.NO_NEED_DIAG1
+        self.need_diag2 = self.NO_NEED_DIAG2
+
+    def _search_key(self):
+        return (
+             self.left, self.right, self.top, self.bottom, self.diag,
+             self.left_colour, self.right_colour, self.top_colour,
+             self.bottom_colour, self.diag_colour,
+             self.need_diag1, self.need_diag2,
+            )
+
+class Pattern(object):
+    # patterns 0x00 - 0x12
+    NO_PATTERN      = 0x00
+    SOLID_PATTERN   = 0x01
+
+    def __init__(self):
+        self.pattern = self.NO_PATTERN
+        self.pattern_fore_colour = 0x40
+        self.pattern_back_colour = 0x41
+
+    def _search_key(self):
+        return (
+            self.pattern,
+            self.pattern_fore_colour,
+            self.pattern_back_colour,
+            )
+
+class Protection(object):
+    def __init__(self):
+        self.cell_locked = 1
+        self.formula_hidden = 0
+
+    def _search_key(self):
+        return (
+            self.cell_locked,
+            self.formula_hidden,
+            )

tablib/packages/xlwt/Row.py

@@ -1,253 +1,253 @@
-# -*- coding: windows-1252 -*-
-
-from . import BIFFRecords
-from . import Style
-from .Cell import StrCell, BlankCell, NumberCell, FormulaCell, MulBlankCell, BooleanCell, ErrorCell, \
-    _get_cells_biff_data_mul
-from . import ExcelFormula
-import datetime as dt
-try:
-    from decimal import Decimal
-except ImportError:
-    # Python 2.3: decimal not supported; create dummy Decimal class
-    class Decimal(object):
-        pass
-
-
-class Row(object):
-    __slots__ = [# private variables
-                 "__idx",
-                 "__parent",
-                 "__parent_wb",
-                 "__cells",
-                 "__min_col_idx",
-                 "__max_col_idx",
-                 "__xf_index",
-                 "__has_default_xf_index",
-                 "__height_in_pixels",
-                 # public variables
-                 "height",
-                 "has_default_height",
-                 "height_mismatch",
-                 "level",
-                 "collapse",
-                 "hidden",
-                 "space_above",
-                 "space_below"]
-
-    def __init__(self, rowx, parent_sheet):
-        if not (isinstance(rowx, int) and 0 <= rowx <= 65535):
-            raise ValueError("row index (%r) not an int in range(65536)" % rowx)
-        self.__idx = rowx
-        self.__parent = parent_sheet
-        self.__parent_wb = parent_sheet.get_parent()
-        self.__cells = {}
-        self.__min_col_idx = 0
-        self.__max_col_idx = 0
-        self.__xf_index = 0x0F
-        self.__has_default_xf_index = 0
-        self.__height_in_pixels = 0x11
-
-        self.height = 0x00FF
-        self.has_default_height = 0x00
-        self.height_mismatch = 0
-        self.level = 0
-        self.collapse = 0
-        self.hidden = 0
-        self.space_above = 0
-        self.space_below = 0
-
-
-    def __adjust_height(self, style):
-        twips = style.font.height
-        points = float(twips)/20.0
-        # Cell height in pixels can be calcuted by following approx. formula:
-        # cell height in pixels = font height in points * 83/50 + 2/5
-        # It works when screen resolution is 96 dpi
-        pix = int(round(points*83.0/50.0 + 2.0/5.0))
-        if pix > self.__height_in_pixels:
-            self.__height_in_pixels = pix
-
-
-    def __adjust_bound_col_idx(self, *args):
-        for arg in args:
-            iarg = int(arg)
-            if not ((0 <= iarg <= 255) and arg == iarg):
-                raise ValueError("column index (%r) not an int in range(256)" % arg)
-            sheet = self.__parent
-            if iarg < self.__min_col_idx:
-                self.__min_col_idx = iarg
-            if iarg > self.__max_col_idx:
-                self.__max_col_idx = iarg
-            if iarg < sheet.first_used_col:
-                sheet.first_used_col = iarg
-            if iarg > sheet.last_used_col:
-                sheet.last_used_col = iarg
-
-    def __excel_date_dt(self, date):
-        if isinstance(date, dt.date) and (not isinstance(date, dt.datetime)):
-            epoch = dt.date(1899, 12, 31)
-        elif isinstance(date, dt.time):
-            date = dt.datetime.combine(dt.datetime(1900, 1, 1), date)
-            epoch = dt.datetime(1900, 1, 1, 0, 0, 0)
-        else:
-            epoch = dt.datetime(1899, 12, 31, 0, 0, 0)
-        delta = date - epoch
-        xldate = delta.days + float(delta.seconds) / (24*60*60)
-        # Add a day for Excel's missing leap day in 1900
-        if xldate > 59:
-            xldate += 1
-        return xldate
-
-    def get_height_in_pixels(self):
-        return self.__height_in_pixels
-
-
-    def set_style(self, style):
-        self.__adjust_height(style)
-        self.__xf_index = self.__parent_wb.add_style(style)
-        self.__has_default_xf_index = 1
-
-
-    def get_xf_index(self):
-        return self.__xf_index
-
-
-    def get_cells_count(self):
-        return len(self.__cells)
-
-
-    def get_min_col(self):
-        return self.__min_col_idx
-
-
-    def get_max_col(self):
-        return self.__max_col_idx
-
-
-    def get_row_biff_data(self):
-        height_options = (self.height & 0x07FFF)
-        height_options |= (self.has_default_height & 0x01) << 15
-
-        options =  (self.level & 0x07) << 0
-        options |= (self.collapse & 0x01) << 4
-        options |= (self.hidden & 0x01) << 5
-        options |= (self.height_mismatch & 0x01) << 6
-        options |= (self.__has_default_xf_index & 0x01) << 7
-        options |= (0x01 & 0x01) << 8
-        options |= (self.__xf_index & 0x0FFF) << 16
-        options |= (self.space_above & 1) << 28
-        options |= (self.space_below & 1) << 29
-
-        return BIFFRecords.RowRecord(self.__idx, self.__min_col_idx,
-            self.__max_col_idx, height_options, options).get()
-
-    def insert_cell(self, col_index, cell_obj):
-        if col_index in self.__cells:
-            if not self.__parent._cell_overwrite_ok:
-                msg = "Attempt to overwrite cell: sheetname=%r rowx=%d colx=%d" \
-                    % (self.__parent.name, self.__idx, col_index)
-                raise Exception(msg)
-            prev_cell_obj = self.__cells[col_index]
-            sst_idx = getattr(prev_cell_obj, 'sst_idx', None)
-            if sst_idx is not None:
-                self.__parent_wb.del_str(sst_idx)
-        self.__cells[col_index] = cell_obj
-
-    def insert_mulcells(self, colx1, colx2, cell_obj):
-        self.insert_cell(colx1, cell_obj)
-        for col_index in range(colx1+1, colx2+1):
-            self.insert_cell(col_index, None)
-
-    def get_cells_biff_data(self):
-        cell_items = [item for item in self.__cells.items() if item[1] is not None]
-        cell_items.sort() # in column order
-        return _get_cells_biff_data_mul(self.__idx, cell_items)
-        # previously:
-        # return ''.join([cell.get_biff_data() for colx, cell in cell_items])
-
-    def get_index(self):
-        return self.__idx
-
-    def set_cell_text(self, colx, value, style=Style.default_style):
-        self.__adjust_height(style)
-        self.__adjust_bound_col_idx(colx)
-        xf_index = self.__parent_wb.add_style(style)
-        self.insert_cell(colx, StrCell(self.__idx, colx, xf_index, self.__parent_wb.add_str(value)))
-
-    def set_cell_blank(self, colx, style=Style.default_style):
-        self.__adjust_height(style)
-        self.__adjust_bound_col_idx(colx)
-        xf_index = self.__parent_wb.add_style(style)
-        self.insert_cell(colx, BlankCell(self.__idx, colx, xf_index))
-
-    def set_cell_mulblanks(self, first_colx, last_colx, style=Style.default_style):
-        assert 0 <= first_colx <= last_colx <= 255
-        self.__adjust_height(style)
-        self.__adjust_bound_col_idx(first_colx, last_colx)
-        xf_index = self.__parent_wb.add_style(style)
-        # ncols = last_colx - first_colx + 1
-        self.insert_mulcells(first_colx, last_colx, MulBlankCell(self.__idx, first_colx, last_colx, xf_index))
-
-    def set_cell_number(self, colx, number, style=Style.default_style):
-        self.__adjust_height(style)
-        self.__adjust_bound_col_idx(colx)
-        xf_index = self.__parent_wb.add_style(style)
-        self.insert_cell(colx, NumberCell(self.__idx, colx, xf_index, number))
-
-    def set_cell_date(self, colx, datetime_obj, style=Style.default_style):
-        self.__adjust_height(style)
-        self.__adjust_bound_col_idx(colx)
-        xf_index = self.__parent_wb.add_style(style)
-        self.insert_cell(colx,
-            NumberCell(self.__idx, colx, xf_index, self.__excel_date_dt(datetime_obj)))
-
-    def set_cell_formula(self, colx, formula, style=Style.default_style, calc_flags=0):
-        self.__adjust_height(style)
-        self.__adjust_bound_col_idx(colx)
-        xf_index = self.__parent_wb.add_style(style)
-        self.__parent_wb.add_sheet_reference(formula)
-        self.insert_cell(colx, FormulaCell(self.__idx, colx, xf_index, formula, calc_flags=0))
-
-    def set_cell_boolean(self, colx, value, style=Style.default_style):
-        self.__adjust_height(style)
-        self.__adjust_bound_col_idx(colx)
-        xf_index = self.__parent_wb.add_style(style)
-        self.insert_cell(colx, BooleanCell(self.__idx, colx, xf_index, bool(value)))
-
-    def set_cell_error(self, colx, error_string_or_code, style=Style.default_style):
-        self.__adjust_height(style)
-        self.__adjust_bound_col_idx(colx)
-        xf_index = self.__parent_wb.add_style(style)
-        self.insert_cell(colx, ErrorCell(self.__idx, colx, xf_index, error_string_or_code))
-
-    def write(self, col, label, style=Style.default_style):
-        self.__adjust_height(style)
-        self.__adjust_bound_col_idx(col)
-        style_index = self.__parent_wb.add_style(style)
-        if isinstance(label, str):
-            if len(label) > 0:
-                self.insert_cell(col,
-                    StrCell(self.__idx, col, style_index, self.__parent_wb.add_str(label))
-                    )
-            else:
-                self.insert_cell(col, BlankCell(self.__idx, col, style_index))
-        elif isinstance(label, bool): # bool is subclass of int; test bool first
-            self.insert_cell(col, BooleanCell(self.__idx, col, style_index, label))
-        elif isinstance(label, (float, int, Decimal)):
-            self.insert_cell(col, NumberCell(self.__idx, col, style_index, label))
-        elif isinstance(label, (dt.datetime, dt.date, dt.time)):
-            date_number = self.__excel_date_dt(label)
-            self.insert_cell(col, NumberCell(self.__idx, col, style_index, date_number))
-        elif label is None:
-            self.insert_cell(col, BlankCell(self.__idx, col, style_index))
-        elif isinstance(label, ExcelFormula.Formula):
-            self.__parent_wb.add_sheet_reference(label)
-            self.insert_cell(col, FormulaCell(self.__idx, col, style_index, label))
-        else:
-            raise Exception("Unexpected data type %r" % type(label))
-
-    write_blanks = set_cell_mulblanks
-
-
-
+# -*- coding: windows-1252 -*-
+
+import BIFFRecords
+import Style
+from Cell import StrCell, BlankCell, NumberCell, FormulaCell, MulBlankCell, BooleanCell, ErrorCell, \
+    _get_cells_biff_data_mul
+import ExcelFormula
+import datetime as dt
+try:
+    from decimal import Decimal
+except ImportError:
+    # Python 2.3: decimal not supported; create dummy Decimal class
+    class Decimal(object):
+        pass
+
+
+class Row(object):
+    __slots__ = [# private variables
+                 "__idx",
+                 "__parent",
+                 "__parent_wb",
+                 "__cells",
+                 "__min_col_idx",
+                 "__max_col_idx",
+                 "__xf_index",
+                 "__has_default_xf_index",
+                 "__height_in_pixels",
+                 # public variables
+                 "height",
+                 "has_default_height",
+                 "height_mismatch",
+                 "level",
+                 "collapse",
+                 "hidden",
+                 "space_above",
+                 "space_below"]
+
+    def __init__(self, rowx, parent_sheet):
+        if not (isinstance(rowx, int) and 0 <= rowx <= 65535):
+            raise ValueError("row index (%r) not an int in range(65536)" % rowx)
+        self.__idx = rowx
+        self.__parent = parent_sheet
+        self.__parent_wb = parent_sheet.get_parent()
+        self.__cells = {}
+        self.__min_col_idx = 0
+        self.__max_col_idx = 0
+        self.__xf_index = 0x0F
+        self.__has_default_xf_index = 0
+        self.__height_in_pixels = 0x11
+
+        self.height = 0x00FF
+        self.has_default_height = 0x00
+        self.height_mismatch = 0
+        self.level = 0
+        self.collapse = 0
+        self.hidden = 0
+        self.space_above = 0
+        self.space_below = 0
+
+
+    def __adjust_height(self, style):
+        twips = style.font.height
+        points = float(twips)/20.0
+        # Cell height in pixels can be calcuted by following approx. formula:
+        # cell height in pixels = font height in points * 83/50 + 2/5
+        # It works when screen resolution is 96 dpi
+        pix = int(round(points*83.0/50.0 + 2.0/5.0))
+        if pix > self.__height_in_pixels:
+            self.__height_in_pixels = pix
+
+
+    def __adjust_bound_col_idx(self, *args):
+        for arg in args:
+            iarg = int(arg)
+            if not ((0 <= iarg <= 255) and arg == iarg):
+                raise ValueError("column index (%r) not an int in range(256)" % arg)
+            sheet = self.__parent
+            if iarg < self.__min_col_idx:
+                self.__min_col_idx = iarg
+            if iarg > self.__max_col_idx:
+                self.__max_col_idx = iarg
+            if iarg < sheet.first_used_col:
+                sheet.first_used_col = iarg
+            if iarg > sheet.last_used_col:
+                sheet.last_used_col = iarg
+
+    def __excel_date_dt(self, date):
+        if isinstance(date, dt.date) and (not isinstance(date, dt.datetime)):
+            epoch = dt.date(1899, 12, 31)
+        elif isinstance(date, dt.time):
+            date = dt.datetime.combine(dt.datetime(1900, 1, 1), date)
+            epoch = dt.datetime(1900, 1, 1, 0, 0, 0)
+        else:
+            epoch = dt.datetime(1899, 12, 31, 0, 0, 0)
+        delta = date - epoch
+        xldate = delta.days + float(delta.seconds) / (24*60*60)
+        # Add a day for Excel's missing leap day in 1900
+        if xldate > 59:
+            xldate += 1
+        return xldate
+
+    def get_height_in_pixels(self):
+        return self.__height_in_pixels
+
+
+    def set_style(self, style):
+        self.__adjust_height(style)
+        self.__xf_index = self.__parent_wb.add_style(style)
+        self.__has_default_xf_index = 1
+
+
+    def get_xf_index(self):
+        return self.__xf_index
+
+
+    def get_cells_count(self):
+        return len(self.__cells)
+
+
+    def get_min_col(self):
+        return self.__min_col_idx
+
+
+    def get_max_col(self):
+        return self.__max_col_idx
+
+
+    def get_row_biff_data(self):
+        height_options = (self.height & 0x07FFF)
+        height_options |= (self.has_default_height & 0x01) << 15
+
+        options =  (self.level & 0x07) << 0
+        options |= (self.collapse & 0x01) << 4
+        options |= (self.hidden & 0x01) << 5
+        options |= (self.height_mismatch & 0x01) << 6
+        options |= (self.__has_default_xf_index & 0x01) << 7
+        options |= (0x01 & 0x01) << 8
+        options |= (self.__xf_index & 0x0FFF) << 16
+        options |= (self.space_above & 1) << 28
+        options |= (self.space_below & 1) << 29
+
+        return BIFFRecords.RowRecord(self.__idx, self.__min_col_idx,
+            self.__max_col_idx, height_options, options).get()
+
+    def insert_cell(self, col_index, cell_obj):
+        if col_index in self.__cells:
+            if not self.__parent._cell_overwrite_ok:
+                msg = "Attempt to overwrite cell: sheetname=%r rowx=%d colx=%d" \
+                    % (self.__parent.name, self.__idx, col_index)
+                raise Exception(msg)
+            prev_cell_obj = self.__cells[col_index]
+            sst_idx = getattr(prev_cell_obj, 'sst_idx', None)
+            if sst_idx is not None:
+                self.__parent_wb.del_str(sst_idx)
+        self.__cells[col_index] = cell_obj
+
+    def insert_mulcells(self, colx1, colx2, cell_obj):
+        self.insert_cell(colx1, cell_obj)
+        for col_index in xrange(colx1+1, colx2+1):
+            self.insert_cell(col_index, None)
+
+    def get_cells_biff_data(self):
+        cell_items = [item for item in self.__cells.iteritems() if item[1] is not None]
+        cell_items.sort() # in column order
+        return _get_cells_biff_data_mul(self.__idx, cell_items)
+        # previously:
+        # return ''.join([cell.get_biff_data() for colx, cell in cell_items])
+
+    def get_index(self):
+        return self.__idx
+
+    def set_cell_text(self, colx, value, style=Style.default_style):
+        self.__adjust_height(style)
+        self.__adjust_bound_col_idx(colx)
+        xf_index = self.__parent_wb.add_style(style)
+        self.insert_cell(colx, StrCell(self.__idx, colx, xf_index, self.__parent_wb.add_str(value)))
+
+    def set_cell_blank(self, colx, style=Style.default_style):
+        self.__adjust_height(style)
+        self.__adjust_bound_col_idx(colx)
+        xf_index = self.__parent_wb.add_style(style)
+        self.insert_cell(colx, BlankCell(self.__idx, colx, xf_index))
+
+    def set_cell_mulblanks(self, first_colx, last_colx, style=Style.default_style):
+        assert 0 <= first_colx <= last_colx <= 255
+        self.__adjust_height(style)
+        self.__adjust_bound_col_idx(first_colx, last_colx)
+        xf_index = self.__parent_wb.add_style(style)
+        # ncols = last_colx - first_colx + 1
+        self.insert_mulcells(first_colx, last_colx, MulBlankCell(self.__idx, first_colx, last_colx, xf_index))
+
+    def set_cell_number(self, colx, number, style=Style.default_style):
+        self.__adjust_height(style)
+        self.__adjust_bound_col_idx(colx)
+        xf_index = self.__parent_wb.add_style(style)
+        self.insert_cell(colx, NumberCell(self.__idx, colx, xf_index, number))
+
+    def set_cell_date(self, colx, datetime_obj, style=Style.default_style):
+        self.__adjust_height(style)
+        self.__adjust_bound_col_idx(colx)
+        xf_index = self.__parent_wb.add_style(style)
+        self.insert_cell(colx,
+            NumberCell(self.__idx, colx, xf_index, self.__excel_date_dt(datetime_obj)))
+
+    def set_cell_formula(self, colx, formula, style=Style.default_style, calc_flags=0):
+        self.__adjust_height(style)
+        self.__adjust_bound_col_idx(colx)
+        xf_index = self.__parent_wb.add_style(style)
+        self.__parent_wb.add_sheet_reference(formula)
+        self.insert_cell(colx, FormulaCell(self.__idx, colx, xf_index, formula, calc_flags=0))
+
+    def set_cell_boolean(self, colx, value, style=Style.default_style):
+        self.__adjust_height(style)
+        self.__adjust_bound_col_idx(colx)
+        xf_index = self.__parent_wb.add_style(style)
+        self.insert_cell(colx, BooleanCell(self.__idx, colx, xf_index, bool(value)))
+
+    def set_cell_error(self, colx, error_string_or_code, style=Style.default_style):
+        self.__adjust_height(style)
+        self.__adjust_bound_col_idx(colx)
+        xf_index = self.__parent_wb.add_style(style)
+        self.insert_cell(colx, ErrorCell(self.__idx, colx, xf_index, error_string_or_code))
+
+    def write(self, col, label, style=Style.default_style):
+        self.__adjust_height(style)
+        self.__adjust_bound_col_idx(col)
+        style_index = self.__parent_wb.add_style(style)
+        if isinstance(label, basestring):
+            if len(label) > 0:
+                self.insert_cell(col,
+                    StrCell(self.__idx, col, style_index, self.__parent_wb.add_str(label))
+                    )
+            else:
+                self.insert_cell(col, BlankCell(self.__idx, col, style_index))
+        elif isinstance(label, bool): # bool is subclass of int; test bool first
+            self.insert_cell(col, BooleanCell(self.__idx, col, style_index, label))
+        elif isinstance(label, (float, int, long, Decimal)):
+            self.insert_cell(col, NumberCell(self.__idx, col, style_index, label))
+        elif isinstance(label, (dt.datetime, dt.date, dt.time)):
+            date_number = self.__excel_date_dt(label)
+            self.insert_cell(col, NumberCell(self.__idx, col, style_index, date_number))
+        elif label is None:
+            self.insert_cell(col, BlankCell(self.__idx, col, style_index))
+        elif isinstance(label, ExcelFormula.Formula):
+            self.__parent_wb.add_sheet_reference(label)
+            self.insert_cell(col, FormulaCell(self.__idx, col, style_index, label))
+        else:
+            raise Exception("Unexpected data type %r" % type(label))
+
+    write_blanks = set_cell_mulblanks
+
+
+

tablib/packages/xlwt/UnicodeUtils.py

@@ -1,81 +1,81 @@
-# -*- coding: utf-8 -*-
-
-'''
-From BIFF8 on, strings are always stored using UTF-16LE  text encoding. The
-character  array  is  a  sequence  of  16-bit  values4.  Additionally it is
-possible  to  use  a  compressed  format, which omits the high bytes of all
-characters, if they are all zero.
-
-The following tables describe the standard format of the entire string, but
-in many records the strings differ from this format. This will be mentioned
-separately. It is possible (but not required) to store Rich-Text formatting
-information  and  Asian  phonetic information inside a Unicode string. This
-results  in  four  different  ways  to  store a string. The character array
-is not zero-terminated.
-
-The  string  consists  of  the  character count (as usual an 8-bit value or
-a  16-bit value), option flags, the character array and optional formatting
-information.  If the string is empty, sometimes the option flags field will
-not occur. This is mentioned at the respective place.
-
-Offset  Size    Contents
-0       1 or 2  Length of the string (character count, ln)
-1 or 2  1       Option flags:
-                  Bit   Mask Contents
-                  0     01H  Character compression (ccompr):
-                               0 = Compressed (8-bit characters)
-                               1 = Uncompressed (16-bit characters)
-                  2     04H  Asian phonetic settings (phonetic):
-                               0 = Does not contain Asian phonetic settings
-                               1 = Contains Asian phonetic settings
-                  3     08H  Rich-Text settings (richtext):
-                               0 = Does not contain Rich-Text settings
-                               1 = Contains Rich-Text settings
-[2 or 3] 2      (optional, only if richtext=1) Number of Rich-Text formatting runs (rt)
-[var.]   4      (optional, only if phonetic=1) Size of Asian phonetic settings block (in bytes, sz)
-var.     ln or
-         2·ln   Character array (8-bit characters or 16-bit characters, dependent on ccompr)
-[var.]   4·rt   (optional, only if richtext=1) List of rt formatting runs
-[var.]   sz     (optional, only if phonetic=1) Asian Phonetic Settings Block
-'''
-
-
-from struct import pack
-
-def upack2(s, encoding='ascii'):
-    # If not unicode, make it so.
-    if isinstance(s, str):
-        us = s
-    else:
-        us = str(s, encoding)
-    # Limit is based on number of content characters
-    # (not on number of bytes in packed result)
-    len_us = len(us)
-    if len_us > 65535:
-        raise Exception('String longer than 65535 characters')
-    try:
-        encs = us.encode('latin1')
-        # Success here means all chars are in U+0000 to U+00FF
-        # inclusive, meaning that we can use "compressed format".
-        flag = 0
-    except UnicodeEncodeError:
-        encs = us.encode('utf_16_le')
-        flag = 1
-    return pack('<HB', len_us, flag) + encs
-
-def upack1(s, encoding='ascii'):
-    # Same as upack2(), but with a one-byte length field.
-    if isinstance(s, str):
-        us = s
-    else:
-        us = str(s, encoding)
-    len_us = len(us)
-    if len_us > 255:
-        raise Exception('String longer than 255 characters')
-    try:
-        encs = us.encode('latin1')
-        flag = 0
-    except UnicodeEncodeError:
-        encs = us.encode('utf_16_le')
-        flag = 1
-    return pack('<BB', len_us, flag) + encs
+# -*- coding: windows-1252 -*-
+
+'''
+From BIFF8 on, strings are always stored using UTF-16LE  text encoding. The
+character  array  is  a  sequence  of  16-bit  values4.  Additionally it is
+possible  to  use  a  compressed  format, which omits the high bytes of all
+characters, if they are all zero.
+
+The following tables describe the standard format of the entire string, but
+in many records the strings differ from this format. This will be mentioned
+separately. It is possible (but not required) to store Rich-Text formatting
+information  and  Asian  phonetic information inside a Unicode string. This
+results  in  four  different  ways  to  store a string. The character array
+is not zero-terminated.
+
+The  string  consists  of  the  character count (as usual an 8-bit value or
+a  16-bit value), option flags, the character array and optional formatting
+information.  If the string is empty, sometimes the option flags field will
+not occur. This is mentioned at the respective place.
+
+Offset  Size    Contents
+0       1 or 2  Length of the string (character count, ln)
+1 or 2  1       Option flags:
+                  Bit   Mask Contents
+                  0     01H  Character compression (ccompr):
+                               0 = Compressed (8-bit characters)
+                               1 = Uncompressed (16-bit characters)
+                  2     04H  Asian phonetic settings (phonetic):
+                               0 = Does not contain Asian phonetic settings
+                               1 = Contains Asian phonetic settings
+                  3     08H  Rich-Text settings (richtext):
+                               0 = Does not contain Rich-Text settings
+                               1 = Contains Rich-Text settings
+[2 or 3] 2      (optional, only if richtext=1) Number of Rich-Text formatting runs (rt)
+[var.]   4      (optional, only if phonetic=1) Size of Asian phonetic settings block (in bytes, sz)
+var.     ln or 
+         2·ln   Character array (8-bit characters or 16-bit characters, dependent on ccompr)
+[var.]   4·rt   (optional, only if richtext=1) List of rt formatting runs 
+[var.]   sz     (optional, only if phonetic=1) Asian Phonetic Settings Block 
+'''
+
+
+from struct import pack
+
+def upack2(s, encoding='ascii'):
+    # If not unicode, make it so.
+    if isinstance(s, unicode):
+        us = s
+    else:
+        us = unicode(s, encoding)
+    # Limit is based on number of content characters
+    # (not on number of bytes in packed result)
+    len_us = len(us)
+    if len_us > 65535:
+        raise Exception('String longer than 65535 characters')
+    try:
+        encs = us.encode('latin1')
+        # Success here means all chars are in U+0000 to U+00FF
+        # inclusive, meaning that we can use "compressed format".
+        flag = 0
+    except UnicodeEncodeError:
+        encs = us.encode('utf_16_le')
+        flag = 1
+    return pack('<HB', len_us, flag) + encs
+
+def upack1(s, encoding='ascii'):
+    # Same as upack2(), but with a one-byte length field.
+    if isinstance(s, unicode):
+        us = s
+    else:
+        us = unicode(s, encoding)
+    len_us = len(us)
+    if len_us > 255:
+        raise Exception('String longer than 255 characters')
+    try:
+        encs = us.encode('latin1')
+        flag = 0
+    except UnicodeEncodeError:
+        encs = us.encode('utf_16_le')
+        flag = 1
+    return pack('<BB', len_us, flag) + encs

tablib/packages/xlwt/Utils.py

@@ -1,196 +1,196 @@
-# pyXLWriter: A library for generating Excel Spreadsheets
-# Copyright (c) 2004 Evgeny Filatov <fufff@users.sourceforge.net>
-# Copyright (c) 2002-2004 John McNamara (Perl Spreadsheet::WriteExcel)
-#
-# This library is free software; you can redistribute it and/or modify it
-# under the terms of the GNU Lesser General Public License as published by
-# the Free Software Foundation; either version 2.1 of the License, or
-# (at your option) any later version.
-#
-# This library is distributed in the hope that it will be useful, but
-# WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser
-# General Public License for more details.
-#
-# You should have received a copy of the GNU Lesser General Public License
-# along with this library; if not, write to the Free Software Foundation,
-# Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
-#----------------------------------------------------------------------------
-# This module was written/ported from PERL Spreadsheet::WriteExcel module
-# The author of the PERL Spreadsheet::WriteExcel module is John McNamara
-# <jmcnamara@cpan.org>
-#----------------------------------------------------------------------------
-# See the README.txt distributed with pyXLWriter for more details.
-
-# Portions are (C) Roman V. Kiseliov, 2005
-
-
-# Utilities for work with reference to cells and with sheetnames
-
-
-__rev_id__ = """$Id: Utils.py 3844 2009-05-20 01:02:54Z sjmachin $"""
-
-import re
-from struct import pack
-from .ExcelMagic import MAX_ROW, MAX_COL
-
-
-_re_cell_ex = re.compile(r"(\$?)([A-I]?[A-Z])(\$?)(\d+)", re.IGNORECASE)
-_re_row_range = re.compile(r"\$?(\d+):\$?(\d+)")
-_re_col_range = re.compile(r"\$?([A-I]?[A-Z]):\$?([A-I]?[A-Z])", re.IGNORECASE)
-_re_cell_range = re.compile(r"\$?([A-I]?[A-Z]\$?\d+):\$?([A-I]?[A-Z]\$?\d+)", re.IGNORECASE)
-_re_cell_ref = re.compile(r"\$?([A-I]?[A-Z]\$?\d+)", re.IGNORECASE)
-
-
-def col_by_name(colname):
-    """
-    """
-    col = 0
-    pow = 1
-    for i in range(len(colname)-1, -1, -1):
-        ch = colname[i]
-        col += (ord(ch) - ord('A') + 1) * pow
-        pow *= 26
-    return col - 1
-
-
-def cell_to_rowcol(cell):
-    """Convert an Excel cell reference string in A1 notation
-    to numeric row/col notation.
-
-    Returns: row, col, row_abs, col_abs
-
-    """
-    m = _re_cell_ex.match(cell)
-    if not m:
-        raise Exception("Ill-formed single_cell reference: %s" % cell)
-    col_abs, col, row_abs, row = m.groups()
-    row_abs = bool(row_abs)
-    col_abs = bool(col_abs)
-    row = int(row) - 1
-    col = col_by_name(col.upper())
-    return row, col, row_abs, col_abs
-
-
-def cell_to_rowcol2(cell):
-    """Convert an Excel cell reference string in A1 notation
-    to numeric row/col notation.
-
-    Returns: row, col
-
-    """
-    m = _re_cell_ex.match(cell)
-    if not m:
-        raise Exception("Error in cell format")
-    col_abs, col, row_abs, row = m.groups()
-    # Convert base26 column string to number
-    # All your Base are belong to us.
-    row = int(row) - 1
-    col = col_by_name(col.upper())
-    return row, col
-
-
-def rowcol_to_cell(row, col, row_abs=False, col_abs=False):
-    """Convert numeric row/col notation to an Excel cell reference string in
-    A1 notation.
-
-    """
-    assert 0 <= row < MAX_ROW # MAX_ROW counts from 1
-    assert 0 <= col < MAX_COL # MAX_COL counts from 1
-    d = col // 26
-    m = col % 26
-    chr1 = ""    # Most significant character in AA1
-    if row_abs:
-        row_abs = '$'
-    else:
-        row_abs = ''
-    if col_abs:
-        col_abs = '$'
-    else:
-        col_abs = ''
-    if d > 0:
-        chr1 = chr(ord('A') + d  - 1)
-    chr2 = chr(ord('A') + m)
-    # Zero index to 1-index
-    return col_abs + chr1 + chr2 + row_abs + str(row + 1)
-
-def rowcol_pair_to_cellrange(row1, col1, row2, col2,
-    row1_abs=False, col1_abs=False, row2_abs=False, col2_abs=False):
-    """Convert two (row,column) pairs
-    into a cell range string in A1:B2 notation.
-
-    Returns: cell range string
-    """
-    assert row1 <= row2
-    assert col1 <= col2
-    return (
-        rowcol_to_cell(row1, col1, row1_abs, col1_abs)
-        + ":"
-        + rowcol_to_cell(row2, col2, row2_abs, col2_abs)
-        )
-
-def cellrange_to_rowcol_pair(cellrange):
-    """Convert cell range string in A1 notation to numeric row/col
-    pair.
-
-    Returns: row1, col1, row2, col2
-
-    """
-    cellrange = cellrange.upper()
-    # Convert a row range: '1:3'
-    res = _re_row_range.match(cellrange)
-    if res:
-        row1 = int(res.group(1)) - 1
-        col1 = 0
-        row2 = int(res.group(2)) - 1
-        col2 = -1
-        return row1, col1, row2, col2
-    # Convert a column range: 'A:A' or 'B:G'.
-    # A range such as A:A is equivalent to A1:A16384, so add rows as required
-    res = _re_col_range.match(cellrange)
-    if res:
-        col1 = col_by_name(res.group(1).upper())
-        row1 = 0
-        col2 = col_by_name(res.group(2).upper())
-        row2 = -1
-        return row1, col1, row2, col2
-    # Convert a cell range: 'A1:B7'
-    res = _re_cell_range.match(cellrange)
-    if res:
-        row1, col1 = cell_to_rowcol2(res.group(1))
-        row2, col2 = cell_to_rowcol2(res.group(2))
-        return row1, col1, row2, col2
-    # Convert a cell reference: 'A1' or 'AD2000'
-    res = _re_cell_ref.match(cellrange)
-    if res:
-        row1, col1 = cell_to_rowcol2(res.group(1))
-        return row1, col1, row1, col1
-    raise Exception("Unknown cell reference %s" % (cell))
-
-
-def cell_to_packed_rowcol(cell):
-    """ pack row and column into the required 4 byte format """
-    row, col, row_abs, col_abs = cell_to_rowcol(cell)
-    if col >= MAX_COL:
-        raise Exception("Column %s greater than IV in formula" % cell)
-    if row >= MAX_ROW: # this for BIFF8. for BIFF7 available 2^14
-        raise Exception("Row %s greater than %d in formula" % (cell, MAX_ROW))
-    col |= int(not row_abs) << 15
-    col |= int(not col_abs) << 14
-    return row, col
-
-# === sheetname functions ===
-
-def valid_sheet_name(sheet_name):
-    if sheet_name == "" or sheet_name[0] == "'" or len(sheet_name) > 31:
-        return False
-    for c in sheet_name:
-        if c in "[]:\\?/*\x00":
-            return False
-    return True
-
-def quote_sheet_name(unquoted_sheet_name):
-    if not valid_sheet_name(unquoted_sheet_name):
-        raise Exception(
-            'attempt to quote an invalid worksheet name %r' % unquoted_sheet_name)
-    return "'" + unquoted_sheet_name.replace("'", "''") + "'"
+# pyXLWriter: A library for generating Excel Spreadsheets
+# Copyright (c) 2004 Evgeny Filatov <fufff@users.sourceforge.net>
+# Copyright (c) 2002-2004 John McNamara (Perl Spreadsheet::WriteExcel)
+#
+# This library is free software; you can redistribute it and/or modify it
+# under the terms of the GNU Lesser General Public License as published by
+# the Free Software Foundation; either version 2.1 of the License, or
+# (at your option) any later version.
+#
+# This library is distributed in the hope that it will be useful, but
+# WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser
+# General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public License
+# along with this library; if not, write to the Free Software Foundation,
+# Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+#----------------------------------------------------------------------------
+# This module was written/ported from PERL Spreadsheet::WriteExcel module
+# The author of the PERL Spreadsheet::WriteExcel module is John McNamara
+# <jmcnamara@cpan.org>
+#----------------------------------------------------------------------------
+# See the README.txt distributed with pyXLWriter for more details.
+
+# Portions are (C) Roman V. Kiseliov, 2005
+
+
+# Utilities for work with reference to cells and with sheetnames
+
+
+__rev_id__ = """$Id: Utils.py 3844 2009-05-20 01:02:54Z sjmachin $"""
+
+import re
+from struct import pack
+from ExcelMagic import MAX_ROW, MAX_COL
+
+
+_re_cell_ex = re.compile(r"(\$?)([A-I]?[A-Z])(\$?)(\d+)", re.IGNORECASE)
+_re_row_range = re.compile(r"\$?(\d+):\$?(\d+)")
+_re_col_range = re.compile(r"\$?([A-I]?[A-Z]):\$?([A-I]?[A-Z])", re.IGNORECASE)
+_re_cell_range = re.compile(r"\$?([A-I]?[A-Z]\$?\d+):\$?([A-I]?[A-Z]\$?\d+)", re.IGNORECASE)
+_re_cell_ref = re.compile(r"\$?([A-I]?[A-Z]\$?\d+)", re.IGNORECASE)
+
+
+def col_by_name(colname):
+    """
+    """
+    col = 0
+    pow = 1
+    for i in xrange(len(colname)-1, -1, -1):
+        ch = colname[i]
+        col += (ord(ch) - ord('A') + 1) * pow
+        pow *= 26
+    return col - 1
+
+
+def cell_to_rowcol(cell):
+    """Convert an Excel cell reference string in A1 notation
+    to numeric row/col notation.
+
+    Returns: row, col, row_abs, col_abs
+
+    """
+    m = _re_cell_ex.match(cell)
+    if not m:
+        raise Exception("Ill-formed single_cell reference: %s" % cell)
+    col_abs, col, row_abs, row = m.groups()
+    row_abs = bool(row_abs)
+    col_abs = bool(col_abs)
+    row = int(row) - 1
+    col = col_by_name(col.upper())
+    return row, col, row_abs, col_abs
+
+
+def cell_to_rowcol2(cell):
+    """Convert an Excel cell reference string in A1 notation
+    to numeric row/col notation.
+
+    Returns: row, col
+
+    """
+    m = _re_cell_ex.match(cell)
+    if not m:
+        raise Exception("Error in cell format")
+    col_abs, col, row_abs, row = m.groups()
+    # Convert base26 column string to number
+    # All your Base are belong to us.
+    row = int(row) - 1
+    col = col_by_name(col.upper())
+    return row, col
+
+
+def rowcol_to_cell(row, col, row_abs=False, col_abs=False):
+    """Convert numeric row/col notation to an Excel cell reference string in
+    A1 notation.
+
+    """
+    assert 0 <= row < MAX_ROW # MAX_ROW counts from 1
+    assert 0 <= col < MAX_COL # MAX_COL counts from 1
+    d = col // 26
+    m = col % 26
+    chr1 = ""    # Most significant character in AA1
+    if row_abs:
+        row_abs = '$'
+    else:
+        row_abs = ''
+    if col_abs:
+        col_abs = '$'
+    else:
+        col_abs = ''
+    if d > 0:
+        chr1 = chr(ord('A') + d  - 1)
+    chr2 = chr(ord('A') + m)
+    # Zero index to 1-index
+    return col_abs + chr1 + chr2 + row_abs + str(row + 1)
+
+def rowcol_pair_to_cellrange(row1, col1, row2, col2,
+    row1_abs=False, col1_abs=False, row2_abs=False, col2_abs=False):
+    """Convert two (row,column) pairs
+    into a cell range string in A1:B2 notation.
+
+    Returns: cell range string
+    """
+    assert row1 <= row2
+    assert col1 <= col2
+    return (
+        rowcol_to_cell(row1, col1, row1_abs, col1_abs)
+        + ":"
+        + rowcol_to_cell(row2, col2, row2_abs, col2_abs)
+        )
+
+def cellrange_to_rowcol_pair(cellrange):
+    """Convert cell range string in A1 notation to numeric row/col
+    pair.
+
+    Returns: row1, col1, row2, col2
+
+    """
+    cellrange = cellrange.upper()
+    # Convert a row range: '1:3'
+    res = _re_row_range.match(cellrange)
+    if res:
+        row1 = int(res.group(1)) - 1
+        col1 = 0
+        row2 = int(res.group(2)) - 1
+        col2 = -1
+        return row1, col1, row2, col2
+    # Convert a column range: 'A:A' or 'B:G'.
+    # A range such as A:A is equivalent to A1:A16384, so add rows as required
+    res = _re_col_range.match(cellrange)
+    if res:
+        col1 = col_by_name(res.group(1).upper())
+        row1 = 0
+        col2 = col_by_name(res.group(2).upper())
+        row2 = -1
+        return row1, col1, row2, col2
+    # Convert a cell range: 'A1:B7'
+    res = _re_cell_range.match(cellrange)
+    if res:
+        row1, col1 = cell_to_rowcol2(res.group(1))
+        row2, col2 = cell_to_rowcol2(res.group(2))
+        return row1, col1, row2, col2
+    # Convert a cell reference: 'A1' or 'AD2000'
+    res = _re_cell_ref.match(cellrange)
+    if res:
+        row1, col1 = cell_to_rowcol2(res.group(1))
+        return row1, col1, row1, col1
+    raise Exception("Unknown cell reference %s" % (cell))
+
+
+def cell_to_packed_rowcol(cell):
+    """ pack row and column into the required 4 byte format """
+    row, col, row_abs, col_abs = cell_to_rowcol(cell)
+    if col >= MAX_COL:
+        raise Exception("Column %s greater than IV in formula" % cell)
+    if row >= MAX_ROW: # this for BIFF8. for BIFF7 available 2^14
+        raise Exception("Row %s greater than %d in formula" % (cell, MAX_ROW))
+    col |= int(not row_abs) << 15
+    col |= int(not col_abs) << 14
+    return row, col
+
+# === sheetname functions ===
+
+def valid_sheet_name(sheet_name):
+    if sheet_name == u"" or sheet_name[0] == u"'" or len(sheet_name) > 31:
+        return False
+    for c in sheet_name:
+        if c in u"[]:\\?/*\x00":
+            return False
+    return True
+
+def quote_sheet_name(unquoted_sheet_name):
+    if not valid_sheet_name(unquoted_sheet_name):
+        raise Exception(
+            'attempt to quote an invalid worksheet name %r' % unquoted_sheet_name)
+    return u"'" + unquoted_sheet_name.replace(u"'", u"''") + u"'"

tablib/packages/xlwt/__init__.py

@@ -1,9 +1,16 @@
-import sys
-
-from .Workbook import Workbook
-from .Worksheet import Worksheet
-from .Row import Row
-from .Column import Column
-from .Formatting import Font, Alignment, Borders, Pattern, Protection
-from .Style import XFStyle, easyxf
-from .ExcelFormula import *
+# -*- coding: windows-1252 -*-
+
+__VERSION__ = '0.7.2'
+
+import sys
+if sys.version_info[:2] < (2, 3):
+    print >> sys.stderr, "Sorry, xlwt requires Python 2.3 or later"
+    sys.exit(1)
+
+from Workbook import Workbook
+from Worksheet import Worksheet
+from Row import Row
+from Column import Column
+from Formatting import Font, Alignment, Borders, Pattern, Protection
+from Style import XFStyle, easyxf
+from ExcelFormula import *

tablib/packages/xlwt/doc/xlwt.html

@@ -0,0 +1,199 @@
+<!DOCTYPE html PUBLIC '-//W3C//DTD XHTML 1.0 Strict//EN' 'http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd'>
+<html>
+<head>
+<meta http-equiv='Content-Type' content='text/html; charset=us-ascii' />
+<title>The xlwt Module</title>
+</head>
+<body>
+<h1>The xlwt Module</h1>
+<p /><p><b>A Python package for generating Microsoft Excel &#8482; spreadsheet files.
+</b></p>
+
+<h2>General information</h2>
+
+<h3>State of Documentation</h3>
+
+<p>
+This documentation is currently incomplete. There may be methods and
+classes not included and any item marked with a <em
+style="color:red;">[NC]</em> is not complete and may have further
+parameters, methods, attributes and functionality that are not
+documented. In these cases, you'll have to refer to the source if the
+documentation provided is insufficient.
+</p>
+
+</p>
+
+<h2>Module Contents <em style="color:red;">[NC]</em></h2>
+<dl>
+  <dt><b>easyxf</b> (function)</dt>
+  <dd>
+  <p>
+  This function is used to create and configure XFStyle objects
+  for use with (for example) the Worksheet.write method.
+  </p>
+  <dl>
+    <dt><i>strg_to_parse</i></dt>
+    <dd>
+      <p>
+      	A string to be parsed to obtain attribute values for Alignment, Borders, Font,
+      	Pattern and Protection objects. Refer to the examples
+      	in the file .../examples/xlwt_easyxf_simple_demo.py and to the xf_dict
+      	dictionary in Style.py. Various synonyms including color/colour, center/centre and gray/grey
+      	are allowed. Case is irrelevant (except maybe in font names). '-' may be used instead
+      	of '_'.<br />
+		Example: "font: bold on; align: wrap on, vert centre, horiz center"
+      </p>
+    </dd>
+    <dt><i>num_format_str</i></dt>
+    <dd>
+      <p>
+		To get the "number format string" of an existing cell whose format you want to reproduce,
+		select the cell and click on Format/Cells/Number/Custom. Otherwise, refer to Excel help.<br />
+		Examples: "#,##0.00", "dd/mm/yyyy"
+      </p>
+    </dd>
+    <dt>Returns:</dt>
+    <dd>
+      An object of the XFstyle class
+    </dd>
+  </dl>
+  <br />
+
+  </dd>
+
+  <dt><b>Workbook</b> (class)  [<a href='#xlwt.Workbook-class'>#</a>]</dt>
+  <dd>
+    <p>The class to instantiate to create a workbook</p>
+    <p>For more information about this class, see <a href='#xlwt.Workbook-class'><i>The Workbook Class</i></a>.</p>
+  </dd>
+  <dt><b>Worksheet</b> (class)  [<a href='#xlwt.Worksheet-class'>#</a>]</dt>
+  <dd>
+    <p>A class to represent the contents of a sheet in a workbook.</p>
+    <p>For more information about this class, see <a href='#xlwt.Worksheet-class'><i>The Worksheet Class</i></a>.</p>
+  </dd>
+</dl>
+
+<h2><a id='xlwt.Workbook-class' name='xlwt.Workbook-class'>The Workbook Class</a><em style="color:red;">[NC]</em></h2>
+<dl>
+<dt><b>Workbook(encoding='ascii',style_compression=0)</b> (class) [<a href='#xlwt.Workbook-class'>#</a>]</dt>
+<dd>
+  <p>
+  This is a class representing a workbook and all its contents.
+  When creating Excel files with xlwt, you will normally start by
+  instantiating an object of this class.
+  </p>
+  <dl>
+    <dt><i>encoding</i></dt>
+    <dd>
+      <em style="color:red;">[NC]</em>
+    </dd>
+    <dt><i>style_compression</i></dt>
+    <dd>
+      <em style="color:red;">[NC]</em>
+    </dd>
+    <dt>Returns:</dt>
+    <dd>
+      An object of the <a href="#xlwt.Workbook-class">Workbook</a> class
+    </dd>
+  </dl>
+  <br />
+</dd>
+<dt><a id='xlwt.Workbook.add_sheet-method' name='xlwt.Workbook.add_sheet-method'><b>add_sheet(sheetname)</b></a> [<a href='#xlwt.Workbook.add_sheet-method'>#</a>]</dt>
+<dd>
+  <p>
+  This method is used to create Worksheets in a Workbook.
+  </p>
+  <dl>
+    <dt><i>sheetname</i></dt>
+    <dd>
+      The name to use for this sheet, as it will appear in the tabs at
+      the bottom of the Excel application.
+    </dd>
+    <dt>Returns:</dt>
+    <dd>
+      An object of the <a href="#xlwt.Worksheet-class">Worksheet</a> class
+    </dd>
+  </dl>
+  <br />
+</dd>
+<dt><a id='xlwt.Workbook.save-method' name='xlwt.Workbook.save-method'><b>save(filename_or_stream)</b></a> [<a href='#xlwt.Workbook.save-method'>#</a>]</dt>
+<dd>
+  <p>
+  This method is used to save Workbook to a file in native Excel format.
+  </p>
+  <dl>
+    <dt><i>filename_or_stream</i></dt>
+    <dd>
+      <p>
+      This can be a string containing a filename of the file, in which case
+      the excel file is saved to disk using the name provided.
+      </p>
+      <p>
+      It can also be a stream object with a write method, such as a
+      StringIO, in which case the data for the excel file is written
+      to the stream.
+      </p>
+    </dd>
+  </dl>
+  <br />
+</dd>
+</dl>
+
+<h2><a id='xlwt.Worksheet-class' name='xlwt.Worksheet-class'>The Worksheet Class</a><em style="color:red;">[NC]</em></h2>
+<dl>
+<dt><b>Worksheet(sheetname, parent_book)</b> (class) [<a href='#xlwt.Worksheet-class'>#</a>]</dt>
+<dd>
+  <p>
+  This is a class representing the contents of a sheet in a workbook.
+  </p>
+  <p>
+   WARNING: You don't normally create instances of this class
+   yourself. They are returned from calls to <a href="#xlwt.Workbook.add_sheet-method">Workbook.add_sheet</a>
+  </p>
+</dd>
+<dt><a id='xlwt.Worksheet.write-method'
+name='xlwt.Worksheet.write-method'><b>write(r, c, label="", style=Style.default_style)</b></a> [<a href='#xlwt.Worksheet.write-method'>#</a>]</dt>
+<dd>
+  <p>
+  This method is used to write a cell to a Worksheet..
+  </p>
+  <dl>
+    <dt><i>r</i></dt>
+    <dd>
+      	The zero-relative number of the row in the worksheet to which the cell should be written.
+    </dd>
+    <dt><i>c</i></dt>
+    <dd>
+      	The zero-relative number of the column in the worksheet to which the cell should be written.
+    </dd>
+    <dt><i>label</i></dt>
+    <dd>
+    	The data value to be written.
+    	An int, long, or decimal.Decimal instance is converted to float.
+    	A unicode instance is written as is.
+    	A str instance is converted to unicode using the encoding (default: 'ascii') specified
+    	when the Workbook instance was created.
+    	A datetime.datetime, datetime.date, or datetime.time instance is converted into Excel date format
+    	(a float representing the number of days since (typically) 1899-12-31T00:00:00,
+    	under the pretence that 1900 was a leap year).
+    	A bool instance will show up as TRUE or FALSE in Excel.
+    	None causes the cell to be blank -- no data, only formatting.
+    	An xlwt.Formula instance causes an Excel formula to be written.
+    	<em style="color:red;">[NC]</em>
+    </dd>
+    <dt><i>style</i></dt>
+    <dd>
+    	A style -- also known as an XF (extended format) -- is an XFStyle object, which encapsulates
+    	the formatting applied to the cell and its contents. XFStyle objects are best set up using the
+    	<i>easyxf</i> function. They may also be set up by setting attributes in
+    	Alignment, Borders, Pattern, Font and Protection objects
+    	then setting those objects and a format string as attributes of an XFStyle object.
+    	<em style="color:red;">[NC]</em>
+    </dd>
+  </dl>
+  <br />
+</dd>
+</dl>
+
+</body></html>

tablib/packages/xlwt/examples/big-16Mb.py

@@ -0,0 +1,35 @@
+#!/usr/bin/env python
+# tries stress SST, SAT and MSAT
+
+from time import *
+from xlwt.Workbook import *
+from xlwt.Style import *
+
+style = XFStyle()
+
+wb = Workbook()
+ws0 = wb.add_sheet('0')
+
+colcount = 200 + 1
+rowcount = 6000 + 1
+
+t0 = time()
+print "\nstart: %s" % ctime(t0)
+
+print "Filling..."
+for col in xrange(colcount):
+    print "[%d]" % col, 
+    for row in xrange(rowcount):
+        #ws0.write(row, col, "BIG(%d, %d)" % (row, col))
+        ws0.write(row, col, "BIG")
+
+t1 = time() - t0
+print "\nsince starting elapsed %.2f s" % (t1)
+
+print "Storing..."
+wb.save('big-16Mb.xls')
+
+t2 = time() - t0
+print "since starting elapsed %.2f s" % (t2)
+
+

tablib/packages/xlwt/examples/big-35Mb.py

@@ -0,0 +1,34 @@
+#!/usr/bin/env python
+# tries stress SST, SAT and MSAT
+
+from time import *
+from xlwt import *
+
+style = XFStyle()
+
+wb = Workbook()
+ws0 = wb.add_sheet('0')
+
+colcount = 200 + 1
+rowcount = 6000 + 1
+
+t0 = time()
+print "\nstart: %s" % ctime(t0)
+
+print "Filling..."
+for col in xrange(colcount):
+    print "[%d]" % col, 
+    for row in xrange(rowcount):
+        ws0.write(row, col, "BIG(%d, %d)" % (row, col))
+        #ws0.write(row, col, "BIG")
+
+t1 = time() - t0
+print "\nsince starting elapsed %.2f s" % (t1)
+
+print "Storing..."
+wb.save('big-35Mb.xls')
+
+t2 = time() - t0
+print "since starting elapsed %.2f s" % (t2)
+
+

tablib/packages/xlwt/examples/blanks.py

@@ -0,0 +1,36 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+font0 = Font()
+font0.name = 'Times New Roman'
+font0.struck_out = True
+font0.bold = True
+
+style0 = XFStyle()
+style0.font = font0
+
+
+wb = Workbook()
+ws0 = wb.add_sheet('0')
+
+ws0.write(1, 1, 'Test', style0)
+
+for i in range(0, 0x53):
+    borders = Borders()
+    borders.left = i
+    borders.right = i
+    borders.top = i
+    borders.bottom = i
+
+    style = XFStyle()
+    style.borders = borders
+
+    ws0.write(i, 2, '', style)
+    ws0.write(i, 3, hex(i), style0)
+
+ws0.write_merge(5, 8, 6, 10, "")
+
+wb.save('blanks.xls')

tablib/packages/xlwt/examples/col_width.py

@@ -0,0 +1,19 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+__rev_id__ = """$Id: col_width.py 3315 2008-03-14 14:44:52Z chris $"""
+
+
+from xlwt import *
+
+w = Workbook()
+ws = w.add_sheet('Hey, Dude')
+
+for i in range(6, 80):
+    fnt = Font()
+    fnt.height = i*20
+    style = XFStyle()
+    style.font = fnt
+    ws.write(1, i, 'Test')
+    ws.col(i).width = 0x0d00 + i
+w.save('col_width.xls')

tablib/packages/xlwt/examples/country.py

@@ -0,0 +1,10 @@
+#!/usr/bin/env python
+# -*- coding: windows-1252 -*-
+# Copyright (C) 2007 John Machin
+
+from xlwt import *
+
+w = Workbook()
+w.country_code = 61
+ws = w.add_sheet('AU')
+w.save('country.xls')

tablib/packages/xlwt/examples/dates.py

@@ -0,0 +1,37 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+from datetime import datetime
+
+w = Workbook()
+ws = w.add_sheet('Hey, Dude')
+
+fmts = [
+    'M/D/YY',
+    'D-MMM-YY',
+    'D-MMM',
+    'MMM-YY',
+    'h:mm AM/PM',
+    'h:mm:ss AM/PM',
+    'h:mm',
+    'h:mm:ss',
+    'M/D/YY h:mm',
+    'mm:ss',
+    '[h]:mm:ss',
+    'mm:ss.0',
+]
+
+i = 0
+for fmt in fmts:
+    ws.write(i, 0, fmt)
+
+    style = XFStyle()
+    style.num_format_str = fmt
+
+    ws.write(i, 4, datetime.now(), style)
+
+    i += 1
+
+w.save('dates.xls')

tablib/packages/xlwt/examples/format.py

@@ -0,0 +1,38 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+font0 = Font()
+font0.name = 'Times New Roman'
+font0.struck_out = True
+font0.bold = True
+
+style0 = XFStyle()
+style0.font = font0
+
+
+wb = Workbook()
+ws0 = wb.add_sheet('0')
+
+ws0.write(1, 1, 'Test', style0)
+
+for i in range(0, 0x53):
+    fnt = Font()
+    fnt.name = 'Arial'
+    fnt.colour_index = i
+    fnt.outline = True
+
+    borders = Borders()
+    borders.left = i
+
+    style = XFStyle()
+    style.font = fnt
+    style.borders = borders
+
+    ws0.write(i, 2, 'colour', style)
+    ws0.write(i, 3, hex(i), style0)
+
+
+wb.save('format.xls')

tablib/packages/xlwt/examples/formula_names.py

@@ -0,0 +1,34 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+from xlwt.ExcelFormulaParser import FormulaParseException
+
+w = Workbook()
+ws = w.add_sheet('F')
+
+## This example is a little silly since the formula building is
+## so simplistic that it often fails because the generated text
+## has the wrong number of parameters for the function being
+## tested.
+
+i = 0
+succeed_count = 0
+fail_count = 0
+for n in sorted(ExcelMagic.std_func_by_name):
+    ws.write(i, 0, n)
+    text = n + "($A$1)"
+    try:
+        formula = Formula(text)
+    except FormulaParseException,e:
+        print "Could not parse %r: %s" % (text,e.args[0])
+        fail_count += 1
+    else:
+        ws.write(i, 3, formula)
+        succeed_count += 1
+    i += 1
+
+w.save('formula_names.xls')
+
+print "succeeded with %i functions, failed with %i" % (succeed_count,fail_count)

tablib/packages/xlwt/examples/formulas.py

@@ -0,0 +1,47 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+w = Workbook()
+ws = w.add_sheet('F')
+
+ws.write(0, 0, Formula("-(1+1)"))
+ws.write(1, 0, Formula("-(1+1)/(-2-2)"))
+ws.write(2, 0, Formula("-(134.8780789+1)"))
+ws.write(3, 0, Formula("-(134.8780789e-10+1)"))
+ws.write(4, 0, Formula("-1/(1+1)+9344"))
+
+ws.write(0, 1, Formula("-(1+1)"))
+ws.write(1, 1, Formula("-(1+1)/(-2-2)"))
+ws.write(2, 1, Formula("-(134.8780789+1)"))
+ws.write(3, 1, Formula("-(134.8780789e-10+1)"))
+ws.write(4, 1, Formula("-1/(1+1)+9344"))
+
+ws.write(0, 2, Formula("A1*B1"))
+ws.write(1, 2, Formula("A2*B2"))
+ws.write(2, 2, Formula("A3*B3"))
+ws.write(3, 2, Formula("A4*B4*sin(pi()/4)"))
+ws.write(4, 2, Formula("A5%*B5*pi()/1000"))
+
+##############
+## NOTE: parameters are separated by semicolon!!!
+##############
+
+
+ws.write(5, 2, Formula("C1+C2+C3+C4+C5/(C1+C2+C3+C4/(C1+C2+C3+C4/(C1+C2+C3+C4)+C5)+C5)-20.3e-2"))
+ws.write(5, 3, Formula("C1^2"))
+ws.write(6, 2, Formula("SUM(C1;C2;;;;;C3;;;C4)"))
+ws.write(6, 3, Formula("SUM($A$1:$C$5)"))
+
+ws.write(7, 0, Formula('"lkjljllkllkl"'))
+ws.write(7, 1, Formula('"yuyiyiyiyi"'))
+ws.write(7, 2, Formula('A8 & B8 & A8'))
+ws.write(8, 2, Formula('now()'))
+
+ws.write(10, 2, Formula('TRUE'))
+ws.write(11, 2, Formula('FALSE'))
+ws.write(12, 3, Formula('IF(A1>A2;3;"hkjhjkhk")'))
+
+w.save('formulas.xls')

tablib/packages/xlwt/examples/hyperlinks.py

@@ -0,0 +1,28 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+f = Font()
+f.height = 20*72
+f.name = 'Verdana'
+f.bold = True
+f.underline = Font.UNDERLINE_DOUBLE
+f.colour_index = 4
+
+h_style = XFStyle()
+h_style.font = f
+
+w = Workbook()
+ws = w.add_sheet('F')
+
+##############
+## NOTE: parameters are separated by semicolon!!!
+##############
+
+n = "HYPERLINK"
+ws.write_merge(1, 1, 1, 10, Formula(n + '("http://www.irs.gov/pub/irs-pdf/f1000.pdf";"f1000.pdf")'), h_style)
+ws.write_merge(2, 2, 2, 25, Formula(n + '("mailto:roman.kiseliov@gmail.com?subject=pyExcelerator-feedback&Body=Hello,%20Roman!";"pyExcelerator-feedback")'), h_style)
+
+w.save("hyperlinks.xls")

tablib/packages/xlwt/examples/image.py

@@ -0,0 +1,12 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+w = Workbook()
+ws = w.add_sheet('Image')
+ws.insert_bitmap('python.bmp', 2, 2)
+ws.insert_bitmap('python.bmp', 10, 2)
+
+w.save('image.xls')

tablib/packages/xlwt/examples/merged.py

@@ -0,0 +1,39 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+fnt = Font()
+fnt.name = 'Arial'
+fnt.colour_index = 4
+fnt.bold = True
+
+borders = Borders()
+borders.left = 6
+borders.right = 6
+borders.top = 6
+borders.bottom = 6
+
+al = Alignment()
+al.horz = Alignment.HORZ_CENTER
+al.vert = Alignment.VERT_CENTER
+
+style = XFStyle()
+style.font = fnt
+style.borders = borders
+style.alignment = al
+
+
+wb = Workbook()
+ws0 = wb.add_sheet('sheet0')
+ws1 = wb.add_sheet('sheet1')
+ws2 = wb.add_sheet('sheet2')
+
+for i in range(0, 0x200, 2):
+    ws0.write_merge(i, i+1, 1, 5, 'test %d' % i, style)
+    ws1.write_merge(i, i, 1, 7, 'test %d' % i, style)
+    ws2.write_merge(i, i+1, 1, 7 + (i%10), 'test %d' % i, style)
+
+
+wb.save('merged.xls')

tablib/packages/xlwt/examples/merged0.py

@@ -0,0 +1,30 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+wb = Workbook()
+ws0 = wb.add_sheet('sheet0')
+
+
+fnt = Font()
+fnt.name = 'Arial'
+fnt.colour_index = 4
+fnt.bold = True
+
+borders = Borders()
+borders.left = 6
+borders.right = 6
+borders.top = 6
+borders.bottom = 6
+
+style = XFStyle()
+style.font = fnt
+style.borders = borders
+
+ws0.write_merge(3, 3, 1, 5, 'test1', style)
+ws0.write_merge(4, 10, 1, 5, 'test2', style)
+ws0.col(1).width = 0x0d00
+
+wb.save('merged0.xls')

tablib/packages/xlwt/examples/merged1.py

@@ -0,0 +1,102 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+wb = Workbook()
+ws0 = wb.add_sheet('sheet0')
+
+fnt1 = Font()
+fnt1.name = 'Verdana'
+fnt1.bold = True
+fnt1.height = 18*0x14
+
+pat1 = Pattern()
+pat1.pattern = Pattern.SOLID_PATTERN
+pat1.pattern_fore_colour = 0x16
+
+brd1 = Borders()
+brd1.left = 0x06
+brd1.right = 0x06
+brd1.top = 0x06
+brd1.bottom = 0x06
+
+fnt2 = Font()
+fnt2.name = 'Verdana'
+fnt2.bold = True
+fnt2.height = 14*0x14
+
+brd2 = Borders()
+brd2.left = 0x01
+brd2.right = 0x01
+brd2.top = 0x01
+brd2.bottom = 0x01
+
+pat2 = Pattern()
+pat2.pattern = Pattern.SOLID_PATTERN
+pat2.pattern_fore_colour = 0x01F
+
+fnt3 = Font()
+fnt3.name = 'Verdana'
+fnt3.bold = True
+fnt3.italic = True
+fnt3.height = 12*0x14
+
+brd3 = Borders()
+brd3.left = 0x07
+brd3.right = 0x07
+brd3.top = 0x07
+brd3.bottom = 0x07
+
+fnt4 = Font()
+
+al1 = Alignment()
+al1.horz = Alignment.HORZ_CENTER
+al1.vert = Alignment.VERT_CENTER
+
+al2 = Alignment()
+al2.horz = Alignment.HORZ_RIGHT
+al2.vert = Alignment.VERT_CENTER
+
+al3 = Alignment()
+al3.horz = Alignment.HORZ_LEFT
+al3.vert = Alignment.VERT_CENTER
+
+style1 = XFStyle()
+style1.font = fnt1
+style1.alignment = al1
+style1.pattern = pat1
+style1.borders = brd1
+
+style2 = XFStyle()
+style2.font = fnt2
+style2.alignment = al1
+style2.pattern = pat2
+style2.borders = brd2
+
+style3 = XFStyle()
+style3.font = fnt3
+style3.alignment = al1
+style3.pattern = pat2
+style3.borders = brd3
+
+price_style = XFStyle()
+price_style.font = fnt4
+price_style.alignment = al2
+price_style.borders = brd3
+price_style.num_format_str = '_(#,##0.00_) "money"'
+
+ware_style = XFStyle()
+ware_style.font = fnt4
+ware_style.alignment = al3
+ware_style.borders = brd3
+
+
+ws0.merge(3, 3, 1, 5, style1)
+ws0.merge(4, 10, 1, 6, style2)
+ws0.merge(14, 16, 1, 7, style3)
+ws0.col(1).width = 0x0d00
+
+
+wb.save('merged1.xls')

tablib/packages/xlwt/examples/mini.py

@@ -0,0 +1,9 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+w = Workbook()
+ws = w.add_sheet('xlwt was here')
+w.save('mini.xls')

tablib/packages/xlwt/examples/num_formats.py

@@ -0,0 +1,60 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+w = Workbook()
+ws = w.add_sheet('Hey, Dude')
+
+fmts = [
+    'general',
+    '0',
+    '0.00',
+    '#,##0',
+    '#,##0.00',
+    '"$"#,##0_);("$"#,##',
+    '"$"#,##0_);[Red]("$"#,##',
+    '"$"#,##0.00_);("$"#,##',
+    '"$"#,##0.00_);[Red]("$"#,##',
+    '0%',
+    '0.00%',
+    '0.00E+00',
+    '# ?/?',
+    '# ??/??',
+    'M/D/YY',
+    'D-MMM-YY',
+    'D-MMM',
+    'MMM-YY',
+    'h:mm AM/PM',
+    'h:mm:ss AM/PM',
+    'h:mm',
+    'h:mm:ss',
+    'M/D/YY h:mm',
+    '_(#,##0_);(#,##0)',
+    '_(#,##0_);[Red](#,##0)',
+    '_(#,##0.00_);(#,##0.00)',
+    '_(#,##0.00_);[Red](#,##0.00)',
+    '_("$"* #,##0_);_("$"* (#,##0);_("$"* "-"_);_(@_)',
+    '_(* #,##0_);_(* (#,##0);_(* "-"_);_(@_)',
+    '_("$"* #,##0.00_);_("$"* (#,##0.00);_("$"* "-"??_);_(@_)',
+    '_(* #,##0.00_);_(* (#,##0.00);_(* "-"??_);_(@_)',
+    'mm:ss',
+    '[h]:mm:ss',
+    'mm:ss.0',
+    '##0.0E+0',
+    '@'   
+]
+
+i = 0
+for fmt in fmts:
+    ws.write(i, 0, fmt)
+
+    style = XFStyle()
+    style.num_format_str = fmt
+
+    ws.write(i, 4, -1278.9078, style)
+
+    i += 1
+
+w.save('num_formats.xls')

tablib/packages/xlwt/examples/numbers.py

@@ -0,0 +1,25 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+w = Workbook()
+ws = w.add_sheet('Hey, Dude')
+
+ws.write(0, 0, 1)
+ws.write(1, 0, 1.23)
+ws.write(2, 0, 12345678)
+ws.write(3, 0, 123456.78)
+
+ws.write(0, 1, -1)
+ws.write(1, 1, -1.23)
+ws.write(2, 1, -12345678)
+ws.write(3, 1, -123456.78)
+
+ws.write(0, 2, -17867868678687.0)
+ws.write(1, 2, -1.23e-5)
+ws.write(2, 2, -12345678.90780980)
+ws.write(3, 2, -123456.78)
+
+w.save('numbers.xls')

tablib/packages/xlwt/examples/outline.py

@@ -0,0 +1,113 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+fnt = Font()
+fnt.name = 'Arial'
+fnt.colour_index = 4
+fnt.bold = True
+
+borders = Borders()
+borders.left = 6
+borders.right = 6
+borders.top = 6
+borders.bottom = 6
+
+style = XFStyle()
+style.font = fnt
+style.borders = borders
+
+wb = Workbook()
+
+ws0 = wb.add_sheet('Rows Outline')
+
+ws0.write_merge(1, 1, 1, 5, 'test 1', style)
+ws0.write_merge(2, 2, 1, 4, 'test 1', style)
+ws0.write_merge(3, 3, 1, 3, 'test 2', style)
+ws0.write_merge(4, 4, 1, 4, 'test 1', style)
+ws0.write_merge(5, 5, 1, 4, 'test 3', style)
+ws0.write_merge(6, 6, 1, 5, 'test 1', style)
+ws0.write_merge(7, 7, 1, 5, 'test 4', style)
+ws0.write_merge(8, 8, 1, 4, 'test 1', style)
+ws0.write_merge(9, 9, 1, 3, 'test 5', style)
+
+ws0.row(1).level = 1
+ws0.row(2).level = 1
+ws0.row(3).level = 2
+ws0.row(4).level = 2
+ws0.row(5).level = 2
+ws0.row(6).level = 2
+ws0.row(7).level = 2
+ws0.row(8).level = 1
+ws0.row(9).level = 1
+
+
+ws1 = wb.add_sheet('Columns Outline')
+
+ws1.write_merge(1, 1, 1, 5, 'test 1', style)
+ws1.write_merge(2, 2, 1, 4, 'test 1', style)
+ws1.write_merge(3, 3, 1, 3, 'test 2', style)
+ws1.write_merge(4, 4, 1, 4, 'test 1', style)
+ws1.write_merge(5, 5, 1, 4, 'test 3', style)
+ws1.write_merge(6, 6, 1, 5, 'test 1', style)
+ws1.write_merge(7, 7, 1, 5, 'test 4', style)
+ws1.write_merge(8, 8, 1, 4, 'test 1', style)
+ws1.write_merge(9, 9, 1, 3, 'test 5', style)
+
+ws1.col(1).level = 1
+ws1.col(2).level = 1
+ws1.col(3).level = 2
+ws1.col(4).level = 2
+ws1.col(5).level = 2
+ws1.col(6).level = 2
+ws1.col(7).level = 2
+ws1.col(8).level = 1
+ws1.col(9).level = 1
+
+
+ws2 = wb.add_sheet('Rows and Columns Outline')
+
+ws2.write_merge(1, 1, 1, 5, 'test 1', style)
+ws2.write_merge(2, 2, 1, 4, 'test 1', style)
+ws2.write_merge(3, 3, 1, 3, 'test 2', style)
+ws2.write_merge(4, 4, 1, 4, 'test 1', style)
+ws2.write_merge(5, 5, 1, 4, 'test 3', style)
+ws2.write_merge(6, 6, 1, 5, 'test 1', style)
+ws2.write_merge(7, 7, 1, 5, 'test 4', style)
+ws2.write_merge(8, 8, 1, 4, 'test 1', style)
+ws2.write_merge(9, 9, 1, 3, 'test 5', style)
+
+ws2.row(1).level = 1
+ws2.row(2).level = 1
+ws2.row(3).level = 2
+ws2.row(4).level = 2
+ws2.row(5).level = 2
+ws2.row(6).level = 2
+ws2.row(7).level = 2
+ws2.row(8).level = 1
+ws2.row(9).level = 1
+
+ws2.write_merge(1, 1, 1, 5, 'test 1', style)
+ws2.write_merge(2, 2, 1, 4, 'test 1', style)
+ws2.write_merge(3, 3, 1, 3, 'test 2', style)
+ws2.write_merge(4, 4, 1, 4, 'test 1', style)
+ws2.write_merge(5, 5, 1, 4, 'test 3', style)
+ws2.write_merge(6, 6, 1, 5, 'test 1', style)
+ws2.write_merge(7, 7, 1, 5, 'test 4', style)
+ws2.write_merge(8, 8, 1, 4, 'test 1', style)
+ws2.write_merge(9, 9, 1, 3, 'test 5', style)
+
+ws2.col(1).level = 1
+ws2.col(2).level = 1
+ws2.col(3).level = 2
+ws2.col(4).level = 2
+ws2.col(5).level = 2
+ws2.col(6).level = 2
+ws2.col(7).level = 2
+ws2.col(8).level = 1
+ws2.col(9).level = 1
+
+
+wb.save('outline.xls')

tablib/packages/xlwt/examples/panes.py

@@ -0,0 +1,58 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+w = Workbook()
+ws1 = w.add_sheet('sheet 1')
+ws2 = w.add_sheet('sheet 2')
+ws3 = w.add_sheet('sheet 3')
+ws4 = w.add_sheet('sheet 4')
+ws5 = w.add_sheet('sheet 5')
+ws6 = w.add_sheet('sheet 6')
+
+for i in range(0x100):
+    ws1.write(i/0x10, i%0x10, i)
+
+for i in range(0x100):
+    ws2.write(i/0x10, i%0x10, i)
+
+for i in range(0x100):
+    ws3.write(i/0x10, i%0x10, i)
+
+for i in range(0x100):
+    ws4.write(i/0x10, i%0x10, i)
+
+for i in range(0x100):
+    ws5.write(i/0x10, i%0x10, i)
+
+for i in range(0x100):
+    ws6.write(i/0x10, i%0x10, i)
+
+ws1.panes_frozen = True
+ws1.horz_split_pos = 2
+
+ws2.panes_frozen = True
+ws2.vert_split_pos = 2
+
+ws3.panes_frozen = True
+ws3.horz_split_pos = 1
+ws3.vert_split_pos = 1
+
+ws4.panes_frozen = False
+ws4.horz_split_pos = 12
+ws4.horz_split_first_visible = 2
+
+ws5.panes_frozen = False
+ws5.vert_split_pos = 40
+ws4.vert_split_first_visible = 2
+
+ws6.panes_frozen = False
+ws6.horz_split_pos = 12
+ws4.horz_split_first_visible = 2
+ws6.vert_split_pos = 40
+ws4.vert_split_first_visible = 2
+
+w.save('panes.xls')
+

tablib/packages/xlwt/examples/parse-fmla.py

@@ -0,0 +1,12 @@
+from xlwt import ExcelFormulaParser, ExcelFormula
+import sys
+
+f = ExcelFormula.Formula(
+""" -((1.80 + 2.898 * 1)/(1.80 + 2.898))*
+AVERAGE((1.80 + 2.898 * 1)/(1.80 + 2.898); 
+        (1.80 + 2.898 * 1)/(1.80 + 2.898); 
+        (1.80 + 2.898 * 1)/(1.80 + 2.898)) + 
+SIN(PI()/4)""")
+
+#for t in f.rpn():
+#    print "%15s %15s" % (ExcelFormulaParser.PtgNames[t[0]], t[1])

tablib/packages/xlwt/examples/protection.py

@@ -0,0 +1,122 @@
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+fnt = Font()
+fnt.name = 'Arial'
+fnt.colour_index = 4
+fnt.bold = True
+
+borders = Borders()
+borders.left = 6
+borders.right = 6
+borders.top = 6
+borders.bottom = 6
+
+style = XFStyle()
+style.font = fnt
+style.borders = borders
+
+wb = Workbook()
+
+ws0 = wb.add_sheet('Rows Outline')
+
+ws0.write_merge(1, 1, 1, 5, 'test 1', style)
+ws0.write_merge(2, 2, 1, 4, 'test 1', style)
+ws0.write_merge(3, 3, 1, 3, 'test 2', style)
+ws0.write_merge(4, 4, 1, 4, 'test 1', style)
+ws0.write_merge(5, 5, 1, 4, 'test 3', style)
+ws0.write_merge(6, 6, 1, 5, 'test 1', style)
+ws0.write_merge(7, 7, 1, 5, 'test 4', style)
+ws0.write_merge(8, 8, 1, 4, 'test 1', style)
+ws0.write_merge(9, 9, 1, 3, 'test 5', style)
+
+ws0.row(1).level = 1
+ws0.row(2).level = 1
+ws0.row(3).level = 2
+ws0.row(4).level = 2
+ws0.row(5).level = 2
+ws0.row(6).level = 2
+ws0.row(7).level = 2
+ws0.row(8).level = 1
+ws0.row(9).level = 1
+
+
+ws1 = wb.add_sheet('Columns Outline')
+
+ws1.write_merge(1, 1, 1, 5, 'test 1', style)
+ws1.write_merge(2, 2, 1, 4, 'test 1', style)
+ws1.write_merge(3, 3, 1, 3, 'test 2', style)
+ws1.write_merge(4, 4, 1, 4, 'test 1', style)
+ws1.write_merge(5, 5, 1, 4, 'test 3', style)
+ws1.write_merge(6, 6, 1, 5, 'test 1', style)
+ws1.write_merge(7, 7, 1, 5, 'test 4', style)
+ws1.write_merge(8, 8, 1, 4, 'test 1', style)
+ws1.write_merge(9, 9, 1, 3, 'test 5', style)
+
+ws1.col(1).level = 1
+ws1.col(2).level = 1
+ws1.col(3).level = 2
+ws1.col(4).level = 2
+ws1.col(5).level = 2
+ws1.col(6).level = 2
+ws1.col(7).level = 2
+ws1.col(8).level = 1
+ws1.col(9).level = 1
+
+
+ws2 = wb.add_sheet('Rows and Columns Outline')
+
+ws2.write_merge(1, 1, 1, 5, 'test 1', style)
+ws2.write_merge(2, 2, 1, 4, 'test 1', style)
+ws2.write_merge(3, 3, 1, 3, 'test 2', style)
+ws2.write_merge(4, 4, 1, 4, 'test 1', style)
+ws2.write_merge(5, 5, 1, 4, 'test 3', style)
+ws2.write_merge(6, 6, 1, 5, 'test 1', style)
+ws2.write_merge(7, 7, 1, 5, 'test 4', style)
+ws2.write_merge(8, 8, 1, 4, 'test 1', style)
+ws2.write_merge(9, 9, 1, 3, 'test 5', style)
+
+ws2.row(1).level = 1
+ws2.row(2).level = 1
+ws2.row(3).level = 2
+ws2.row(4).level = 2
+ws2.row(5).level = 2
+ws2.row(6).level = 2
+ws2.row(7).level = 2
+ws2.row(8).level = 1
+ws2.row(9).level = 1
+
+ws2.col(1).level = 1
+ws2.col(2).level = 1
+ws2.col(3).level = 2
+ws2.col(4).level = 2
+ws2.col(5).level = 2
+ws2.col(6).level = 2
+ws2.col(7).level = 2
+ws2.col(8).level = 1
+ws2.col(9).level = 1
+
+
+ws0.protect = True
+ws0.wnd_protect = True
+ws0.obj_protect = True
+ws0.scen_protect = True
+ws0.password = "123456"
+
+ws1.protect = True
+ws1.wnd_protect = True
+ws1.obj_protect = True
+ws1.scen_protect = True
+ws1.password = "abcdefghij"
+
+ws2.protect = True
+ws2.wnd_protect = True
+ws2.obj_protect = True
+ws2.scen_protect = True
+ws2.password = "ok"
+
+wb.protect = True
+wb.wnd_protect = True
+wb.obj_protect = True
+wb.save('protection.xls')

tablib/packages/xlwt/examples/row_styles.py

@@ -0,0 +1,17 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+w = Workbook()
+ws = w.add_sheet('Hey, Dude')
+
+for i in range(6, 80):
+    fnt = Font()
+    fnt.height = i*20
+    style = XFStyle()
+    style.font = fnt
+    ws.write(i, 1, 'Test')
+    ws.row(i).set_style(style)
+w.save('row_styles.xls')

tablib/packages/xlwt/examples/row_styles_empty.py

@@ -0,0 +1,18 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+__rev_id__ = """$Id: row_styles_empty.py 3309 2008-03-14 11:04:30Z chris $"""
+
+
+from pyExcelerator import *
+
+w = Workbook()
+ws = w.add_sheet('Hey, Dude')
+
+for i in range(6, 80):
+    fnt = Font()
+    fnt.height = i*20
+    style = XFStyle()
+    style.font = fnt
+    ws.row(i).set_style(style)
+w.save('row_styles_empty.xls')

tablib/packages/xlwt/examples/simple.py

@@ -0,0 +1,24 @@
+import xlwt
+from datetime import datetime
+
+font0 = xlwt.Font()
+font0.name = 'Times New Roman'
+font0.colour_index = 2
+font0.bold = True
+
+style0 = xlwt.XFStyle()
+style0.font = font0
+
+style1 = xlwt.XFStyle()
+style1.num_format_str = 'D-MMM-YY'
+
+wb = xlwt.Workbook()
+ws = wb.add_sheet('A Test Sheet')
+
+ws.write(0, 0, 'Test', style0)
+ws.write(1, 0, datetime.now(), style1)
+ws.write(2, 0, 1)
+ws.write(2, 1, 1)
+ws.write(2, 2, xlwt.Formula("A3+B3"))
+
+wb.save('example.xls')

tablib/packages/xlwt/examples/sst.py

@@ -0,0 +1,52 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+font0 = Formatting.Font()
+font0.name = 'Arial'
+font1 = Formatting.Font()
+font1.name = 'Arial Cyr'
+font2 = Formatting.Font()
+font2.name = 'Times New Roman'
+font3 = Formatting.Font()
+font3.name = 'Courier New Cyr'
+
+num_format0 = '0.00000'
+num_format1 = '0.000000'
+num_format2 = '0.0000000'
+num_format3 = '0.00000000'
+
+st0 = XFStyle()
+st1 = XFStyle()
+st2 = XFStyle()
+st3 = XFStyle()
+st4 = XFStyle()
+
+st0.font = font0
+st0.num_format = num_format0
+
+st1.font = font1
+st1.num_format = num_format1
+
+st2.font = font2
+st2.num_format = num_format2
+
+st3.font = font3
+st3.num_format = num_format3
+
+wb = Workbook()
+
+wb.add_style(st0)
+wb.add_style(st1)
+wb.add_style(st2)
+wb.add_style(st3)
+
+ws0 = wb.add_sheet('0')
+ws0.write(0, 0, 'Olya'*0x4000, st0)
+
+#for i in range(0, 0x10):
+#    ws0.write(i, 2, ('%d'%i)*0x4000, st1)
+    
+wb.save('sst.xls')

tablib/packages/xlwt/examples/unicode0.py

@@ -0,0 +1,12 @@
+#!/usr/bin/env python
+import xlwt
+
+# Strings passed to (for example) Worksheet.write can be unicode objects,
+# or str (8-bit) objects, which are then decoded into unicode.
+# The encoding to be used defaults to 'ascii'. This can be overridden
+# when the Workbook instance is created:
+
+book = xlwt.Workbook(encoding='cp1251')
+sheet = book.add_sheet('cp1251-demo')
+sheet.write(0, 0, '\xce\xeb\xff')
+book.save('unicode0.xls')

tablib/packages/xlwt/examples/unicode1.py

@@ -0,0 +1,28 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+w = Workbook()
+ws1 = w.add_sheet(u'\N{GREEK SMALL LETTER ALPHA}\N{GREEK SMALL LETTER BETA}\N{GREEK SMALL LETTER GAMMA}')
+
+ws1.write(0, 0, u'\N{GREEK SMALL LETTER ALPHA}\N{GREEK SMALL LETTER BETA}\N{GREEK SMALL LETTER GAMMA}')
+ws1.write(1, 1, u'\N{GREEK SMALL LETTER DELTA}x = 1 + \N{GREEK SMALL LETTER DELTA}')
+
+ws1.write(2,0, u'A\u2262\u0391.')     # RFC2152 example
+ws1.write(3,0, u'Hi Mom -\u263a-!')   # RFC2152 example
+ws1.write(4,0, u'\u65E5\u672C\u8A9E') # RFC2152 example
+ws1.write(5,0, u'Item 3 is \u00a31.') # RFC2152 example
+ws1.write(8,0, u'\N{INTEGRAL}')       # RFC2152 example
+
+w.add_sheet(u'A\u2262\u0391.')     # RFC2152 example
+w.add_sheet(u'Hi Mom -\u263a-!')   # RFC2152 example
+one_more_ws = w.add_sheet(u'\u65E5\u672C\u8A9E') # RFC2152 example
+w.add_sheet(u'Item 3 is \u00a31.') # RFC2152 example
+
+one_more_ws.write(0, 0, u'\u2665\u2665')
+
+w.add_sheet(u'\N{GREEK SMALL LETTER ETA WITH TONOS}')
+w.save('unicode1.xls')
+

tablib/packages/xlwt/examples/unicode2.py

@@ -0,0 +1,19 @@
+#!/usr/bin/env python
+# -*- coding: windows-1251 -*-
+# Copyright (C) 2005 Kiseliov Roman
+
+from xlwt import *
+
+w = Workbook()
+ws1 = w.add_sheet(u'\N{GREEK SMALL LETTER ALPHA}\N{GREEK SMALL LETTER BETA}\N{GREEK SMALL LETTER GAMMA}\u2665\u041e\u041b\u042f\u2665')
+
+fnt = Font()
+fnt.height = 26*20
+style = XFStyle()
+style.font = fnt
+
+for i in range(0x10000):
+    ws1.write(i/0x10, i%0x10, unichr(i), style)
+
+w.save('unicode2.xls')
+

tablib/packages/xlwt/examples/wsprops.py

@@ -0,0 +1,155 @@
+props = \
+[
+        'name',
+        'parent',
+        'rows',
+        'cols',
+        'merged_ranges',
+        'bmp_rec',
+        'show_formulas',
+        'show_grid',
+        'show_headers',
+        'panes_frozen',
+        'show_empty_as_zero',
+        'auto_colour_grid',
+        'cols_right_to_left',
+        'show_outline',
+        'remove_splits',
+        'selected',
+        'hidden',
+        'page_preview',
+        'first_visible_row',
+        'first_visible_col',
+        'grid_colour',
+        'preview_magn',
+        'normal_magn',
+        'row_gut_width',
+        'col_gut_height',
+        'show_auto_page_breaks',
+        'dialogue_sheet',
+        'auto_style_outline',
+        'outline_below',
+        'outline_right',
+        'fit_num_pages',
+        'show_row_outline',
+        'show_col_outline',
+        'alt_expr_eval',
+        'alt_formula_entries',
+        'row_default_height',
+        'col_default_width',
+        'calc_mode',
+        'calc_count',
+        'RC_ref_mode',
+        'iterations_on',
+        'delta',
+        'save_recalc',
+        'print_headers',
+        'print_grid',
+        'grid_set',
+        'vert_page_breaks',
+        'horz_page_breaks',
+        'header_str',
+        'footer_str',
+        'print_centered_vert',
+        'print_centered_horz',
+        'left_margin',
+        'right_margin',
+        'top_margin',
+        'bottom_margin',
+        'paper_size_code',
+        'print_scaling',
+        'start_page_number',
+        'fit_width_to_pages',
+        'fit_height_to_pages',
+        'print_in_rows',
+        'portrait',
+        'print_not_colour',
+        'print_draft',
+        'print_notes',
+        'print_notes_at_end',
+        'print_omit_errors',
+        'print_hres',
+        'print_vres',
+        'header_margin',
+        'footer_margin',
+        'copies_num',
+]
+
+from xlwt import *
+
+wb = Workbook()
+ws = wb.add_sheet('sheet')
+
+print ws.name
+print ws.parent
+print ws.rows
+print ws.cols
+print ws.merged_ranges
+print ws.bmp_rec
+print ws.show_formulas
+print ws.show_grid
+print ws.show_headers
+print ws.panes_frozen
+print ws.show_empty_as_zero
+print ws.auto_colour_grid
+print ws.cols_right_to_left
+print ws.show_outline
+print ws.remove_splits
+print ws.selected
+# print ws.hidden
+print ws.page_preview
+print ws.first_visible_row
+print ws.first_visible_col
+print ws.grid_colour
+print ws.preview_magn
+print ws.normal_magn
+#print ws.row_gut_width
+#print ws.col_gut_height
+print ws.show_auto_page_breaks
+print ws.dialogue_sheet
+print ws.auto_style_outline
+print ws.outline_below
+print ws.outline_right
+print ws.fit_num_pages
+print ws.show_row_outline
+print ws.show_col_outline
+print ws.alt_expr_eval
+print ws.alt_formula_entries
+print ws.row_default_height
+print ws.col_default_width
+print ws.calc_mode
+print ws.calc_count
+print ws.RC_ref_mode
+print ws.iterations_on
+print ws.delta
+print ws.save_recalc
+print ws.print_headers
+print ws.print_grid
+#print ws.grid_set
+print ws.vert_page_breaks
+print ws.horz_page_breaks
+print ws.header_str
+print ws.footer_str
+print ws.print_centered_vert
+print ws.print_centered_horz
+print ws.left_margin
+print ws.right_margin
+print ws.top_margin
+print ws.bottom_margin
+print ws.paper_size_code
+print ws.print_scaling
+print ws.start_page_number
+print ws.fit_width_to_pages
+print ws.fit_height_to_pages
+print ws.print_in_rows
+print ws.portrait
+print ws.print_colour
+print ws.print_draft
+print ws.print_notes
+print ws.print_notes_at_end
+print ws.print_omit_errors
+print ws.print_hres
+print ws.print_vres
+print ws.header_margin
+print ws.footer_margin
+print ws.copies_num

tablib/packages/xlwt/examples/xlwt_easyxf_simple_demo.py

@@ -0,0 +1,46 @@
+
+# Write an XLS file with a single worksheet, containing
+# a heading row and some rows of data.
+
+import xlwt
+import datetime
+ezxf = xlwt.easyxf
+
+def write_xls(file_name, sheet_name, headings, data, heading_xf, data_xfs):
+    book = xlwt.Workbook()
+    sheet = book.add_sheet(sheet_name)
+    rowx = 0
+    for colx, value in enumerate(headings):
+        sheet.write(rowx, colx, value, heading_xf)
+    sheet.set_panes_frozen(True) # frozen headings instead of split panes
+    sheet.set_horz_split_pos(rowx+1) # in general, freeze after last heading row
+    sheet.set_remove_splits(True) # if user does unfreeze, don't leave a split there
+    for row in data:
+        rowx += 1
+        for colx, value in enumerate(row):
+            sheet.write(rowx, colx, value, data_xfs[colx])
+    book.save(file_name)
+
+if __name__ == '__main__':
+    import sys
+    mkd = datetime.date
+    hdngs = ['Date', 'Stock Code', 'Quantity', 'Unit Price', 'Value', 'Message']
+    kinds =  'date    text          int         price         money    text'.split()
+    data = [
+        [mkd(2007, 7, 1), 'ABC', 1000, 1.234567, 1234.57, ''],
+        [mkd(2007, 12, 31), 'XYZ', -100, 4.654321, -465.43, 'Goods returned'],
+        ] + [
+            [mkd(2008, 6, 30), 'PQRCD', 100, 2.345678, 234.57, ''],
+        ] * 100
+
+    heading_xf = ezxf('font: bold on; align: wrap on, vert centre, horiz center')
+    kind_to_xf_map = {
+        'date': ezxf(num_format_str='yyyy-mm-dd'),
+        'int': ezxf(num_format_str='#,##0'),
+        'money': ezxf('font: italic on; pattern: pattern solid, fore-colour grey25',
+            num_format_str='$#,##0.00'),
+        'price': ezxf(num_format_str='#0.000000'),
+        'text': ezxf(),
+        }
+    data_xfs = [kind_to_xf_map[k] for k in kinds]
+    write_xls('xlwt_easyxf_simple_demo.xls', 'Demo', hdngs, data, heading_xf, data_xfs)

tablib/packages/xlwt/excel-formula.g

@@ -0,0 +1,374 @@
+header {
+    import struct
+    import Utils
+    from UnicodeUtils import upack1
+    from ExcelMagic import *
+
+    _RVAdelta =     {"R": 0, "V": 0x20, "A": 0x40}
+    _RVAdeltaRef =  {"R": 0, "V": 0x20, "A": 0x40, "D": 0x20}
+    _RVAdeltaArea = {"R": 0, "V": 0x20, "A": 0x40, "D": 0}
+
+    
+    class FormulaParseException(Exception):
+        """
+        An exception indicating that a Formula could not be successfully parsed.
+        """
+}
+
+header "ExcelFormulaParser.__init__" {
+    self.rpn = ""
+    self.sheet_references = []
+    self.xcall_references = []
+}
+
+options {
+    language  = "Python";
+}
+
+class ExcelFormulaParser extends Parser;
+options {
+    k = 2;
+    defaultErrorHandler = false;
+    buildAST = false;
+}
+
+
+tokens {
+    TRUE_CONST;
+    FALSE_CONST;
+    STR_CONST;
+    NUM_CONST;
+    INT_CONST;
+
+    FUNC_IF;
+    FUNC_CHOOSE;
+    NAME;
+    QUOTENAME;
+
+    EQ;
+    NE;
+    GT;
+    LT;
+    GE;
+    LE;
+
+    ADD;
+    SUB;
+    MUL;
+    DIV;
+
+    POWER;
+    PERCENT;
+
+    LP;
+    RP;
+
+    LB;
+    RB;
+
+    COLON;
+    COMMA;
+    SEMICOLON;
+    REF2D;
+    REF2D_R1C1;
+    BANG;
+}
+
+formula
+    : expr["V"]
+    ;
+
+expr[arg_type]
+    : // {print "\n**expr %s" % arg_type}
+    prec0_expr[arg_type]
+        (
+            (
+                  EQ { op = struct.pack('B', ptgEQ) }
+                | NE { op = struct.pack('B', ptgNE) }
+                | GT { op = struct.pack('B', ptgGT) }
+                | LT { op = struct.pack('B', ptgLT) }
+                | GE { op = struct.pack('B', ptgGE) }
+                | LE { op = struct.pack('B', ptgLE) }
+            )
+            prec0_expr[arg_type] { self.rpn += op }
+        )*
+    ;
+
+prec0_expr[arg_type]
+    : prec1_expr[arg_type]
+        (
+            (
+                CONCAT { op = struct.pack('B', ptgConcat) }
+            )
+            prec1_expr[arg_type] { self.rpn += op }
+        )*
+    ;
+
+prec1_expr[arg_type]
+    : // {print "**prec1_expr1 %s" % arg_type}
+    prec2_expr[arg_type]
+    // {print "**prec1_expr2 %s" % arg_type}
+        (
+            (
+                  ADD { op = struct.pack('B', ptgAdd) }
+                | SUB { op = struct.pack('B', ptgSub) }
+            )
+            // {print "**prec1_expr3 %s" % arg_type}
+            prec2_expr[arg_type]
+            { self.rpn += op;
+            // print "**prec1_expr4 %s" % arg_type
+            }
+        )*
+    ;
+
+
+prec2_expr[arg_type]
+    : prec3_expr[arg_type]
+        (
+            (
+                  MUL { op = struct.pack('B', ptgMul) }
+                | DIV { op = struct.pack('B', ptgDiv) }
+            )
+            prec3_expr[arg_type] { self.rpn += op }
+        )*
+    ;
+
+prec3_expr[arg_type]
+    : prec4_expr[arg_type]
+        (
+            (
+                POWER { op = struct.pack('B', ptgPower) }
+            )
+            prec4_expr[arg_type] { self.rpn += op }
+        )*
+    ;
+
+prec4_expr[arg_type]
+    : prec5_expr[arg_type]
+        (
+            PERCENT { self.rpn += struct.pack('B', ptgPercent) }
+        )?
+    ;
+
+prec5_expr[arg_type]
+    : primary[arg_type]
+    | SUB primary[arg_type] { self.rpn += struct.pack('B', ptgUminus) }
+    ;
+
+primary[arg_type]
+    : TRUE_CONST
+        {
+            self.rpn += struct.pack("2B", ptgBool, 1)
+        }
+    | FALSE_CONST
+        {
+            self.rpn += struct.pack("2B", ptgBool, 0)
+        }
+    | str_tok:STR_CONST
+        {
+            self.rpn += struct.pack("B", ptgStr) + upack1(str_tok.text[1:-1].replace("\"\"", "\""))
+        }
+    | int_tok:INT_CONST
+        {
+            // print "**int_const", int_tok.text
+            int_value = int(int_tok.text)
+            if int_value <= 65535:
+                self.rpn += struct.pack("<BH", ptgInt, int_value)
+            else:
+                self.rpn += struct.pack("<Bd", ptgNum, float(int_value))
+        }
+    | num_tok:NUM_CONST
+        {
+            self.rpn += struct.pack("<Bd", ptgNum, float(num_tok.text))
+        }
+    | ref2d_tok:REF2D
+        {
+            // print "**ref2d %s %s" % (ref2d_tok.text, arg_type)
+            r, c = Utils.cell_to_packed_rowcol(ref2d_tok.text)
+            ptg = ptgRefR + _RVAdeltaRef[arg_type]
+            self.rpn += struct.pack("<B2H", ptg, r, c)
+        }
+    | ref2d1_tok:REF2D COLON ref2d2_tok:REF2D
+        {
+            r1, c1 = Utils.cell_to_packed_rowcol(ref2d1_tok.text)
+            r2, c2 = Utils.cell_to_packed_rowcol(ref2d2_tok.text)
+            ptg = ptgAreaR + _RVAdeltaArea[arg_type]
+            self.rpn += struct.pack("<B4H", ptg, r1, r2, c1, c2)
+        }
+    | sheet1 = sheet
+        { 
+            sheet2 = sheet1
+        }
+        ( COLON sheet2 = sheet )? BANG ref3d_ref2d: REF2D
+        {
+            ptg = ptgRef3dR + _RVAdeltaRef[arg_type]
+            rpn_ref2d = ""
+            r1, c1 = Utils.cell_to_packed_rowcol(ref3d_ref2d.text)
+            rpn_ref2d = struct.pack("<3H", 0x0000, r1, c1)
+        }
+        ( COLON ref3d_ref2d2: REF2D
+            {
+                ptg = ptgArea3dR + _RVAdeltaArea[arg_type]
+                r2, c2 = Utils.cell_to_packed_rowcol(ref3d_ref2d2.text)
+                rpn_ref2d = struct.pack("<5H", 0x0000, r1, r2, c1, c2)
+            }
+        )?
+        {
+            self.rpn += struct.pack("<B", ptg)
+            self.sheet_references.append((sheet1, sheet2, len(self.rpn)))
+            self.rpn += rpn_ref2d
+        }
+    | FUNC_IF
+        LP expr["V"] (SEMICOLON | COMMA)
+        {
+            self.rpn += struct.pack("<BBH", ptgAttr, 0x02, 0) // tAttrIf
+            pos0 = len(self.rpn) - 2
+        }
+        expr[arg_type] (SEMICOLON | COMMA)
+        {
+            self.rpn += struct.pack("<BBH", ptgAttr, 0x08, 0) // tAttrSkip
+            pos1 = len(self.rpn) - 2
+            self.rpn = self.rpn[:pos0] + struct.pack("<H", pos1-pos0) + self.rpn[pos0+2:]
+        }
+        expr[arg_type] RP
+        {
+            self.rpn += struct.pack("<BBH", ptgAttr, 0x08, 3) // tAttrSkip
+            self.rpn += struct.pack("<BBH", ptgFuncVarR, 3, 1) // 3 = nargs, 1 = IF func
+            pos2 = len(self.rpn)
+            self.rpn = self.rpn[:pos1] + struct.pack("<H", pos2-(pos1+2)-1) + self.rpn[pos1+2:]
+        }
+    | FUNC_CHOOSE
+        {
+            arg_type = "R"
+            rpn_chunks = []
+        }
+        LP expr["V"] // first argument (the selector)
+        {
+            rpn_start = len(self.rpn)
+            ref_markers = [len(self.sheet_references)]
+        }
+        (
+            (SEMICOLON | COMMA)
+                { mark = len(self.rpn) }
+                (
+                  expr[arg_type]
+                | { self.rpn += struct.pack("B", ptgMissArg) }
+                )
+                {
+                    rpn_chunks.append(self.rpn[mark:])
+                    ref_markers.append(len(self.sheet_references))
+                }
+        )*
+        RP
+        {
+            self.rpn = self.rpn[:rpn_start]
+            nc = len(rpn_chunks)
+            chunklens = [len(chunk) for chunk in rpn_chunks]
+            skiplens = [0] * nc
+            skiplens[-1] = 3
+            for ic in xrange(nc-1, 0, -1):
+                skiplens[ic-1] = skiplens[ic] + chunklens[ic] + 4
+            jump_pos = [2 * nc + 2]
+            for ic in xrange(nc):
+                jump_pos.append(jump_pos[-1] + chunklens[ic] + 4)
+            chunk_shift = 2 * nc + 6 // size of tAttrChoose
+            for ic in xrange(nc):
+                for refx in xrange(ref_markers[ic], ref_markers[ic+1]):
+                    ref = self.sheet_references[refx]
+                    self.sheet_references[refx] = (ref[0], ref[1], ref[2] + chunk_shift)
+                chunk_shift += 4 // size of tAttrSkip
+            choose_rpn = []
+            choose_rpn.append(struct.pack("<BBH", ptgAttr, 0x04, nc)) // 0x04 is tAttrChoose
+            choose_rpn.append(struct.pack("<%dH" % (nc+1), *jump_pos))
+            for ic in xrange(nc):
+                choose_rpn.append(rpn_chunks[ic])
+                choose_rpn.append(struct.pack("<BBH", ptgAttr, 0x08, skiplens[ic])) // 0x08 is tAttrSkip
+            choose_rpn.append(struct.pack("<BBH", ptgFuncVarV, nc+1, 100)) // 100 is CHOOSE fn
+            self.rpn += "".join(choose_rpn)
+        }
+    | name_tok:NAME
+        {
+            raise Exception("[formula] found unexpected NAME token (%r)" % name_tok.txt)
+            // #### TODO: handle references to defined names here
+        }
+    | func_tok:NAME
+        {
+            func_toku = func_tok.text.upper()
+            if func_toku in all_funcs_by_name:
+                (opcode,
+                min_argc,
+                max_argc,
+                func_type,
+                arg_type_str) = all_funcs_by_name[func_toku]
+                arg_type_list = list(arg_type_str)
+            else:
+                raise Exception("[formula] unknown function (%s)" % func_tok.text)
+            // print "**func_tok1 %s %s" % (func_toku, func_type)
+            xcall = opcode < 0
+            if xcall:
+                // The name of the add-in function is passed as the 1st arg
+                // of the hidden XCALL function
+                self.xcall_references.append((func_toku, len(self.rpn) + 1))
+                self.rpn += struct.pack("<BHHH",
+                    ptgNameXR,
+                    0xadde, // ##PATCHME## index to REF entry in EXTERNSHEET record
+                    0xefbe, // ##PATCHME## one-based index to EXTERNNAME record
+                    0x0000) // unused
+        }
+        LP arg_count = expr_list[arg_type_list, min_argc, max_argc] RP
+        {
+            if arg_count > max_argc or arg_count < min_argc:
+                raise Exception, "%d parameters for function: %s" % (arg_count, func_tok.text)
+            if xcall:
+                func_ptg = ptgFuncVarR + _RVAdelta[func_type]
+                self.rpn += struct.pack("<2BH", func_ptg, arg_count + 1, 255) // 255 is magic XCALL function
+            elif min_argc == max_argc:
+                func_ptg = ptgFuncR + _RVAdelta[func_type]
+                self.rpn += struct.pack("<BH", func_ptg, opcode)
+            elif arg_count == 1 and func_tok.text.upper() == "SUM":
+                self.rpn += struct.pack("<BBH", ptgAttr, 0x10, 0) // tAttrSum
+            else:
+                func_ptg = ptgFuncVarR + _RVAdelta[func_type]
+                self.rpn += struct.pack("<2BH", func_ptg, arg_count, opcode)
+        }
+    | LP expr[arg_type] RP
+        {
+            self.rpn += struct.pack("B", ptgParen)
+        }
+    ;
+
+expr_list[arg_type_list, min_argc, max_argc] returns [arg_cnt]
+    {
+        arg_cnt = 0
+        arg_type = arg_type_list[arg_cnt]
+        // print "**expr_list1[%d] req=%s" % (arg_cnt, arg_type)
+    }
+    : expr[arg_type] { arg_cnt += 1 }
+    (
+        {
+            if arg_cnt < len(arg_type_list):
+                arg_type = arg_type_list[arg_cnt]
+            else:
+                arg_type = arg_type_list[-1]
+            if arg_type == "+":
+                arg_type = arg_type_list[-2]
+            // print "**expr_list2[%d] req=%s" % (arg_cnt, arg_type)
+        }
+        (SEMICOLON | COMMA)
+            (
+                  expr[arg_type]
+                | { self.rpn += struct.pack("B", ptgMissArg) }
+            )
+            { arg_cnt += 1 }
+    )*
+    |
+    ;
+
+sheet returns[ref]
+    : sheet_ref_name: NAME
+    	{ ref = sheet_ref_name.text }
+    | sheet_ref_int: INT_CONST
+    	{ ref = sheet_ref_int.text }
+    | sheet_ref_quote: QUOTENAME
+    	{ ref = sheet_ref_quote.text[1:-1].replace("''", "'") }
+    ;

tablib/packages/yaml/__init__.py

@@ -1,12 +1,12 @@
 
-from .error import *
+from error import *
 
-from .tokens import *
-from .events import *
-from .nodes import *
+from tokens import *
+from events import *
+from nodes import *
 
-from .loader import *
-from .dumper import *
+from loader import *
+from dumper import *
 
 __version__ = '3.09'
 
@@ -149,9 +149,9 @@ def dump_all(documents, stream=None, Dumper=Dumper,
     getvalue = None
     if stream is None:
         if encoding is None:
-            from io import StringIO
+            from StringIO import StringIO
         else:
-            from io import StringIO
+            from cStringIO import StringIO
         stream = StringIO()
         getvalue = stream.getvalue
     dumper = Dumper(stream, default_style=default_style,

tablib/packages/yaml/composer.py

@@ -1,14 +1,14 @@
 
 __all__ = ['Composer', 'ComposerError']
 
-from .error import MarkedYAMLError
-from .events import *
-from .nodes import *
+from error import MarkedYAMLError
+from events import *
+from nodes import *
 
 class ComposerError(MarkedYAMLError):
     pass
 
-class Composer:
+class Composer(object):
 
     def __init__(self):
         self.anchors = {}
@@ -66,14 +66,14 @@ class Composer:
             anchor = event.anchor
             if anchor not in self.anchors:
                 raise ComposerError(None, None, "found undefined alias %r"
-                        % anchor, event.start_mark)
+                        % anchor.encode('utf-8'), event.start_mark)
             return self.anchors[anchor]
         event = self.peek_event()
         anchor = event.anchor
         if anchor is not None:
             if anchor in self.anchors:
                 raise ComposerError("found duplicate anchor %r; first occurence"
-                        % anchor, self.anchors[anchor].start_mark,
+                        % anchor.encode('utf-8'), self.anchors[anchor].start_mark,
                         "second occurence", event.start_mark)
         self.descend_resolver(parent, index)
         if self.check_event(ScalarEvent):
@@ -88,7 +88,7 @@ class Composer:
     def compose_scalar_node(self, anchor):
         event = self.get_event()
         tag = event.tag
-        if tag is None or tag == '!':
+        if tag is None or tag == u'!':
             tag = self.resolve(ScalarNode, event.value, event.implicit)
         node = ScalarNode(tag, event.value,
                 event.start_mark, event.end_mark, style=event.style)
@@ -99,7 +99,7 @@ class Composer:
     def compose_sequence_node(self, anchor):
         start_event = self.get_event()
         tag = start_event.tag
-        if tag is None or tag == '!':
+        if tag is None or tag == u'!':
             tag = self.resolve(SequenceNode, None, start_event.implicit)
         node = SequenceNode(tag, [],
                 start_event.start_mark, None,
@@ -117,7 +117,7 @@ class Composer:
     def compose_mapping_node(self, anchor):
         start_event = self.get_event()
         tag = start_event.tag
-        if tag is None or tag == '!':
+        if tag is None or tag == u'!':
             tag = self.resolve(MappingNode, None, start_event.implicit)
         node = MappingNode(tag, [],
                 start_event.start_mark, None,

tablib/packages/yaml/constructor.py

@@ -2,15 +2,22 @@
 __all__ = ['BaseConstructor', 'SafeConstructor', 'Constructor',
     'ConstructorError']
 
-from .error import *
-from .nodes import *
+from error import *
+from nodes import *
 
-import collections, datetime, base64, binascii, re, sys, types
+import datetime
+
+try:
+    set
+except NameError:
+    from sets import Set as set
+
+import binascii, re, sys, types
 
 class ConstructorError(MarkedYAMLError):
     pass
 
-class BaseConstructor:
+class BaseConstructor(object):
 
     yaml_constructors = {}
     yaml_multi_constructors = {}
@@ -88,7 +95,7 @@ class BaseConstructor:
             data = constructor(self, tag_suffix, node)
         if isinstance(data, types.GeneratorType):
             generator = data
-            data = next(generator)
+            data = generator.next()
             if self.deep_construct:
                 for dummy in generator:
                     pass
@@ -123,9 +130,11 @@ class BaseConstructor:
         mapping = {}
         for key_node, value_node in node.value:
             key = self.construct_object(key_node, deep=deep)
-            if not isinstance(key, collections.Hashable):
+            try:
+                hash(key)
+            except TypeError, exc:
                 raise ConstructorError("while constructing a mapping", node.start_mark,
-                        "found unhashable key", key_node.start_mark)
+                        "found unacceptable key (%s)" % exc, key_node.start_mark)
             value = self.construct_object(value_node, deep=deep)
             mapping[key] = value
         return mapping
@@ -142,33 +151,33 @@ class BaseConstructor:
             pairs.append((key, value))
         return pairs
 
-    @classmethod
     def add_constructor(cls, tag, constructor):
         if not 'yaml_constructors' in cls.__dict__:
             cls.yaml_constructors = cls.yaml_constructors.copy()
         cls.yaml_constructors[tag] = constructor
+    add_constructor = classmethod(add_constructor)
 
-    @classmethod
     def add_multi_constructor(cls, tag_prefix, multi_constructor):
         if not 'yaml_multi_constructors' in cls.__dict__:
             cls.yaml_multi_constructors = cls.yaml_multi_constructors.copy()
         cls.yaml_multi_constructors[tag_prefix] = multi_constructor
+    add_multi_constructor = classmethod(add_multi_constructor)
 
 class SafeConstructor(BaseConstructor):
 
     def construct_scalar(self, node):
         if isinstance(node, MappingNode):
             for key_node, value_node in node.value:
-                if key_node.tag == 'tag:yaml.org,2002:value':
+                if key_node.tag == u'tag:yaml.org,2002:value':
                     return self.construct_scalar(value_node)
-        return super().construct_scalar(node)
+        return BaseConstructor.construct_scalar(self, node)
 
     def flatten_mapping(self, node):
         merge = []
         index = 0
         while index < len(node.value):
             key_node, value_node = node.value[index]
-            if key_node.tag == 'tag:yaml.org,2002:merge':
+            if key_node.tag == u'tag:yaml.org,2002:merge':
                 del node.value[index]
                 if isinstance(value_node, MappingNode):
                     self.flatten_mapping(value_node)
@@ -190,8 +199,8 @@ class SafeConstructor(BaseConstructor):
                     raise ConstructorError("while constructing a mapping", node.start_mark,
                             "expected a mapping or list of mappings for merging, but found %s"
                             % value_node.id, value_node.start_mark)
-            elif key_node.tag == 'tag:yaml.org,2002:value':
-                key_node.tag = 'tag:yaml.org,2002:str'
+            elif key_node.tag == u'tag:yaml.org,2002:value':
+                key_node.tag = u'tag:yaml.org,2002:str'
                 index += 1
             else:
                 index += 1
@@ -201,19 +210,19 @@ class SafeConstructor(BaseConstructor):
     def construct_mapping(self, node, deep=False):
         if isinstance(node, MappingNode):
             self.flatten_mapping(node)
-        return super().construct_mapping(node, deep=deep)
+        return BaseConstructor.construct_mapping(self, node, deep=deep)
 
     def construct_yaml_null(self, node):
         self.construct_scalar(node)
         return None
 
     bool_values = {
-        'yes':      True,
-        'no':       False,
-        'true':     True,
-        'false':    False,
-        'on':       True,
-        'off':      False,
+        u'yes':     True,
+        u'no':      False,
+        u'true':    True,
+        u'false':   False,
+        u'on':      True,
+        u'off':     False,
     }
 
     def construct_yaml_bool(self, node):
@@ -221,7 +230,7 @@ class SafeConstructor(BaseConstructor):
         return self.bool_values[value.lower()]
 
     def construct_yaml_int(self, node):
-        value = self.construct_scalar(node)
+        value = str(self.construct_scalar(node))
         value = value.replace('_', '')
         sign = +1
         if value[0] == '-':
@@ -254,7 +263,7 @@ class SafeConstructor(BaseConstructor):
     nan_value = -inf_value/inf_value   # Trying to make a quiet NaN (like C99).
 
     def construct_yaml_float(self, node):
-        value = self.construct_scalar(node)
+        value = str(self.construct_scalar(node))
         value = value.replace('_', '').lower()
         sign = +1
         if value[0] == '-':
@@ -278,23 +287,15 @@ class SafeConstructor(BaseConstructor):
             return sign*float(value)
 
     def construct_yaml_binary(self, node):
+        value = self.construct_scalar(node)
         try:
-            value = self.construct_scalar(node).encode('ascii')
-        except UnicodeEncodeError as exc:
+            return str(value).decode('base64')
+        except (binascii.Error, UnicodeEncodeError), exc:
             raise ConstructorError(None, None,
-                    "failed to convert base64 data into ascii: %s" % exc,
-                    node.start_mark)
-        try:
-            if hasattr(base64, 'decodebytes'):
-                return base64.decodebytes(value)
-            else:
-                return base64.decodestring(value)
-        except binascii.Error as exc:
-            raise ConstructorError(None, None,
-                    "failed to decode base64 data: %s" % exc, node.start_mark)
+                    "failed to decode base64 data: %s" % exc, node.start_mark) 
 
     timestamp_regexp = re.compile(
-            r'''^(?P<year>[0-9][0-9][0-9][0-9])
+            ur'''^(?P<year>[0-9][0-9][0-9][0-9])
                 -(?P<month>[0-9][0-9]?)
                 -(?P<day>[0-9][0-9]?)
                 (?:(?:[Tt]|[ \t]+)
@@ -385,7 +386,11 @@ class SafeConstructor(BaseConstructor):
         data.update(value)
 
     def construct_yaml_str(self, node):
-        return self.construct_scalar(node)
+        value = self.construct_scalar(node)
+        try:
+            return value.encode('ascii')
+        except UnicodeEncodeError:
+            return value
 
     def construct_yaml_seq(self, node):
         data = []
@@ -410,55 +415,55 @@ class SafeConstructor(BaseConstructor):
 
     def construct_undefined(self, node):
         raise ConstructorError(None, None,
-                "could not determine a constructor for the tag %r" % node.tag,
+                "could not determine a constructor for the tag %r" % node.tag.encode('utf-8'),
                 node.start_mark)
 
 SafeConstructor.add_constructor(
-        'tag:yaml.org,2002:null',
+        u'tag:yaml.org,2002:null',
         SafeConstructor.construct_yaml_null)
 
 SafeConstructor.add_constructor(
-        'tag:yaml.org,2002:bool',
+        u'tag:yaml.org,2002:bool',
         SafeConstructor.construct_yaml_bool)
 
 SafeConstructor.add_constructor(
-        'tag:yaml.org,2002:int',
+        u'tag:yaml.org,2002:int',
         SafeConstructor.construct_yaml_int)
 
 SafeConstructor.add_constructor(
-        'tag:yaml.org,2002:float',
+        u'tag:yaml.org,2002:float',
         SafeConstructor.construct_yaml_float)
 
 SafeConstructor.add_constructor(
-        'tag:yaml.org,2002:binary',
+        u'tag:yaml.org,2002:binary',
         SafeConstructor.construct_yaml_binary)
 
 SafeConstructor.add_constructor(
-        'tag:yaml.org,2002:timestamp',
+        u'tag:yaml.org,2002:timestamp',
         SafeConstructor.construct_yaml_timestamp)
 
 SafeConstructor.add_constructor(
-        'tag:yaml.org,2002:omap',
+        u'tag:yaml.org,2002:omap',
         SafeConstructor.construct_yaml_omap)
 
 SafeConstructor.add_constructor(
-        'tag:yaml.org,2002:pairs',
+        u'tag:yaml.org,2002:pairs',
         SafeConstructor.construct_yaml_pairs)
 
 SafeConstructor.add_constructor(
-        'tag:yaml.org,2002:set',
+        u'tag:yaml.org,2002:set',
         SafeConstructor.construct_yaml_set)
 
 SafeConstructor.add_constructor(
-        'tag:yaml.org,2002:str',
+        u'tag:yaml.org,2002:str',
         SafeConstructor.construct_yaml_str)
 
 SafeConstructor.add_constructor(
-        'tag:yaml.org,2002:seq',
+        u'tag:yaml.org,2002:seq',
         SafeConstructor.construct_yaml_seq)
 
 SafeConstructor.add_constructor(
-        'tag:yaml.org,2002:map',
+        u'tag:yaml.org,2002:map',
         SafeConstructor.construct_yaml_map)
 
 SafeConstructor.add_constructor(None,
@@ -467,29 +472,13 @@ SafeConstructor.add_constructor(None,
 class Constructor(SafeConstructor):
 
     def construct_python_str(self, node):
-        return self.construct_scalar(node)
+        return self.construct_scalar(node).encode('utf-8')
 
     def construct_python_unicode(self, node):
         return self.construct_scalar(node)
 
-    def construct_python_bytes(self, node):
-        try:
-            value = self.construct_scalar(node).encode('ascii')
-        except UnicodeEncodeError as exc:
-            raise ConstructorError(None, None,
-                    "failed to convert base64 data into ascii: %s" % exc,
-                    node.start_mark)
-        try:
-            if hasattr(base64, 'decodebytes'):
-                return base64.decodebytes(value)
-            else:
-                return base64.decodestring(value)
-        except binascii.Error as exc:
-            raise ConstructorError(None, None,
-                    "failed to decode base64 data: %s" % exc, node.start_mark)
-
     def construct_python_long(self, node):
-        return self.construct_yaml_int(node)
+        return long(self.construct_yaml_int(node))
 
     def construct_python_complex(self, node):
        return complex(self.construct_scalar(node))
@@ -503,46 +492,54 @@ class Constructor(SafeConstructor):
                     "expected non-empty name appended to the tag", mark)
         try:
             __import__(name)
-        except ImportError as exc:
+        except ImportError, exc:
             raise ConstructorError("while constructing a Python module", mark,
-                    "cannot find module %r (%s)" % (name, exc), mark)
+                    "cannot find module %r (%s)" % (name.encode('utf-8'), exc), mark)
         return sys.modules[name]
 
     def find_python_name(self, name, mark):
         if not name:
             raise ConstructorError("while constructing a Python object", mark,
                     "expected non-empty name appended to the tag", mark)
-        if '.' in name:
-            module_name, object_name = name.rsplit('.', 1)
+        if u'.' in name:
+            # Python 2.4 only
+            #module_name, object_name = name.rsplit('.', 1)
+            items = name.split('.')
+            object_name = items.pop()
+            module_name = '.'.join(items)
         else:
-            module_name = 'builtins'
+            module_name = '__builtin__'
             object_name = name
         try:
             __import__(module_name)
-        except ImportError as exc:
+        except ImportError, exc:
             raise ConstructorError("while constructing a Python object", mark,
-                    "cannot find module %r (%s)" % (module_name, exc), mark)
+                    "cannot find module %r (%s)" % (module_name.encode('utf-8'), exc), mark)
         module = sys.modules[module_name]
         if not hasattr(module, object_name):
             raise ConstructorError("while constructing a Python object", mark,
-                    "cannot find %r in the module %r"
-                    % (object_name, module.__name__), mark)
+                    "cannot find %r in the module %r" % (object_name.encode('utf-8'),
+                        module.__name__), mark)
         return getattr(module, object_name)
 
     def construct_python_name(self, suffix, node):
         value = self.construct_scalar(node)
         if value:
             raise ConstructorError("while constructing a Python name", node.start_mark,
-                    "expected the empty value, but found %r" % value, node.start_mark)
+                    "expected the empty value, but found %r" % value.encode('utf-8'),
+                    node.start_mark)
         return self.find_python_name(suffix, node.start_mark)
 
     def construct_python_module(self, suffix, node):
         value = self.construct_scalar(node)
         if value:
             raise ConstructorError("while constructing a Python module", node.start_mark,
-                    "expected the empty value, but found %r" % value, node.start_mark)
+                    "expected the empty value, but found %r" % value.encode('utf-8'),
+                    node.start_mark)
         return self.find_python_module(suffix, node.start_mark)
 
+    class classobj: pass
+
     def make_python_instance(self, suffix, node,
             args=None, kwds=None, newobj=False):
         if not args:
@@ -550,7 +547,12 @@ class Constructor(SafeConstructor):
         if not kwds:
             kwds = {}
         cls = self.find_python_name(suffix, node.start_mark)
-        if newobj and isinstance(cls, type):
+        if newobj and isinstance(cls, type(self.classobj))  \
+                and not args and not kwds:
+            instance = self.classobj()
+            instance.__class__ = cls
+            return instance
+        elif newobj and isinstance(cls, type):
             return cls.__new__(cls, *args, **kwds)
         else:
             return cls(*args, **kwds)
@@ -617,70 +619,66 @@ class Constructor(SafeConstructor):
         return self.construct_python_object_apply(suffix, node, newobj=True)
 
 Constructor.add_constructor(
-    'tag:yaml.org,2002:python/none',
+    u'tag:yaml.org,2002:python/none',
     Constructor.construct_yaml_null)
 
 Constructor.add_constructor(
-    'tag:yaml.org,2002:python/bool',
+    u'tag:yaml.org,2002:python/bool',
     Constructor.construct_yaml_bool)
 
 Constructor.add_constructor(
-    'tag:yaml.org,2002:python/str',
+    u'tag:yaml.org,2002:python/str',
     Constructor.construct_python_str)
 
 Constructor.add_constructor(
-    'tag:yaml.org,2002:python/unicode',
+    u'tag:yaml.org,2002:python/unicode',
     Constructor.construct_python_unicode)
 
 Constructor.add_constructor(
-    'tag:yaml.org,2002:python/bytes',
-    Constructor.construct_python_bytes)
-
-Constructor.add_constructor(
-    'tag:yaml.org,2002:python/int',
+    u'tag:yaml.org,2002:python/int',
     Constructor.construct_yaml_int)
 
 Constructor.add_constructor(
-    'tag:yaml.org,2002:python/long',
+    u'tag:yaml.org,2002:python/long',
     Constructor.construct_python_long)
 
 Constructor.add_constructor(
-    'tag:yaml.org,2002:python/float',
+    u'tag:yaml.org,2002:python/float',
     Constructor.construct_yaml_float)
 
 Constructor.add_constructor(
-    'tag:yaml.org,2002:python/complex',
+    u'tag:yaml.org,2002:python/complex',
     Constructor.construct_python_complex)
 
 Constructor.add_constructor(
-    'tag:yaml.org,2002:python/list',
+    u'tag:yaml.org,2002:python/list',
     Constructor.construct_yaml_seq)
 
 Constructor.add_constructor(
-    'tag:yaml.org,2002:python/tuple',
+    u'tag:yaml.org,2002:python/tuple',
     Constructor.construct_python_tuple)
 
 Constructor.add_constructor(
-    'tag:yaml.org,2002:python/dict',
+    u'tag:yaml.org,2002:python/dict',
     Constructor.construct_yaml_map)
 
 Constructor.add_multi_constructor(
-    'tag:yaml.org,2002:python/name:',
+    u'tag:yaml.org,2002:python/name:',
     Constructor.construct_python_name)
 
 Constructor.add_multi_constructor(
-    'tag:yaml.org,2002:python/module:',
+    u'tag:yaml.org,2002:python/module:',
     Constructor.construct_python_module)
 
 Constructor.add_multi_constructor(
-    'tag:yaml.org,2002:python/object:',
+    u'tag:yaml.org,2002:python/object:',
     Constructor.construct_python_object)
 
 Constructor.add_multi_constructor(
-    'tag:yaml.org,2002:python/object/apply:',
+    u'tag:yaml.org,2002:python/object/apply:',
     Constructor.construct_python_object_apply)
 
 Constructor.add_multi_constructor(
-    'tag:yaml.org,2002:python/object/new:',
+    u'tag:yaml.org,2002:python/object/new:',
     Constructor.construct_python_object_new)
 

tablib/packages/yaml/cyaml.py

@@ -4,12 +4,12 @@ __all__ = ['CBaseLoader', 'CSafeLoader', 'CLoader',
 
 from _yaml import CParser, CEmitter
 
-from .constructor import *
+from constructor import *
 
-from .serializer import *
-from .representer import *
+from serializer import *
+from representer import *
 
-from .resolver import *
+from resolver import *
 
 class CBaseLoader(CParser, BaseConstructor, BaseResolver):
 

tablib/packages/yaml/dumper.py

@@ -1,10 +1,10 @@
 
 __all__ = ['BaseDumper', 'SafeDumper', 'Dumper']
 
-from .emitter import *
-from .serializer import *
-from .representer import *
-from .resolver import *
+from emitter import *
+from serializer import *
+from representer import *
+from resolver import *
 
 class BaseDumper(Emitter, Serializer, BaseRepresenter, BaseResolver):
 

tablib/packages/yaml/emitter.py

@@ -8,13 +8,13 @@
 
 __all__ = ['Emitter', 'EmitterError']
 
-from .error import YAMLError
-from .events import *
+from error import YAMLError
+from events import *
 
 class EmitterError(YAMLError):
     pass
 
-class ScalarAnalysis:
+class ScalarAnalysis(object):
     def __init__(self, scalar, empty, multiline,
             allow_flow_plain, allow_block_plain,
             allow_single_quoted, allow_double_quoted,
@@ -28,11 +28,11 @@ class ScalarAnalysis:
         self.allow_double_quoted = allow_double_quoted
         self.allow_block = allow_block
 
-class Emitter:
+class Emitter(object):
 
     DEFAULT_TAG_PREFIXES = {
-        '!' : '!',
-        'tag:yaml.org,2002:' : '!!',
+        u'!' : u'!',
+        u'tag:yaml.org,2002:' : u'!!',
     }
 
     def __init__(self, stream, canonical=None, indent=None, width=None,
@@ -88,8 +88,8 @@ class Emitter:
         self.best_width = 80
         if width and width > self.best_indent*2:
             self.best_width = width
-        self.best_line_break = '\n'
-        if line_break in ['\r', '\n', '\r\n']:
+        self.best_line_break = u'\n'
+        if line_break in [u'\r', u'\n', u'\r\n']:
             self.best_line_break = line_break
 
         # Tag prefixes.
@@ -154,7 +154,7 @@ class Emitter:
 
     def expect_stream_start(self):
         if isinstance(self.event, StreamStartEvent):
-            if self.event.encoding and not hasattr(self.stream, 'encoding'):
+            if self.event.encoding and not getattr(self.stream, 'encoding', None):
                 self.encoding = self.event.encoding
             self.write_stream_start()
             self.state = self.expect_first_document_start
@@ -173,14 +173,15 @@ class Emitter:
     def expect_document_start(self, first=False):
         if isinstance(self.event, DocumentStartEvent):
             if (self.event.version or self.event.tags) and self.open_ended:
-                self.write_indicator('...', True)
+                self.write_indicator(u'...', True)
                 self.write_indent()
             if self.event.version:
                 version_text = self.prepare_version(self.event.version)
                 self.write_version_directive(version_text)
             self.tag_prefixes = self.DEFAULT_TAG_PREFIXES.copy()
             if self.event.tags:
-                handles = sorted(self.event.tags.keys())
+                handles = self.event.tags.keys()
+                handles.sort()
                 for handle in handles:
                     prefix = self.event.tags[handle]
                     self.tag_prefixes[prefix] = handle
@@ -192,13 +193,13 @@ class Emitter:
                     and not self.check_empty_document())
             if not implicit:
                 self.write_indent()
-                self.write_indicator('---', True)
+                self.write_indicator(u'---', True)
                 if self.canonical:
                     self.write_indent()
             self.state = self.expect_document_root
         elif isinstance(self.event, StreamEndEvent):
             if self.open_ended:
-                self.write_indicator('...', True)
+                self.write_indicator(u'...', True)
                 self.write_indent()
             self.write_stream_end()
             self.state = self.expect_nothing
@@ -210,7 +211,7 @@ class Emitter:
         if isinstance(self.event, DocumentEndEvent):
             self.write_indent()
             if self.event.explicit:
-                self.write_indicator('...', True)
+                self.write_indicator(u'...', True)
                 self.write_indent()
             self.flush_stream()
             self.state = self.expect_document_start
@@ -233,7 +234,7 @@ class Emitter:
         if isinstance(self.event, AliasEvent):
             self.expect_alias()
         elif isinstance(self.event, (ScalarEvent, CollectionStartEvent)):
-            self.process_anchor('&')
+            self.process_anchor(u'&')
             self.process_tag()
             if isinstance(self.event, ScalarEvent):
                 self.expect_scalar()
@@ -255,7 +256,7 @@ class Emitter:
     def expect_alias(self):
         if self.event.anchor is None:
             raise EmitterError("anchor is not specified for alias")
-        self.process_anchor('*')
+        self.process_anchor(u'*')
         self.state = self.states.pop()
 
     def expect_scalar(self):
@@ -267,7 +268,7 @@ class Emitter:
     # Flow sequence handlers.
 
     def expect_flow_sequence(self):
-        self.write_indicator('[', True, whitespace=True)
+        self.write_indicator(u'[', True, whitespace=True)
         self.flow_level += 1
         self.increase_indent(flow=True)
         self.state = self.expect_first_flow_sequence_item
@@ -276,7 +277,7 @@ class Emitter:
         if isinstance(self.event, SequenceEndEvent):
             self.indent = self.indents.pop()
             self.flow_level -= 1
-            self.write_indicator(']', False)
+            self.write_indicator(u']', False)
             self.state = self.states.pop()
         else:
             if self.canonical or self.column > self.best_width:
@@ -289,12 +290,12 @@ class Emitter:
             self.indent = self.indents.pop()
             self.flow_level -= 1
             if self.canonical:
-                self.write_indicator(',', False)
+                self.write_indicator(u',', False)
                 self.write_indent()
-            self.write_indicator(']', False)
+            self.write_indicator(u']', False)
             self.state = self.states.pop()
         else:
-            self.write_indicator(',', False)
+            self.write_indicator(u',', False)
             if self.canonical or self.column > self.best_width:
                 self.write_indent()
             self.states.append(self.expect_flow_sequence_item)
@@ -303,7 +304,7 @@ class Emitter:
     # Flow mapping handlers.
 
     def expect_flow_mapping(self):
-        self.write_indicator('{', True, whitespace=True)
+        self.write_indicator(u'{', True, whitespace=True)
         self.flow_level += 1
         self.increase_indent(flow=True)
         self.state = self.expect_first_flow_mapping_key
@@ -312,7 +313,7 @@ class Emitter:
         if isinstance(self.event, MappingEndEvent):
             self.indent = self.indents.pop()
             self.flow_level -= 1
-            self.write_indicator('}', False)
+            self.write_indicator(u'}', False)
             self.state = self.states.pop()
         else:
             if self.canonical or self.column > self.best_width:
@@ -321,7 +322,7 @@ class Emitter:
                 self.states.append(self.expect_flow_mapping_simple_value)
                 self.expect_node(mapping=True, simple_key=True)
             else:
-                self.write_indicator('?', True)
+                self.write_indicator(u'?', True)
                 self.states.append(self.expect_flow_mapping_value)
                 self.expect_node(mapping=True)
 
@@ -330,31 +331,31 @@ class Emitter:
             self.indent = self.indents.pop()
             self.flow_level -= 1
             if self.canonical:
-                self.write_indicator(',', False)
+                self.write_indicator(u',', False)
                 self.write_indent()
-            self.write_indicator('}', False)
+            self.write_indicator(u'}', False)
             self.state = self.states.pop()
         else:
-            self.write_indicator(',', False)
+            self.write_indicator(u',', False)
             if self.canonical or self.column > self.best_width:
                 self.write_indent()
             if not self.canonical and self.check_simple_key():
                 self.states.append(self.expect_flow_mapping_simple_value)
                 self.expect_node(mapping=True, simple_key=True)
             else:
-                self.write_indicator('?', True)
+                self.write_indicator(u'?', True)
                 self.states.append(self.expect_flow_mapping_value)
                 self.expect_node(mapping=True)
 
     def expect_flow_mapping_simple_value(self):
-        self.write_indicator(':', False)
+        self.write_indicator(u':', False)
         self.states.append(self.expect_flow_mapping_key)
         self.expect_node(mapping=True)
 
     def expect_flow_mapping_value(self):
         if self.canonical or self.column > self.best_width:
             self.write_indent()
-        self.write_indicator(':', True)
+        self.write_indicator(u':', True)
         self.states.append(self.expect_flow_mapping_key)
         self.expect_node(mapping=True)
 
@@ -374,7 +375,7 @@ class Emitter:
             self.state = self.states.pop()
         else:
             self.write_indent()
-            self.write_indicator('-', True, indention=True)
+            self.write_indicator(u'-', True, indention=True)
             self.states.append(self.expect_block_sequence_item)
             self.expect_node(sequence=True)
 
@@ -397,18 +398,18 @@ class Emitter:
                 self.states.append(self.expect_block_mapping_simple_value)
                 self.expect_node(mapping=True, simple_key=True)
             else:
-                self.write_indicator('?', True, indention=True)
+                self.write_indicator(u'?', True, indention=True)
                 self.states.append(self.expect_block_mapping_value)
                 self.expect_node(mapping=True)
 
     def expect_block_mapping_simple_value(self):
-        self.write_indicator(':', False)
+        self.write_indicator(u':', False)
         self.states.append(self.expect_block_mapping_key)
         self.expect_node(mapping=True)
 
     def expect_block_mapping_value(self):
         self.write_indent()
-        self.write_indicator(':', True, indention=True)
+        self.write_indicator(u':', True, indention=True)
         self.states.append(self.expect_block_mapping_key)
         self.expect_node(mapping=True)
 
@@ -427,7 +428,7 @@ class Emitter:
             return False
         event = self.events[0]
         return (isinstance(event, ScalarEvent) and event.anchor is None
-                and event.tag is None and event.implicit and event.value == '')
+                and event.tag is None and event.implicit and event.value == u'')
 
     def check_simple_key(self):
         length = 0
@@ -472,7 +473,7 @@ class Emitter:
                 self.prepared_tag = None
                 return
             if self.event.implicit[0] and tag is None:
-                tag = '!'
+                tag = u'!'
                 self.prepared_tag = None
         else:
             if (not self.canonical or tag is None) and self.event.implicit:
@@ -535,18 +536,19 @@ class Emitter:
         major, minor = version
         if major != 1:
             raise EmitterError("unsupported YAML version: %d.%d" % (major, minor))
-        return '%d.%d' % (major, minor)
+        return u'%d.%d' % (major, minor)
 
     def prepare_tag_handle(self, handle):
         if not handle:
             raise EmitterError("tag handle must not be empty")
-        if handle[0] != '!' or handle[-1] != '!':
-            raise EmitterError("tag handle must start and end with '!': %r" % handle)
+        if handle[0] != u'!' or handle[-1] != u'!':
+            raise EmitterError("tag handle must start and end with '!': %r"
+                    % (handle.encode('utf-8')))
         for ch in handle[1:-1]:
-            if not ('0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z'    \
-                    or ch in '-_'):
+            if not (u'0' <= ch <= u'9' or u'A' <= ch <= u'Z' or u'a' <= ch <= u'z'  \
+                    or ch in u'-_'):
                 raise EmitterError("invalid character %r in the tag handle: %r"
-                        % (ch, handle))
+                        % (ch.encode('utf-8'), handle.encode('utf-8')))
         return handle
 
     def prepare_tag_prefix(self, prefix):
@@ -554,12 +556,12 @@ class Emitter:
             raise EmitterError("tag prefix must not be empty")
         chunks = []
         start = end = 0
-        if prefix[0] == '!':
+        if prefix[0] == u'!':
             end = 1
         while end < len(prefix):
             ch = prefix[end]
-            if '0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z' \
-                    or ch in '-;/?!:@&=+$,_.~*\'()[]':
+            if u'0' <= ch <= u'9' or u'A' <= ch <= u'Z' or u'a' <= ch <= u'z'   \
+                    or ch in u'-;/?!:@&=+$,_.~*\'()[]':
                 end += 1
             else:
                 if start < end:
@@ -567,31 +569,32 @@ class Emitter:
                 start = end = end+1
                 data = ch.encode('utf-8')
                 for ch in data:
-                    chunks.append('%%%02X' % ord(ch))
+                    chunks.append(u'%%%02X' % ord(ch))
         if start < end:
             chunks.append(prefix[start:end])
-        return ''.join(chunks)
+        return u''.join(chunks)
 
     def prepare_tag(self, tag):
         if not tag:
             raise EmitterError("tag must not be empty")
-        if tag == '!':
+        if tag == u'!':
             return tag
         handle = None
         suffix = tag
-        prefixes = sorted(self.tag_prefixes.keys())
+        prefixes = self.tag_prefixes.keys()
+        prefixes.sort()
         for prefix in prefixes:
             if tag.startswith(prefix)   \
-                    and (prefix == '!' or len(prefix) < len(tag)):
+                    and (prefix == u'!' or len(prefix) < len(tag)):
                 handle = self.tag_prefixes[prefix]
                 suffix = tag[len(prefix):]
         chunks = []
         start = end = 0
         while end < len(suffix):
             ch = suffix[end]
-            if '0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z' \
-                    or ch in '-;/?:@&=+$,_.~*\'()[]'   \
-                    or (ch == '!' and handle != '!'):
+            if u'0' <= ch <= u'9' or u'A' <= ch <= u'Z' or u'a' <= ch <= u'z'   \
+                    or ch in u'-;/?:@&=+$,_.~*\'()[]'   \
+                    or (ch == u'!' and handle != u'!'):
                 end += 1
             else:
                 if start < end:
@@ -599,23 +602,23 @@ class Emitter:
                 start = end = end+1
                 data = ch.encode('utf-8')
                 for ch in data:
-                    chunks.append('%%%02X' % ord(ch))
+                    chunks.append(u'%%%02X' % ord(ch))
         if start < end:
             chunks.append(suffix[start:end])
-        suffix_text = ''.join(chunks)
+        suffix_text = u''.join(chunks)
         if handle:
-            return '%s%s' % (handle, suffix_text)
+            return u'%s%s' % (handle, suffix_text)
         else:
-            return '!<%s>' % suffix_text
+            return u'!<%s>' % suffix_text
 
     def prepare_anchor(self, anchor):
         if not anchor:
             raise EmitterError("anchor must not be empty")
         for ch in anchor:
-            if not ('0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z'    \
-                    or ch in '-_'):
+            if not (u'0' <= ch <= u'9' or u'A' <= ch <= u'Z' or u'a' <= ch <= u'z'  \
+                    or ch in u'-_'):
                 raise EmitterError("invalid character %r in the anchor: %r"
-                        % (ch, anchor))
+                        % (ch.encode('utf-8'), anchor.encode('utf-8')))
         return anchor
 
     def analyze_scalar(self, scalar):
@@ -642,7 +645,7 @@ class Emitter:
         space_break = False
 
         # Check document indicators.
-        if scalar.startswith('---') or scalar.startswith('...'):
+        if scalar.startswith(u'---') or scalar.startswith(u'...'):
             block_indicators = True
             flow_indicators = True
 
@@ -651,7 +654,7 @@ class Emitter:
 
         # Last character or followed by a whitespace.
         followed_by_whitespace = (len(scalar) == 1 or
-                scalar[1] in '\0 \t\r\n\x85\u2028\u2029')
+                scalar[1] in u'\0 \t\r\n\x85\u2028\u2029')
 
         # The previous character is a space.
         previous_space = False
@@ -666,34 +669,34 @@ class Emitter:
             # Check for indicators.
             if index == 0:
                 # Leading indicators are special characters.
-                if ch in '#,[]{}&*!|>\'\"%@`': 
+                if ch in u'#,[]{}&*!|>\'\"%@`': 
                     flow_indicators = True
                     block_indicators = True
-                if ch in '?:':
+                if ch in u'?:':
                     flow_indicators = True
                     if followed_by_whitespace:
                         block_indicators = True
-                if ch == '-' and followed_by_whitespace:
+                if ch == u'-' and followed_by_whitespace:
                     flow_indicators = True
                     block_indicators = True
             else:
                 # Some indicators cannot appear within a scalar as well.
-                if ch in ',?[]{}':
+                if ch in u',?[]{}':
                     flow_indicators = True
-                if ch == ':':
+                if ch == u':':
                     flow_indicators = True
                     if followed_by_whitespace:
                         block_indicators = True
-                if ch == '#' and preceeded_by_whitespace:
+                if ch == u'#' and preceeded_by_whitespace:
                     flow_indicators = True
                     block_indicators = True
 
             # Check for line breaks, special, and unicode characters.
-            if ch in '\n\x85\u2028\u2029':
+            if ch in u'\n\x85\u2028\u2029':
                 line_breaks = True
-            if not (ch == '\n' or '\x20' <= ch <= '\x7E'):
-                if (ch == '\x85' or '\xA0' <= ch <= '\uD7FF'
-                        or '\uE000' <= ch <= '\uFFFD') and ch != '\uFEFF':
+            if not (ch == u'\n' or u'\x20' <= ch <= u'\x7E'):
+                if (ch == u'\x85' or u'\xA0' <= ch <= u'\uD7FF'
+                        or u'\uE000' <= ch <= u'\uFFFD') and ch != u'\uFEFF':
                     unicode_characters = True
                     if not self.allow_unicode:
                         special_characters = True
@@ -701,7 +704,7 @@ class Emitter:
                     special_characters = True
 
             # Detect important whitespace combinations.
-            if ch == ' ':
+            if ch == u' ':
                 if index == 0:
                     leading_space = True
                 if index == len(scalar)-1:
@@ -710,7 +713,7 @@ class Emitter:
                     break_space = True
                 previous_space = True
                 previous_break = False
-            elif ch in '\n\x85\u2028\u2029':
+            elif ch in u'\n\x85\u2028\u2029':
                 if index == 0:
                     leading_break = True
                 if index == len(scalar)-1:
@@ -725,9 +728,9 @@ class Emitter:
 
             # Prepare for the next character.
             index += 1
-            preceeded_by_whitespace = (ch in '\0 \t\r\n\x85\u2028\u2029')
+            preceeded_by_whitespace = (ch in u'\0 \t\r\n\x85\u2028\u2029')
             followed_by_whitespace = (index+1 >= len(scalar) or
-                    scalar[index+1] in '\0 \t\r\n\x85\u2028\u2029')
+                    scalar[index+1] in u'\0 \t\r\n\x85\u2028\u2029')
 
         # Let's decide what styles are allowed.
         allow_flow_plain = True
@@ -786,7 +789,7 @@ class Emitter:
     def write_stream_start(self):
         # Write BOM if needed.
         if self.encoding and self.encoding.startswith('utf-16'):
-            self.stream.write('\uFEFF'.encode(self.encoding))
+            self.stream.write(u'\uFEFF'.encode(self.encoding))
 
     def write_stream_end(self):
         self.flush_stream()
@@ -796,7 +799,7 @@ class Emitter:
         if self.whitespace or not need_whitespace:
             data = indicator
         else:
-            data = ' '+indicator
+            data = u' '+indicator
         self.whitespace = whitespace
         self.indention = self.indention and indention
         self.column += len(data)
@@ -812,7 +815,7 @@ class Emitter:
             self.write_line_break()
         if self.column < indent:
             self.whitespace = True
-            data = ' '*(indent-self.column)
+            data = u' '*(indent-self.column)
             self.column = indent
             if self.encoding:
                 data = data.encode(self.encoding)
@@ -830,14 +833,14 @@ class Emitter:
         self.stream.write(data)
 
     def write_version_directive(self, version_text):
-        data = '%%YAML %s' % version_text
+        data = u'%%YAML %s' % version_text
         if self.encoding:
             data = data.encode(self.encoding)
         self.stream.write(data)
         self.write_line_break()
 
     def write_tag_directive(self, handle_text, prefix_text):
-        data = '%%TAG %s %s' % (handle_text, prefix_text)
+        data = u'%%TAG %s %s' % (handle_text, prefix_text)
         if self.encoding:
             data = data.encode(self.encoding)
         self.stream.write(data)
@@ -846,7 +849,7 @@ class Emitter:
     # Scalar streams.
 
     def write_single_quoted(self, text, split=True):
-        self.write_indicator('\'', True)
+        self.write_indicator(u'\'', True)
         spaces = False
         breaks = False
         start = end = 0
@@ -855,7 +858,7 @@ class Emitter:
             if end < len(text):
                 ch = text[end]
             if spaces:
-                if ch is None or ch != ' ':
+                if ch is None or ch != u' ':
                     if start+1 == end and self.column > self.best_width and split   \
                             and start != 0 and end != len(text):
                         self.write_indent()
@@ -867,18 +870,18 @@ class Emitter:
                         self.stream.write(data)
                     start = end
             elif breaks:
-                if ch is None or ch not in '\n\x85\u2028\u2029':
-                    if text[start] == '\n':
+                if ch is None or ch not in u'\n\x85\u2028\u2029':
+                    if text[start] == u'\n':
                         self.write_line_break()
                     for br in text[start:end]:
-                        if br == '\n':
+                        if br == u'\n':
                             self.write_line_break()
                         else:
                             self.write_line_break(br)
                     self.write_indent()
                     start = end
             else:
-                if ch is None or ch in ' \n\x85\u2028\u2029' or ch == '\'':
+                if ch is None or ch in u' \n\x85\u2028\u2029' or ch == u'\'':
                     if start < end:
                         data = text[start:end]
                         self.column += len(data)
@@ -886,49 +889,49 @@ class Emitter:
                             data = data.encode(self.encoding)
                         self.stream.write(data)
                         start = end
-            if ch == '\'':
-                data = '\'\''
+            if ch == u'\'':
+                data = u'\'\''
                 self.column += 2
                 if self.encoding:
                     data = data.encode(self.encoding)
                 self.stream.write(data)
                 start = end + 1
             if ch is not None:
-                spaces = (ch == ' ')
-                breaks = (ch in '\n\x85\u2028\u2029')
+                spaces = (ch == u' ')
+                breaks = (ch in u'\n\x85\u2028\u2029')
             end += 1
-        self.write_indicator('\'', False)
+        self.write_indicator(u'\'', False)
 
     ESCAPE_REPLACEMENTS = {
-        '\0':       '0',
-        '\x07':     'a',
-        '\x08':     'b',
-        '\x09':     't',
-        '\x0A':     'n',
-        '\x0B':     'v',
-        '\x0C':     'f',
-        '\x0D':     'r',
-        '\x1B':     'e',
-        '\"':       '\"',
-        '\\':       '\\',
-        '\x85':     'N',
-        '\xA0':     '_',
-        '\u2028':   'L',
-        '\u2029':   'P',
+        u'\0':      u'0',
+        u'\x07':    u'a',
+        u'\x08':    u'b',
+        u'\x09':    u't',
+        u'\x0A':    u'n',
+        u'\x0B':    u'v',
+        u'\x0C':    u'f',
+        u'\x0D':    u'r',
+        u'\x1B':    u'e',
+        u'\"':      u'\"',
+        u'\\':      u'\\',
+        u'\x85':    u'N',
+        u'\xA0':    u'_',
+        u'\u2028':  u'L',
+        u'\u2029':  u'P',
     }
 
     def write_double_quoted(self, text, split=True):
-        self.write_indicator('"', True)
+        self.write_indicator(u'"', True)
         start = end = 0
         while end <= len(text):
             ch = None
             if end < len(text):
                 ch = text[end]
-            if ch is None or ch in '"\\\x85\u2028\u2029\uFEFF' \
-                    or not ('\x20' <= ch <= '\x7E'
+            if ch is None or ch in u'"\\\x85\u2028\u2029\uFEFF' \
+                    or not (u'\x20' <= ch <= u'\x7E'
                         or (self.allow_unicode
-                            and ('\xA0' <= ch <= '\uD7FF'
-                                or '\uE000' <= ch <= '\uFFFD'))):
+                            and (u'\xA0' <= ch <= u'\uD7FF'
+                                or u'\uE000' <= ch <= u'\uFFFD'))):
                 if start < end:
                     data = text[start:end]
                     self.column += len(data)
@@ -938,21 +941,21 @@ class Emitter:
                     start = end
                 if ch is not None:
                     if ch in self.ESCAPE_REPLACEMENTS:
-                        data = '\\'+self.ESCAPE_REPLACEMENTS[ch]
-                    elif ch <= '\xFF':
-                        data = '\\x%02X' % ord(ch)
-                    elif ch <= '\uFFFF':
-                        data = '\\u%04X' % ord(ch)
+                        data = u'\\'+self.ESCAPE_REPLACEMENTS[ch]
+                    elif ch <= u'\xFF':
+                        data = u'\\x%02X' % ord(ch)
+                    elif ch <= u'\uFFFF':
+                        data = u'\\u%04X' % ord(ch)
                     else:
-                        data = '\\U%08X' % ord(ch)
+                        data = u'\\U%08X' % ord(ch)
                     self.column += len(data)
                     if self.encoding:
                         data = data.encode(self.encoding)
                     self.stream.write(data)
                     start = end+1
-            if 0 < end < len(text)-1 and (ch == ' ' or start >= end)    \
+            if 0 < end < len(text)-1 and (ch == u' ' or start >= end)   \
                     and self.column+(end-start) > self.best_width and split:
-                data = text[start:end]+'\\'
+                data = text[start:end]+u'\\'
                 if start < end:
                     start = end
                 self.column += len(data)
@@ -962,30 +965,30 @@ class Emitter:
                 self.write_indent()
                 self.whitespace = False
                 self.indention = False
-                if text[start] == ' ':
-                    data = '\\'
+                if text[start] == u' ':
+                    data = u'\\'
                     self.column += len(data)
                     if self.encoding:
                         data = data.encode(self.encoding)
                     self.stream.write(data)
             end += 1
-        self.write_indicator('"', False)
+        self.write_indicator(u'"', False)
 
     def determine_block_hints(self, text):
-        hints = ''
+        hints = u''
         if text:
-            if text[0] in ' \n\x85\u2028\u2029':
-                hints += str(self.best_indent)
-            if text[-1] not in '\n\x85\u2028\u2029':
-                hints += '-'
-            elif len(text) == 1 or text[-2] in '\n\x85\u2028\u2029':
-                hints += '+'
+            if text[0] in u' \n\x85\u2028\u2029':
+                hints += unicode(self.best_indent)
+            if text[-1] not in u'\n\x85\u2028\u2029':
+                hints += u'-'
+            elif len(text) == 1 or text[-2] in u'\n\x85\u2028\u2029':
+                hints += u'+'
         return hints
 
     def write_folded(self, text):
         hints = self.determine_block_hints(text)
-        self.write_indicator('>'+hints, True)
-        if hints[-1:] == '+':
+        self.write_indicator(u'>'+hints, True)
+        if hints[-1:] == u'+':
             self.open_ended = True
         self.write_line_break()
         leading_space = True
@@ -997,13 +1000,13 @@ class Emitter:
             if end < len(text):
                 ch = text[end]
             if breaks:
-                if ch is None or ch not in '\n\x85\u2028\u2029':
-                    if not leading_space and ch is not None and ch != ' '   \
-                            and text[start] == '\n':
+                if ch is None or ch not in u'\n\x85\u2028\u2029':
+                    if not leading_space and ch is not None and ch != u' '  \
+                            and text[start] == u'\n':
                         self.write_line_break()
-                    leading_space = (ch == ' ')
+                    leading_space = (ch == u' ')
                     for br in text[start:end]:
-                        if br == '\n':
+                        if br == u'\n':
                             self.write_line_break()
                         else:
                             self.write_line_break(br)
@@ -1011,7 +1014,7 @@ class Emitter:
                         self.write_indent()
                     start = end
             elif spaces:
-                if ch != ' ':
+                if ch != u' ':
                     if start+1 == end and self.column > self.best_width:
                         self.write_indent()
                     else:
@@ -1022,7 +1025,7 @@ class Emitter:
                         self.stream.write(data)
                     start = end
             else:
-                if ch is None or ch in ' \n\x85\u2028\u2029':
+                if ch is None or ch in u' \n\x85\u2028\u2029':
                     data = text[start:end]
                     self.column += len(data)
                     if self.encoding:
@@ -1032,14 +1035,14 @@ class Emitter:
                         self.write_line_break()
                     start = end
             if ch is not None:
-                breaks = (ch in '\n\x85\u2028\u2029')
-                spaces = (ch == ' ')
+                breaks = (ch in u'\n\x85\u2028\u2029')
+                spaces = (ch == u' ')
             end += 1
 
     def write_literal(self, text):
         hints = self.determine_block_hints(text)
-        self.write_indicator('|'+hints, True)
-        if hints[-1:] == '+':
+        self.write_indicator(u'|'+hints, True)
+        if hints[-1:] == u'+':
             self.open_ended = True
         self.write_line_break()
         breaks = True
@@ -1049,9 +1052,9 @@ class Emitter:
             if end < len(text):
                 ch = text[end]
             if breaks:
-                if ch is None or ch not in '\n\x85\u2028\u2029':
+                if ch is None or ch not in u'\n\x85\u2028\u2029':
                     for br in text[start:end]:
-                        if br == '\n':
+                        if br == u'\n':
                             self.write_line_break()
                         else:
                             self.write_line_break(br)
@@ -1059,7 +1062,7 @@ class Emitter:
                         self.write_indent()
                     start = end
             else:
-                if ch is None or ch in '\n\x85\u2028\u2029':
+                if ch is None or ch in u'\n\x85\u2028\u2029':
                     data = text[start:end]
                     if self.encoding:
                         data = data.encode(self.encoding)
@@ -1068,7 +1071,7 @@ class Emitter:
                         self.write_line_break()
                     start = end
             if ch is not None:
-                breaks = (ch in '\n\x85\u2028\u2029')
+                breaks = (ch in u'\n\x85\u2028\u2029')
             end += 1
 
     def write_plain(self, text, split=True):
@@ -1077,7 +1080,7 @@ class Emitter:
         if not text:
             return
         if not self.whitespace:
-            data = ' '
+            data = u' '
             self.column += len(data)
             if self.encoding:
                 data = data.encode(self.encoding)
@@ -1092,7 +1095,7 @@ class Emitter:
             if end < len(text):
                 ch = text[end]
             if spaces:
-                if ch != ' ':
+                if ch != u' ':
                     if start+1 == end and self.column > self.best_width and split:
                         self.write_indent()
                         self.whitespace = False
@@ -1105,11 +1108,11 @@ class Emitter:
                         self.stream.write(data)
                     start = end
             elif breaks:
-                if ch not in '\n\x85\u2028\u2029':
-                    if text[start] == '\n':
+                if ch not in u'\n\x85\u2028\u2029':
+                    if text[start] == u'\n':
                         self.write_line_break()
                     for br in text[start:end]:
-                        if br == '\n':
+                        if br == u'\n':
                             self.write_line_break()
                         else:
                             self.write_line_break(br)
@@ -1118,7 +1121,7 @@ class Emitter:
                     self.indention = False
                     start = end
             else:
-                if ch is None or ch in ' \n\x85\u2028\u2029':
+                if ch is None or ch in u' \n\x85\u2028\u2029':
                     data = text[start:end]
                     self.column += len(data)
                     if self.encoding:
@@ -1126,7 +1129,7 @@ class Emitter:
                     self.stream.write(data)
                     start = end
             if ch is not None:
-                spaces = (ch == ' ')
-                breaks = (ch in '\n\x85\u2028\u2029')
+                spaces = (ch == u' ')
+                breaks = (ch in u'\n\x85\u2028\u2029')
             end += 1
 

tablib/packages/yaml/error.py

@@ -1,7 +1,7 @@
 
 __all__ = ['Mark', 'YAMLError', 'MarkedYAMLError']
 
-class Mark:
+class Mark(object):
 
     def __init__(self, name, index, line, column, buffer, pointer):
         self.name = name
@@ -16,7 +16,7 @@ class Mark:
             return None
         head = ''
         start = self.pointer
-        while start > 0 and self.buffer[start-1] not in '\0\r\n\x85\u2028\u2029':
+        while start > 0 and self.buffer[start-1] not in u'\0\r\n\x85\u2028\u2029':
             start -= 1
             if self.pointer-start > max_length/2-1:
                 head = ' ... '
@@ -24,13 +24,13 @@ class Mark:
                 break
         tail = ''
         end = self.pointer
-        while end < len(self.buffer) and self.buffer[end] not in '\0\r\n\x85\u2028\u2029':
+        while end < len(self.buffer) and self.buffer[end] not in u'\0\r\n\x85\u2028\u2029':
             end += 1
             if end-self.pointer > max_length/2-1:
                 tail = ' ... '
                 end -= 5
                 break
-        snippet = self.buffer[start:end]
+        snippet = self.buffer[start:end].encode('utf-8')
         return ' '*indent + head + snippet + tail + '\n'  \
                 + ' '*(indent+self.pointer-start+len(head)) + '^'
 

tablib/packages/yaml/loader.py

@@ -1,12 +1,12 @@
 
 __all__ = ['BaseLoader', 'SafeLoader', 'Loader']
 
-from .reader import *
-from .scanner import *
-from .parser import *
-from .composer import *
-from .constructor import *
-from .resolver import *
+from reader import *
+from scanner import *
+from parser import *
+from composer import *
+from constructor import *
+from resolver import *
 
 class BaseLoader(Reader, Scanner, Parser, Composer, BaseConstructor, BaseResolver):
 

tablib/packages/yaml/parser.py

@@ -61,21 +61,21 @@
 
 __all__ = ['Parser', 'ParserError']
 
-from .error import MarkedYAMLError
-from .tokens import *
-from .events import *
-from .scanner import *
+from error import MarkedYAMLError
+from tokens import *
+from events import *
+from scanner import *
 
 class ParserError(MarkedYAMLError):
     pass
 
-class Parser:
+class Parser(object):
     # Since writing a recursive-descendant parser is a straightforward task, we
     # do not give many comments here.
 
     DEFAULT_TAGS = {
-        '!':   '!',
-        '!!':  'tag:yaml.org,2002:',
+        u'!':   u'!',
+        u'!!':  u'tag:yaml.org,2002:',
     }
 
     def __init__(self):
@@ -214,7 +214,7 @@ class Parser:
         self.tag_handles = {}
         while self.check_token(DirectiveToken):
             token = self.get_token()
-            if token.name == 'YAML':
+            if token.name == u'YAML':
                 if self.yaml_version is not None:
                     raise ParserError(None, None,
                             "found duplicate YAML directive", token.start_mark)
@@ -224,11 +224,11 @@ class Parser:
                             "found incompatible YAML document (version 1.* is required)",
                             token.start_mark)
                 self.yaml_version = token.value
-            elif token.name == 'TAG':
+            elif token.name == u'TAG':
                 handle, prefix = token.value
                 if handle in self.tag_handles:
                     raise ParserError(None, None,
-                            "duplicate tag handle %r" % handle,
+                            "duplicate tag handle %r" % handle.encode('utf-8'),
                             token.start_mark)
                 self.tag_handles[handle] = prefix
         if self.tag_handles:
@@ -298,19 +298,19 @@ class Parser:
                 if handle is not None:
                     if handle not in self.tag_handles:
                         raise ParserError("while parsing a node", start_mark,
-                                "found undefined tag handle %r" % handle,
+                                "found undefined tag handle %r" % handle.encode('utf-8'),
                                 tag_mark)
                     tag = self.tag_handles[handle]+suffix
                 else:
                     tag = suffix
-            #if tag == '!':
+            #if tag == u'!':
             #    raise ParserError("while parsing a node", start_mark,
             #            "found non-specific tag '!'", tag_mark,
             #            "Please check 'http://pyyaml.org/wiki/YAMLNonSpecificTag' and share your opinion.")
             if start_mark is None:
                 start_mark = end_mark = self.peek_token().start_mark
             event = None
-            implicit = (tag is None or tag == '!')
+            implicit = (tag is None or tag == u'!')
             if indentless_sequence and self.check_token(BlockEntryToken):
                 end_mark = self.peek_token().end_mark
                 event = SequenceStartEvent(anchor, tag, implicit,
@@ -320,7 +320,7 @@ class Parser:
                 if self.check_token(ScalarToken):
                     token = self.get_token()
                     end_mark = token.end_mark
-                    if (token.plain and tag is None) or tag == '!':
+                    if (token.plain and tag is None) or tag == u'!':
                         implicit = (True, False)
                     elif tag is None:
                         implicit = (False, True)
@@ -352,7 +352,7 @@ class Parser:
                 elif anchor is not None or tag is not None:
                     # Empty scalars are allowed even if a tag or an anchor is
                     # specified.
-                    event = ScalarEvent(anchor, tag, (implicit, False), '',
+                    event = ScalarEvent(anchor, tag, (implicit, False), u'',
                             start_mark, end_mark)
                     self.state = self.states.pop()
                 else:
@@ -580,5 +580,5 @@ class Parser:
         return self.process_empty_scalar(self.peek_token().start_mark)
 
     def process_empty_scalar(self, mark):
-        return ScalarEvent(None, None, (True, False), '', mark, mark)
+        return ScalarEvent(None, None, (True, False), u'', mark, mark)
 

tablib/packages/yaml/reader.py

@@ -17,10 +17,45 @@
 
 __all__ = ['Reader', 'ReaderError']
 
-from .error import YAMLError, Mark
+from error import YAMLError, Mark
 
 import codecs, re
 
+# Unfortunately, codec functions in Python 2.3 does not support the `finish`
+# arguments, so we have to write our own wrappers.
+
+try:
+    codecs.utf_8_decode('', 'strict', False)
+    from codecs import utf_8_decode, utf_16_le_decode, utf_16_be_decode
+
+except TypeError:
+
+    def utf_16_le_decode(data, errors, finish=False):
+        if not finish and len(data) % 2 == 1:
+            data = data[:-1]
+        return codecs.utf_16_le_decode(data, errors)
+
+    def utf_16_be_decode(data, errors, finish=False):
+        if not finish and len(data) % 2 == 1:
+            data = data[:-1]
+        return codecs.utf_16_be_decode(data, errors)
+
+    def utf_8_decode(data, errors, finish=False):
+        if not finish:
+            # We are trying to remove a possible incomplete multibyte character
+            # from the suffix of the data.
+            # The first byte of a multi-byte sequence is in the range 0xc0 to 0xfd.
+            # All further bytes are in the range 0x80 to 0xbf.
+            # UTF-8 encoded UCS characters may be up to six bytes long.
+            count = 0
+            while count < 5 and count < len(data)   \
+                    and '\x80' <= data[-count-1] <= '\xBF':
+                count -= 1
+            if count < 5 and count < len(data)  \
+                    and '\xC0' <= data[-count-1] <= '\xFD':
+                data = data[:-count-1]
+        return codecs.utf_8_decode(data, errors)
+
 class ReaderError(YAMLError):
 
     def __init__(self, name, position, character, encoding, reason):
@@ -31,7 +66,7 @@ class ReaderError(YAMLError):
         self.reason = reason
 
     def __str__(self):
-        if isinstance(self.character, bytes):
+        if isinstance(self.character, str):
             return "'%s' codec can't decode byte #x%02x: %s\n"  \
                     "  in \"%s\", position %d"    \
                     % (self.encoding, ord(self.character), self.reason,
@@ -44,13 +79,13 @@ class ReaderError(YAMLError):
 
 class Reader(object):
     # Reader:
-    # - determines the data encoding and converts it to a unicode string,
+    # - determines the data encoding and converts it to unicode,
     # - checks if characters are in allowed range,
     # - adds '\0' to the end.
 
     # Reader accepts
-    #  - a `bytes` object,
     #  - a `str` object,
+    #  - a `unicode` object,
     #  - a file-like object with its `read` method returning `str`,
     #  - a file-like object with its `read` method returning `unicode`.
 
@@ -61,7 +96,7 @@ class Reader(object):
         self.stream = None
         self.stream_pointer = 0
         self.eof = True
-        self.buffer = ''
+        self.buffer = u''
         self.pointer = 0
         self.raw_buffer = None
         self.raw_decode = None
@@ -69,19 +104,19 @@ class Reader(object):
         self.index = 0
         self.line = 0
         self.column = 0
-        if isinstance(stream, str):
+        if isinstance(stream, unicode):
             self.name = "<unicode string>"
             self.check_printable(stream)
-            self.buffer = stream+'\0'
-        elif isinstance(stream, bytes):
-            self.name = "<byte string>"
+            self.buffer = stream+u'\0'
+        elif isinstance(stream, str):
+            self.name = "<string>"
             self.raw_buffer = stream
             self.determine_encoding()
         else:
             self.stream = stream
             self.name = getattr(stream, 'name', "<file>")
             self.eof = False
-            self.raw_buffer = None
+            self.raw_buffer = ''
             self.determine_encoding()
 
     def peek(self, index=0):
@@ -103,11 +138,11 @@ class Reader(object):
             ch = self.buffer[self.pointer]
             self.pointer += 1
             self.index += 1
-            if ch in '\n\x85\u2028\u2029'  \
-                    or (ch == '\r' and self.buffer[self.pointer] != '\n'):
+            if ch in u'\n\x85\u2028\u2029'  \
+                    or (ch == u'\r' and self.buffer[self.pointer] != u'\n'):
                 self.line += 1
                 self.column = 0
-            elif ch != '\uFEFF':
+            elif ch != u'\uFEFF':
                 self.column += 1
             length -= 1
 
@@ -120,21 +155,21 @@ class Reader(object):
                     None, None)
 
     def determine_encoding(self):
-        while not self.eof and (self.raw_buffer is None or len(self.raw_buffer) < 2):
+        while not self.eof and len(self.raw_buffer) < 2:
             self.update_raw()
-        if isinstance(self.raw_buffer, bytes):
+        if not isinstance(self.raw_buffer, unicode):
             if self.raw_buffer.startswith(codecs.BOM_UTF16_LE):
-                self.raw_decode = codecs.utf_16_le_decode
+                self.raw_decode = utf_16_le_decode
                 self.encoding = 'utf-16-le'
             elif self.raw_buffer.startswith(codecs.BOM_UTF16_BE):
-                self.raw_decode = codecs.utf_16_be_decode
+                self.raw_decode = utf_16_be_decode
                 self.encoding = 'utf-16-be'
             else:
-                self.raw_decode = codecs.utf_8_decode
+                self.raw_decode = utf_8_decode
                 self.encoding = 'utf-8'
         self.update(1)
 
-    NON_PRINTABLE = re.compile('[^\x09\x0A\x0D\x20-\x7E\x85\xA0-\uD7FF\uE000-\uFFFD]')
+    NON_PRINTABLE = re.compile(u'[^\x09\x0A\x0D\x20-\x7E\x85\xA0-\uD7FF\uE000-\uFFFD]')
     def check_printable(self, data):
         match = self.NON_PRINTABLE.search(data)
         if match:
@@ -155,8 +190,8 @@ class Reader(object):
                 try:
                     data, converted = self.raw_decode(self.raw_buffer,
                             'strict', self.eof)
-                except UnicodeDecodeError as exc:
-                    character = self.raw_buffer[exc.start]
+                except UnicodeDecodeError, exc:
+                    character = exc.object[exc.start]
                     if self.stream is not None:
                         position = self.stream_pointer-len(self.raw_buffer)+exc.start
                     else:
@@ -170,18 +205,16 @@ class Reader(object):
             self.buffer += data
             self.raw_buffer = self.raw_buffer[converted:]
             if self.eof:
-                self.buffer += '\0'
+                self.buffer += u'\0'
                 self.raw_buffer = None
                 break
 
-    def update_raw(self, size=4096):
+    def update_raw(self, size=1024):
         data = self.stream.read(size)
-        if self.raw_buffer is None:
-            self.raw_buffer = data
-        else:
+        if data:
             self.raw_buffer += data
-        self.stream_pointer += len(data)
-        if not data:
+            self.stream_pointer += len(data)
+        else:
             self.eof = True
 
 #try:

tablib/packages/yaml/representer.py

@@ -2,15 +2,22 @@
 __all__ = ['BaseRepresenter', 'SafeRepresenter', 'Representer',
     'RepresenterError']
 
-from .error import *
-from .nodes import *
+from error import *
+from nodes import *
 
-import datetime, sys, copyreg, types, base64
+import datetime
+
+try:
+    set
+except NameError:
+    from sets import Set as set
+
+import sys, copy_reg, types
 
 class RepresenterError(YAMLError):
     pass
 
-class BaseRepresenter:
+class BaseRepresenter(object):
 
     yaml_representers = {}
     yaml_multi_representers = {}
@@ -29,6 +36,12 @@ class BaseRepresenter:
         self.object_keeper = []
         self.alias_key = None
 
+    def get_classobj_bases(self, cls):
+        bases = [cls]
+        for base in cls.__bases__:
+            bases.extend(self.get_classobj_bases(base))
+        return bases
+
     def represent_data(self, data):
         if self.ignore_aliases(data):
             self.alias_key = None
@@ -43,6 +56,8 @@ class BaseRepresenter:
             #self.represented_objects[alias_key] = None
             self.object_keeper.append(data)
         data_types = type(data).__mro__
+        if type(data) is types.InstanceType:
+            data_types = self.get_classobj_bases(data.__class__)+list(data_types)
         if data_types[0] in self.yaml_representers:
             node = self.yaml_representers[data_types[0]](self, data)
         else:
@@ -56,22 +71,22 @@ class BaseRepresenter:
                 elif None in self.yaml_representers:
                     node = self.yaml_representers[None](self, data)
                 else:
-                    node = ScalarNode(None, str(data))
+                    node = ScalarNode(None, unicode(data))
         #if alias_key is not None:
         #    self.represented_objects[alias_key] = node
         return node
 
-    @classmethod
     def add_representer(cls, data_type, representer):
         if not 'yaml_representers' in cls.__dict__:
             cls.yaml_representers = cls.yaml_representers.copy()
         cls.yaml_representers[data_type] = representer
+    add_representer = classmethod(add_representer)
 
-    @classmethod
     def add_multi_representer(cls, data_type, representer):
         if not 'yaml_multi_representers' in cls.__dict__:
             cls.yaml_multi_representers = cls.yaml_multi_representers.copy()
         cls.yaml_multi_representers[data_type] = representer
+    add_multi_representer = classmethod(add_multi_representer)
 
     def represent_scalar(self, tag, value, style=None):
         if style is None:
@@ -106,11 +121,8 @@ class BaseRepresenter:
             self.represented_objects[self.alias_key] = node
         best_style = True
         if hasattr(mapping, 'items'):
-            mapping = list(mapping.items())
-            try:
-                mapping = sorted(mapping)
-            except TypeError:
-                pass
+            mapping = mapping.items()
+            mapping.sort()
         for item_key, item_value in mapping:
             node_key = self.represent_data(item_key)
             node_value = self.represent_data(item_value)
@@ -134,31 +146,44 @@ class SafeRepresenter(BaseRepresenter):
     def ignore_aliases(self, data):
         if data in [None, ()]:
             return True
-        if isinstance(data, (str, bytes, bool, int, float)):
+        if isinstance(data, (str, unicode, bool, int, float)):
             return True
 
     def represent_none(self, data):
-        return self.represent_scalar('tag:yaml.org,2002:null', 'null')
+        return self.represent_scalar(u'tag:yaml.org,2002:null',
+                u'null')
 
     def represent_str(self, data):
-        return self.represent_scalar('tag:yaml.org,2002:str', data)
+        tag = None
+        style = None
+        try:
+            data = unicode(data, 'ascii')
+            tag = u'tag:yaml.org,2002:str'
+        except UnicodeDecodeError:
+            try:
+                data = unicode(data, 'utf-8')
+                tag = u'tag:yaml.org,2002:str'
+            except UnicodeDecodeError:
+                data = data.encode('base64')
+                tag = u'tag:yaml.org,2002:binary'
+                style = '|'
+        return self.represent_scalar(tag, data, style=style)
 
-    def represent_binary(self, data):
-        if hasattr(base64, 'encodebytes'):
-            data = base64.encodebytes(data).decode('ascii')
-        else:
-            data = base64.encodestring(data).decode('ascii')
-        return self.represent_scalar('tag:yaml.org,2002:binary', data, style='|')
+    def represent_unicode(self, data):
+        return self.represent_scalar(u'tag:yaml.org,2002:str', data)
 
     def represent_bool(self, data):
         if data:
-            value = 'true'
+            value = u'true'
         else:
-            value = 'false'
-        return self.represent_scalar('tag:yaml.org,2002:bool', value)
+            value = u'false'
+        return self.represent_scalar(u'tag:yaml.org,2002:bool', value)
 
     def represent_int(self, data):
-        return self.represent_scalar('tag:yaml.org,2002:int', str(data))
+        return self.represent_scalar(u'tag:yaml.org,2002:int', unicode(data))
+
+    def represent_long(self, data):
+        return self.represent_scalar(u'tag:yaml.org,2002:int', unicode(data))
 
     inf_value = 1e300
     while repr(inf_value) != repr(inf_value*inf_value):
@@ -166,13 +191,13 @@ class SafeRepresenter(BaseRepresenter):
 
     def represent_float(self, data):
         if data != data or (data == 0.0 and data == 1.0):
-            value = '.nan'
+            value = u'.nan'
         elif data == self.inf_value:
-            value = '.inf'
+            value = u'.inf'
         elif data == -self.inf_value:
-            value = '-.inf'
+            value = u'-.inf'
         else:
-            value = repr(data).lower()
+            value = unicode(repr(data)).lower()
             # Note that in some cases `repr(data)` represents a float number
             # without the decimal parts.  For instance:
             #   >>> repr(1e17)
@@ -180,9 +205,9 @@ class SafeRepresenter(BaseRepresenter):
             # Unfortunately, this is not a valid float representation according
             # to the definition of the `!!float` tag.  We fix this by adding
             # '.0' before the 'e' symbol.
-            if '.' not in value and 'e' in value:
-                value = value.replace('e', '.0e', 1)
-        return self.represent_scalar('tag:yaml.org,2002:float', value)
+            if u'.' not in value and u'e' in value:
+                value = value.replace(u'e', u'.0e', 1)
+        return self.represent_scalar(u'tag:yaml.org,2002:float', value)
 
     def represent_list(self, data):
         #pairs = (len(data) > 0 and isinstance(data, list))
@@ -192,7 +217,7 @@ class SafeRepresenter(BaseRepresenter):
         #            pairs = False
         #            break
         #if not pairs:
-            return self.represent_sequence('tag:yaml.org,2002:seq', data)
+            return self.represent_sequence(u'tag:yaml.org,2002:seq', data)
         #value = []
         #for item_key, item_value in data:
         #    value.append(self.represent_mapping(u'tag:yaml.org,2002:map',
@@ -200,21 +225,21 @@ class SafeRepresenter(BaseRepresenter):
         #return SequenceNode(u'tag:yaml.org,2002:pairs', value)
 
     def represent_dict(self, data):
-        return self.represent_mapping('tag:yaml.org,2002:map', data)
+        return self.represent_mapping(u'tag:yaml.org,2002:map', data)
 
     def represent_set(self, data):
         value = {}
         for key in data:
             value[key] = None
-        return self.represent_mapping('tag:yaml.org,2002:set', value)
+        return self.represent_mapping(u'tag:yaml.org,2002:set', value)
 
     def represent_date(self, data):
-        value = data.isoformat()
-        return self.represent_scalar('tag:yaml.org,2002:timestamp', value)
+        value = unicode(data.isoformat())
+        return self.represent_scalar(u'tag:yaml.org,2002:timestamp', value)
 
     def represent_datetime(self, data):
-        value = data.isoformat(' ')
-        return self.represent_scalar('tag:yaml.org,2002:timestamp', value)
+        value = unicode(data.isoformat(' '))
+        return self.represent_scalar(u'tag:yaml.org,2002:timestamp', value)
 
     def represent_yaml_object(self, tag, data, cls, flow_style=None):
         if hasattr(data, '__getstate__'):
@@ -232,8 +257,8 @@ SafeRepresenter.add_representer(type(None),
 SafeRepresenter.add_representer(str,
         SafeRepresenter.represent_str)
 
-SafeRepresenter.add_representer(bytes,
-        SafeRepresenter.represent_binary)
+SafeRepresenter.add_representer(unicode,
+        SafeRepresenter.represent_unicode)
 
 SafeRepresenter.add_representer(bool,
         SafeRepresenter.represent_bool)
@@ -241,6 +266,9 @@ SafeRepresenter.add_representer(bool,
 SafeRepresenter.add_representer(int,
         SafeRepresenter.represent_int)
 
+SafeRepresenter.add_representer(long,
+        SafeRepresenter.represent_long)
+
 SafeRepresenter.add_representer(float,
         SafeRepresenter.represent_float)
 
@@ -267,27 +295,99 @@ SafeRepresenter.add_representer(None,
 
 class Representer(SafeRepresenter):
 
+    def represent_str(self, data):
+        tag = None
+        style = None
+        try:
+            data = unicode(data, 'ascii')
+            tag = u'tag:yaml.org,2002:str'
+        except UnicodeDecodeError:
+            try:
+                data = unicode(data, 'utf-8')
+                tag = u'tag:yaml.org,2002:python/str'
+            except UnicodeDecodeError:
+                data = data.encode('base64')
+                tag = u'tag:yaml.org,2002:binary'
+                style = '|'
+        return self.represent_scalar(tag, data, style=style)
+
+    def represent_unicode(self, data):
+        tag = None
+        try:
+            data.encode('ascii')
+            tag = u'tag:yaml.org,2002:python/unicode'
+        except UnicodeEncodeError:
+            tag = u'tag:yaml.org,2002:str'
+        return self.represent_scalar(tag, data)
+
+    def represent_long(self, data):
+        tag = u'tag:yaml.org,2002:int'
+        if int(data) is not data:
+            tag = u'tag:yaml.org,2002:python/long'
+        return self.represent_scalar(tag, unicode(data))
+
     def represent_complex(self, data):
         if data.imag == 0.0:
-            data = '%r' % data.real
+            data = u'%r' % data.real
         elif data.real == 0.0:
-            data = '%rj' % data.imag
+            data = u'%rj' % data.imag
         elif data.imag > 0:
-            data = '%r+%rj' % (data.real, data.imag)
+            data = u'%r+%rj' % (data.real, data.imag)
         else:
-            data = '%r%rj' % (data.real, data.imag)
-        return self.represent_scalar('tag:yaml.org,2002:python/complex', data)
+            data = u'%r%rj' % (data.real, data.imag)
+        return self.represent_scalar(u'tag:yaml.org,2002:python/complex', data)
 
     def represent_tuple(self, data):
-        return self.represent_sequence('tag:yaml.org,2002:python/tuple', data)
+        return self.represent_sequence(u'tag:yaml.org,2002:python/tuple', data)
 
     def represent_name(self, data):
-        name = '%s.%s' % (data.__module__, data.__name__)
-        return self.represent_scalar('tag:yaml.org,2002:python/name:'+name, '')
+        name = u'%s.%s' % (data.__module__, data.__name__)
+        return self.represent_scalar(u'tag:yaml.org,2002:python/name:'+name, u'')
 
     def represent_module(self, data):
         return self.represent_scalar(
-                'tag:yaml.org,2002:python/module:'+data.__name__, '')
+                u'tag:yaml.org,2002:python/module:'+data.__name__, u'')
+
+    def represent_instance(self, data):
+        # For instances of classic classes, we use __getinitargs__ and
+        # __getstate__ to serialize the data.
+
+        # If data.__getinitargs__ exists, the object must be reconstructed by
+        # calling cls(**args), where args is a tuple returned by
+        # __getinitargs__. Otherwise, the cls.__init__ method should never be
+        # called and the class instance is created by instantiating a trivial
+        # class and assigning to the instance's __class__ variable.
+
+        # If data.__getstate__ exists, it returns the state of the object.
+        # Otherwise, the state of the object is data.__dict__.
+
+        # We produce either a !!python/object or !!python/object/new node.
+        # If data.__getinitargs__ does not exist and state is a dictionary, we
+        # produce a !!python/object node . Otherwise we produce a
+        # !!python/object/new node.
+
+        cls = data.__class__
+        class_name = u'%s.%s' % (cls.__module__, cls.__name__)
+        args = None
+        state = None
+        if hasattr(data, '__getinitargs__'):
+            args = list(data.__getinitargs__())
+        if hasattr(data, '__getstate__'):
+            state = data.__getstate__()
+        else:
+            state = data.__dict__
+        if args is None and isinstance(state, dict):
+            return self.represent_mapping(
+                    u'tag:yaml.org,2002:python/object:'+class_name, state)
+        if isinstance(state, dict) and not state:
+            return self.represent_sequence(
+                    u'tag:yaml.org,2002:python/object/new:'+class_name, args)
+        value = {}
+        if args:
+            value['args'] = args
+        value['state'] = state
+        return self.represent_mapping(
+                u'tag:yaml.org,2002:python/object/new:'+class_name, value)
 
     def represent_object(self, data):
         # We use __reduce__ API to save the data. data.__reduce__ returns
@@ -307,8 +407,8 @@ class Representer(SafeRepresenter):
         # !!python/object/apply node.
 
         cls = type(data)
-        if cls in copyreg.dispatch_table:
-            reduce = copyreg.dispatch_table[cls](data)
+        if cls in copy_reg.dispatch_table:
+            reduce = copy_reg.dispatch_table[cls](data)
         elif hasattr(data, '__reduce_ex__'):
             reduce = data.__reduce_ex__(2)
         elif hasattr(data, '__reduce__'):
@@ -327,16 +427,16 @@ class Representer(SafeRepresenter):
         if function.__name__ == '__newobj__':
             function = args[0]
             args = args[1:]
-            tag = 'tag:yaml.org,2002:python/object/new:'
+            tag = u'tag:yaml.org,2002:python/object/new:'
             newobj = True
         else:
-            tag = 'tag:yaml.org,2002:python/object/apply:'
+            tag = u'tag:yaml.org,2002:python/object/apply:'
             newobj = False
-        function_name = '%s.%s' % (function.__module__, function.__name__)
+        function_name = u'%s.%s' % (function.__module__, function.__name__)
         if not args and not listitems and not dictitems \
                 and isinstance(state, dict) and newobj:
             return self.represent_mapping(
-                    'tag:yaml.org,2002:python/object:'+function_name, state)
+                    u'tag:yaml.org,2002:python/object:'+function_name, state)
         if not listitems and not dictitems  \
                 and isinstance(state, dict) and not state:
             return self.represent_sequence(tag+function_name, args)
@@ -351,6 +451,15 @@ class Representer(SafeRepresenter):
             value['dictitems'] = dictitems
         return self.represent_mapping(tag+function_name, value)
 
+Representer.add_representer(str,
+        Representer.represent_str)
+
+Representer.add_representer(unicode,
+        Representer.represent_unicode)
+
+Representer.add_representer(long,
+        Representer.represent_long)
+
 Representer.add_representer(complex,
         Representer.represent_complex)
 
@@ -360,6 +469,9 @@ Representer.add_representer(tuple,
 Representer.add_representer(type,
         Representer.represent_name)
 
+Representer.add_representer(types.ClassType,
+        Representer.represent_name)
+
 Representer.add_representer(types.FunctionType,
         Representer.represent_name)
 
@@ -369,6 +481,9 @@ Representer.add_representer(types.BuiltinFunctionType,
 Representer.add_representer(types.ModuleType,
         Representer.represent_module)
 
+Representer.add_multi_representer(types.InstanceType,
+        Representer.represent_instance)
+
 Representer.add_multi_representer(object,
         Representer.represent_object)
 

tablib/packages/yaml/resolver.py

@@ -1,19 +1,19 @@
 
 __all__ = ['BaseResolver', 'Resolver']
 
-from .error import *
-from .nodes import *
+from error import *
+from nodes import *
 
 import re
 
 class ResolverError(YAMLError):
     pass
 
-class BaseResolver:
+class BaseResolver(object):
 
-    DEFAULT_SCALAR_TAG = 'tag:yaml.org,2002:str'
-    DEFAULT_SEQUENCE_TAG = 'tag:yaml.org,2002:seq'
-    DEFAULT_MAPPING_TAG = 'tag:yaml.org,2002:map'
+    DEFAULT_SCALAR_TAG = u'tag:yaml.org,2002:str'
+    DEFAULT_SEQUENCE_TAG = u'tag:yaml.org,2002:seq'
+    DEFAULT_MAPPING_TAG = u'tag:yaml.org,2002:map'
 
     yaml_implicit_resolvers = {}
     yaml_path_resolvers = {}
@@ -22,7 +22,6 @@ class BaseResolver:
         self.resolver_exact_paths = []
         self.resolver_prefix_paths = []
 
-    @classmethod
     def add_implicit_resolver(cls, tag, regexp, first):
         if not 'yaml_implicit_resolvers' in cls.__dict__:
             cls.yaml_implicit_resolvers = cls.yaml_implicit_resolvers.copy()
@@ -30,8 +29,8 @@ class BaseResolver:
             first = [None]
         for ch in first:
             cls.yaml_implicit_resolvers.setdefault(ch, []).append((tag, regexp))
+    add_implicit_resolver = classmethod(add_implicit_resolver)
 
-    @classmethod
     def add_path_resolver(cls, tag, path, kind=None):
         # Note: `add_path_resolver` is experimental.  The API could be changed.
         # `new_path` is a pattern that is matched against the path from the
@@ -67,10 +66,10 @@ class BaseResolver:
             elif node_check is dict:
                 node_check = MappingNode
             elif node_check not in [ScalarNode, SequenceNode, MappingNode]  \
-                    and not isinstance(node_check, str) \
+                    and not isinstance(node_check, basestring)  \
                     and node_check is not None:
                 raise ResolverError("Invalid node checker: %s" % node_check)
-            if not isinstance(index_check, (str, int))  \
+            if not isinstance(index_check, (basestring, int))   \
                     and index_check is not None:
                 raise ResolverError("Invalid index checker: %s" % index_check)
             new_path.append((node_check, index_check))
@@ -84,6 +83,7 @@ class BaseResolver:
                 and kind is not None:
             raise ResolverError("Invalid node kind: %s" % kind)
         cls.yaml_path_resolvers[tuple(new_path), kind] = tag
+    add_path_resolver = classmethod(add_path_resolver)
 
     def descend_resolver(self, current_node, current_index):
         if not self.yaml_path_resolvers:
@@ -117,7 +117,7 @@ class BaseResolver:
     def check_resolver_prefix(self, depth, path, kind,
             current_node, current_index):
         node_check, index_check = path[depth-1]
-        if isinstance(node_check, str):
+        if isinstance(node_check, basestring):
             if current_node.tag != node_check:
                 return
         elif node_check is not None:
@@ -128,7 +128,7 @@ class BaseResolver:
         if (index_check is False or index_check is None)    \
                 and current_index is None:
             return
-        if isinstance(index_check, str):
+        if isinstance(index_check, basestring):
             if not (isinstance(current_index, ScalarNode)
                     and index_check == current_index.value):
                 return
@@ -139,8 +139,8 @@ class BaseResolver:
 
     def resolve(self, kind, value, implicit):
         if kind is ScalarNode and implicit[0]:
-            if value == '':
-                resolvers = self.yaml_implicit_resolvers.get('', [])
+            if value == u'':
+                resolvers = self.yaml_implicit_resolvers.get(u'', [])
             else:
                 resolvers = self.yaml_implicit_resolvers.get(value[0], [])
             resolvers += self.yaml_implicit_resolvers.get(None, [])
@@ -165,60 +165,60 @@ class Resolver(BaseResolver):
     pass
 
 Resolver.add_implicit_resolver(
-        'tag:yaml.org,2002:bool',
-        re.compile(r'''^(?:yes|Yes|YES|no|No|NO
+        u'tag:yaml.org,2002:bool',
+        re.compile(ur'''^(?:yes|Yes|YES|no|No|NO
                     |true|True|TRUE|false|False|FALSE
                     |on|On|ON|off|Off|OFF)$''', re.X),
-        list('yYnNtTfFoO'))
+        list(u'yYnNtTfFoO'))
 
 Resolver.add_implicit_resolver(
-        'tag:yaml.org,2002:float',
-        re.compile(r'''^(?:[-+]?(?:[0-9][0-9_]*)\.[0-9_]*(?:[eE][-+][0-9]+)?
+        u'tag:yaml.org,2002:float',
+        re.compile(ur'''^(?:[-+]?(?:[0-9][0-9_]*)\.[0-9_]*(?:[eE][-+][0-9]+)?
                     |\.[0-9_]+(?:[eE][-+][0-9]+)?
                     |[-+]?[0-9][0-9_]*(?::[0-5]?[0-9])+\.[0-9_]*
                     |[-+]?\.(?:inf|Inf|INF)
                     |\.(?:nan|NaN|NAN))$''', re.X),
-        list('-+0123456789.'))
+        list(u'-+0123456789.'))
 
 Resolver.add_implicit_resolver(
-        'tag:yaml.org,2002:int',
-        re.compile(r'''^(?:[-+]?0b[0-1_]+
+        u'tag:yaml.org,2002:int',
+        re.compile(ur'''^(?:[-+]?0b[0-1_]+
                     |[-+]?0[0-7_]+
                     |[-+]?(?:0|[1-9][0-9_]*)
                     |[-+]?0x[0-9a-fA-F_]+
                     |[-+]?[1-9][0-9_]*(?::[0-5]?[0-9])+)$''', re.X),
-        list('-+0123456789'))
+        list(u'-+0123456789'))
 
 Resolver.add_implicit_resolver(
-        'tag:yaml.org,2002:merge',
-        re.compile(r'^(?:<<)$'),
-        ['<'])
+        u'tag:yaml.org,2002:merge',
+        re.compile(ur'^(?:<<)$'),
+        [u'<'])
 
 Resolver.add_implicit_resolver(
-        'tag:yaml.org,2002:null',
-        re.compile(r'''^(?: ~
+        u'tag:yaml.org,2002:null',
+        re.compile(ur'''^(?: ~
                     |null|Null|NULL
                     | )$''', re.X),
-        ['~', 'n', 'N', ''])
+        [u'~', u'n', u'N', u''])
 
 Resolver.add_implicit_resolver(
-        'tag:yaml.org,2002:timestamp',
-        re.compile(r'''^(?:[0-9][0-9][0-9][0-9]-[0-9][0-9]-[0-9][0-9]
+        u'tag:yaml.org,2002:timestamp',
+        re.compile(ur'''^(?:[0-9][0-9][0-9][0-9]-[0-9][0-9]-[0-9][0-9]
                     |[0-9][0-9][0-9][0-9] -[0-9][0-9]? -[0-9][0-9]?
                      (?:[Tt]|[ \t]+)[0-9][0-9]?
                      :[0-9][0-9] :[0-9][0-9] (?:\.[0-9]*)?
                      (?:[ \t]*(?:Z|[-+][0-9][0-9]?(?::[0-9][0-9])?))?)$''', re.X),
-        list('0123456789'))
+        list(u'0123456789'))
 
 Resolver.add_implicit_resolver(
-        'tag:yaml.org,2002:value',
-        re.compile(r'^(?:=)$'),
-        ['='])
+        u'tag:yaml.org,2002:value',
+        re.compile(ur'^(?:=)$'),
+        [u'='])
 
 # The following resolver is only for documentation purposes. It cannot work
 # because plain scalars cannot start with '!', '&', or '*'.
 Resolver.add_implicit_resolver(
-        'tag:yaml.org,2002:yaml',
-        re.compile(r'^(?:!|&|\*)$'),
-        list('!&*'))
+        u'tag:yaml.org,2002:yaml',
+        re.compile(ur'^(?:!|&|\*)$'),
+        list(u'!&*'))
 

tablib/packages/yaml/scanner.py

@@ -26,13 +26,13 @@
 
 __all__ = ['Scanner', 'ScannerError']
 
-from .error import MarkedYAMLError
-from .tokens import *
+from error import MarkedYAMLError
+from tokens import *
 
 class ScannerError(MarkedYAMLError):
     pass
 
-class SimpleKey:
+class SimpleKey(object):
     # See below simple keys treatment.
 
     def __init__(self, token_number, required, index, line, column, mark):
@@ -43,7 +43,7 @@ class SimpleKey:
         self.column = column
         self.mark = mark
 
-class Scanner:
+class Scanner(object):
 
     def __init__(self):
         """Initialize the scanner."""
@@ -166,85 +166,85 @@ class Scanner:
         ch = self.peek()
 
         # Is it the end of stream?
-        if ch == '\0':
+        if ch == u'\0':
             return self.fetch_stream_end()
 
         # Is it a directive?
-        if ch == '%' and self.check_directive():
+        if ch == u'%' and self.check_directive():
             return self.fetch_directive()
 
         # Is it the document start?
-        if ch == '-' and self.check_document_start():
+        if ch == u'-' and self.check_document_start():
             return self.fetch_document_start()
 
         # Is it the document end?
-        if ch == '.' and self.check_document_end():
+        if ch == u'.' and self.check_document_end():
             return self.fetch_document_end()
 
         # TODO: support for BOM within a stream.
-        #if ch == '\uFEFF':
+        #if ch == u'\uFEFF':
         #    return self.fetch_bom()    <-- issue BOMToken
 
         # Note: the order of the following checks is NOT significant.
 
         # Is it the flow sequence start indicator?
-        if ch == '[':
+        if ch == u'[':
             return self.fetch_flow_sequence_start()
 
         # Is it the flow mapping start indicator?
-        if ch == '{':
+        if ch == u'{':
             return self.fetch_flow_mapping_start()
 
         # Is it the flow sequence end indicator?
-        if ch == ']':
+        if ch == u']':
             return self.fetch_flow_sequence_end()
 
         # Is it the flow mapping end indicator?
-        if ch == '}':
+        if ch == u'}':
             return self.fetch_flow_mapping_end()
 
         # Is it the flow entry indicator?
-        if ch == ',':
+        if ch == u',':
             return self.fetch_flow_entry()
 
         # Is it the block entry indicator?
-        if ch == '-' and self.check_block_entry():
+        if ch == u'-' and self.check_block_entry():
             return self.fetch_block_entry()
 
         # Is it the key indicator?
-        if ch == '?' and self.check_key():
+        if ch == u'?' and self.check_key():
             return self.fetch_key()
 
         # Is it the value indicator?
-        if ch == ':' and self.check_value():
+        if ch == u':' and self.check_value():
             return self.fetch_value()
 
         # Is it an alias?
-        if ch == '*':
+        if ch == u'*':
             return self.fetch_alias()
 
         # Is it an anchor?
-        if ch == '&':
+        if ch == u'&':
             return self.fetch_anchor()
 
         # Is it a tag?
-        if ch == '!':
+        if ch == u'!':
             return self.fetch_tag()
 
         # Is it a literal scalar?
-        if ch == '|' and not self.flow_level:
+        if ch == u'|' and not self.flow_level:
             return self.fetch_literal()
 
         # Is it a folded scalar?
-        if ch == '>' and not self.flow_level:
+        if ch == u'>' and not self.flow_level:
             return self.fetch_folded()
 
         # Is it a single quoted scalar?
-        if ch == '\'':
+        if ch == u'\'':
             return self.fetch_single()
 
         # Is it a double quoted scalar?
-        if ch == '\"':
+        if ch == u'\"':
             return self.fetch_double()
 
         # It must be a plain scalar then.
@@ -253,8 +253,8 @@ class Scanner:
 
         # No? It's an error. Let's produce a nice error message.
         raise ScannerError("while scanning for the next token", None,
-                "found character %r that cannot start any token" % ch,
-                self.get_mark())
+                "found character %r that cannot start any token"
+                % ch.encode('utf-8'), self.get_mark())
 
     # Simple keys treatment.
 
@@ -280,7 +280,7 @@ class Scanner:
         # - should be no longer than 1024 characters.
         # Disabling this procedure will allow simple keys of any length and
         # height (may cause problems if indentation is broken though).
-        for level in list(self.possible_simple_keys):
+        for level in self.possible_simple_keys.keys():
             key = self.possible_simple_keys[level]
             if key.line != self.line  \
                     or self.index-key.index > 1024:
@@ -692,22 +692,22 @@ class Scanner:
 
         # DOCUMENT-START:   ^ '---' (' '|'\n')
         if self.column == 0:
-            if self.prefix(3) == '---'  \
-                    and self.peek(3) in '\0 \t\r\n\x85\u2028\u2029':
+            if self.prefix(3) == u'---'  \
+                    and self.peek(3) in u'\0 \t\r\n\x85\u2028\u2029':
                 return True
 
     def check_document_end(self):
 
         # DOCUMENT-END:     ^ '...' (' '|'\n')
         if self.column == 0:
-            if self.prefix(3) == '...'  \
-                    and self.peek(3) in '\0 \t\r\n\x85\u2028\u2029':
+            if self.prefix(3) == u'...'  \
+                    and self.peek(3) in u'\0 \t\r\n\x85\u2028\u2029':
                 return True
 
     def check_block_entry(self):
 
         # BLOCK-ENTRY:      '-' (' '|'\n')
-        return self.peek(1) in '\0 \t\r\n\x85\u2028\u2029'
+        return self.peek(1) in u'\0 \t\r\n\x85\u2028\u2029'
 
     def check_key(self):
 
@@ -717,7 +717,7 @@ class Scanner:
 
         # KEY(block context):   '?' (' '|'\n')
         else:
-            return self.peek(1) in '\0 \t\r\n\x85\u2028\u2029'
+            return self.peek(1) in u'\0 \t\r\n\x85\u2028\u2029'
 
     def check_value(self):
 
@@ -727,7 +727,7 @@ class Scanner:
 
         # VALUE(block context): ':' (' '|'\n')
         else:
-            return self.peek(1) in '\0 \t\r\n\x85\u2028\u2029'
+            return self.peek(1) in u'\0 \t\r\n\x85\u2028\u2029'
 
     def check_plain(self):
 
@@ -744,9 +744,9 @@ class Scanner:
         # '-' character) because we want the flow context to be space
         # independent.
         ch = self.peek()
-        return ch not in '\0 \t\r\n\x85\u2028\u2029-?:,[]{}#&*!|>\'\"%@`'  \
-                or (self.peek(1) not in '\0 \t\r\n\x85\u2028\u2029'
-                        and (ch == '-' or (not self.flow_level and ch in '?:')))
+        return ch not in u'\0 \t\r\n\x85\u2028\u2029-?:,[]{}#&*!|>\'\"%@`'  \
+                or (self.peek(1) not in u'\0 \t\r\n\x85\u2028\u2029'
+                        and (ch == u'-' or (not self.flow_level and ch in u'?:')))
 
     # Scanners.
 
@@ -770,14 +770,14 @@ class Scanner:
         # `unwind_indent` before issuing BLOCK-END.
         # Scanners for block, flow, and plain scalars need to be modified.
 
-        if self.index == 0 and self.peek() == '\uFEFF':
+        if self.index == 0 and self.peek() == u'\uFEFF':
             self.forward()
         found = False
         while not found:
-            while self.peek() == ' ':
+            while self.peek() == u' ':
                 self.forward()
-            if self.peek() == '#':
-                while self.peek() not in '\0\r\n\x85\u2028\u2029':
+            if self.peek() == u'#':
+                while self.peek() not in u'\0\r\n\x85\u2028\u2029':
                     self.forward()
             if self.scan_line_break():
                 if not self.flow_level:
@@ -791,15 +791,15 @@ class Scanner:
         self.forward()
         name = self.scan_directive_name(start_mark)
         value = None
-        if name == 'YAML':
+        if name == u'YAML':
             value = self.scan_yaml_directive_value(start_mark)
             end_mark = self.get_mark()
-        elif name == 'TAG':
+        elif name == u'TAG':
             value = self.scan_tag_directive_value(start_mark)
             end_mark = self.get_mark()
         else:
             end_mark = self.get_mark()
-            while self.peek() not in '\0\r\n\x85\u2028\u2029':
+            while self.peek() not in u'\0\r\n\x85\u2028\u2029':
                 self.forward()
         self.scan_directive_ignored_line(start_mark)
         return DirectiveToken(name, value, start_mark, end_mark)
@@ -808,48 +808,51 @@ class Scanner:
         # See the specification for details.
         length = 0
         ch = self.peek(length)
-        while '0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z'  \
-                or ch in '-_':
+        while u'0' <= ch <= u'9' or u'A' <= ch <= u'Z' or u'a' <= ch <= u'z'    \
+                or ch in u'-_':
             length += 1
             ch = self.peek(length)
         if not length:
             raise ScannerError("while scanning a directive", start_mark,
                     "expected alphabetic or numeric character, but found %r"
-                    % ch, self.get_mark())
+                    % ch.encode('utf-8'), self.get_mark())
         value = self.prefix(length)
         self.forward(length)
         ch = self.peek()
-        if ch not in '\0 \r\n\x85\u2028\u2029':
+        if ch not in u'\0 \r\n\x85\u2028\u2029':
             raise ScannerError("while scanning a directive", start_mark,
                     "expected alphabetic or numeric character, but found %r"
-                    % ch, self.get_mark())
+                    % ch.encode('utf-8'), self.get_mark())
         return value
 
     def scan_yaml_directive_value(self, start_mark):
         # See the specification for details.
-        while self.peek() == ' ':
+        while self.peek() == u' ':
             self.forward()
         major = self.scan_yaml_directive_number(start_mark)
         if self.peek() != '.':
             raise ScannerError("while scanning a directive", start_mark,
-                    "expected a digit or '.', but found %r" % self.peek(),
+                    "expected a digit or '.', but found %r"
+                    % self.peek().encode('utf-8'),
                     self.get_mark())
         self.forward()
         minor = self.scan_yaml_directive_number(start_mark)
-        if self.peek() not in '\0 \r\n\x85\u2028\u2029':
+        if self.peek() not in u'\0 \r\n\x85\u2028\u2029':
             raise ScannerError("while scanning a directive", start_mark,
-                    "expected a digit or ' ', but found %r" % self.peek(),
+                    "expected a digit or ' ', but found %r"
+                    % self.peek().encode('utf-8'),
                     self.get_mark())
         return (major, minor)
 
     def scan_yaml_directive_number(self, start_mark):
         # See the specification for details.
         ch = self.peek()
-        if not ('0' <= ch <= '9'):
+        if not (u'0' <= ch <= u'9'):
             raise ScannerError("while scanning a directive", start_mark,
-                    "expected a digit, but found %r" % ch, self.get_mark())
+                    "expected a digit, but found %r" % ch.encode('utf-8'),
+                    self.get_mark())
         length = 0
-        while '0' <= self.peek(length) <= '9':
+        while u'0' <= self.peek(length) <= u'9':
             length += 1
         value = int(self.prefix(length))
         self.forward(length)
@@ -857,10 +860,10 @@ class Scanner:
 
     def scan_tag_directive_value(self, start_mark):
         # See the specification for details.
-        while self.peek() == ' ':
+        while self.peek() == u' ':
             self.forward()
         handle = self.scan_tag_directive_handle(start_mark)
-        while self.peek() == ' ':
+        while self.peek() == u' ':
             self.forward()
         prefix = self.scan_tag_directive_prefix(start_mark)
         return (handle, prefix)
@@ -869,32 +872,34 @@ class Scanner:
         # See the specification for details.
         value = self.scan_tag_handle('directive', start_mark)
         ch = self.peek()
-        if ch != ' ':
+        if ch != u' ':
             raise ScannerError("while scanning a directive", start_mark,
-                    "expected ' ', but found %r" % ch, self.get_mark())
+                    "expected ' ', but found %r" % ch.encode('utf-8'),
+                    self.get_mark())
         return value
 
     def scan_tag_directive_prefix(self, start_mark):
         # See the specification for details.
         value = self.scan_tag_uri('directive', start_mark)
         ch = self.peek()
-        if ch not in '\0 \r\n\x85\u2028\u2029':
+        if ch not in u'\0 \r\n\x85\u2028\u2029':
             raise ScannerError("while scanning a directive", start_mark,
-                    "expected ' ', but found %r" % ch, self.get_mark())
+                    "expected ' ', but found %r" % ch.encode('utf-8'),
+                    self.get_mark())
         return value
 
     def scan_directive_ignored_line(self, start_mark):
         # See the specification for details.
-        while self.peek() == ' ':
+        while self.peek() == u' ':
             self.forward()
-        if self.peek() == '#':
-            while self.peek() not in '\0\r\n\x85\u2028\u2029':
+        if self.peek() == u'#':
+            while self.peek() not in u'\0\r\n\x85\u2028\u2029':
                 self.forward()
         ch = self.peek()
-        if ch not in '\0\r\n\x85\u2028\u2029':
+        if ch not in u'\0\r\n\x85\u2028\u2029':
             raise ScannerError("while scanning a directive", start_mark,
                     "expected a comment or a line break, but found %r"
-                        % ch, self.get_mark())
+                        % ch.encode('utf-8'), self.get_mark())
         self.scan_line_break()
 
     def scan_anchor(self, TokenClass):
@@ -908,28 +913,28 @@ class Scanner:
         # Therefore we restrict aliases to numbers and ASCII letters.
         start_mark = self.get_mark()
         indicator = self.peek()
-        if indicator == '*':
+        if indicator == u'*':
             name = 'alias'
         else:
             name = 'anchor'
         self.forward()
         length = 0
         ch = self.peek(length)
-        while '0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z'  \
-                or ch in '-_':
+        while u'0' <= ch <= u'9' or u'A' <= ch <= u'Z' or u'a' <= ch <= u'z'    \
+                or ch in u'-_':
             length += 1
             ch = self.peek(length)
         if not length:
             raise ScannerError("while scanning an %s" % name, start_mark,
                     "expected alphabetic or numeric character, but found %r"
-                    % ch, self.get_mark())
+                    % ch.encode('utf-8'), self.get_mark())
         value = self.prefix(length)
         self.forward(length)
         ch = self.peek()
-        if ch not in '\0 \t\r\n\x85\u2028\u2029?:,]}%@`':
+        if ch not in u'\0 \t\r\n\x85\u2028\u2029?:,]}%@`':
             raise ScannerError("while scanning an %s" % name, start_mark,
                     "expected alphabetic or numeric character, but found %r"
-                    % ch, self.get_mark())
+                    % ch.encode('utf-8'), self.get_mark())
         end_mark = self.get_mark()
         return TokenClass(value, start_mark, end_mark)
 
@@ -937,39 +942,40 @@ class Scanner:
         # See the specification for details.
         start_mark = self.get_mark()
         ch = self.peek(1)
-        if ch == '<':
+        if ch == u'<':
             handle = None
             self.forward(2)
             suffix = self.scan_tag_uri('tag', start_mark)
-            if self.peek() != '>':
+            if self.peek() != u'>':
                 raise ScannerError("while parsing a tag", start_mark,
-                        "expected '>', but found %r" % self.peek(),
+                        "expected '>', but found %r" % self.peek().encode('utf-8'),
                         self.get_mark())
             self.forward()
-        elif ch in '\0 \t\r\n\x85\u2028\u2029':
+        elif ch in u'\0 \t\r\n\x85\u2028\u2029':
             handle = None
-            suffix = '!'
+            suffix = u'!'
             self.forward()
         else:
             length = 1
             use_handle = False
-            while ch not in '\0 \r\n\x85\u2028\u2029':
-                if ch == '!':
+            while ch not in u'\0 \r\n\x85\u2028\u2029':
+                if ch == u'!':
                     use_handle = True
                     break
                 length += 1
                 ch = self.peek(length)
-            handle = '!'
+            handle = u'!'
             if use_handle:
                 handle = self.scan_tag_handle('tag', start_mark)
             else:
-                handle = '!'
+                handle = u'!'
                 self.forward()
             suffix = self.scan_tag_uri('tag', start_mark)
         ch = self.peek()
-        if ch not in '\0 \r\n\x85\u2028\u2029':
+        if ch not in u'\0 \r\n\x85\u2028\u2029':
             raise ScannerError("while scanning a tag", start_mark,
-                    "expected ' ', but found %r" % ch, self.get_mark())
+                    "expected ' ', but found %r" % ch.encode('utf-8'),
+                    self.get_mark())
         value = (handle, suffix)
         end_mark = self.get_mark()
         return TagToken(value, start_mark, end_mark)
@@ -1000,39 +1006,39 @@ class Scanner:
         else:
             indent = min_indent+increment-1
             breaks, end_mark = self.scan_block_scalar_breaks(indent)
-        line_break = ''
+        line_break = u''
 
         # Scan the inner part of the block scalar.
-        while self.column == indent and self.peek() != '\0':
+        while self.column == indent and self.peek() != u'\0':
             chunks.extend(breaks)
-            leading_non_space = self.peek() not in ' \t'
+            leading_non_space = self.peek() not in u' \t'
             length = 0
-            while self.peek(length) not in '\0\r\n\x85\u2028\u2029':
+            while self.peek(length) not in u'\0\r\n\x85\u2028\u2029':
                 length += 1
             chunks.append(self.prefix(length))
             self.forward(length)
             line_break = self.scan_line_break()
             breaks, end_mark = self.scan_block_scalar_breaks(indent)
-            if self.column == indent and self.peek() != '\0':
+            if self.column == indent and self.peek() != u'\0':
 
                 # Unfortunately, folding rules are ambiguous.
                 #
                 # This is the folding according to the specification:
                 
-                if folded and line_break == '\n'    \
-                        and leading_non_space and self.peek() not in ' \t':
+                if folded and line_break == u'\n'   \
+                        and leading_non_space and self.peek() not in u' \t':
                     if not breaks:
-                        chunks.append(' ')
+                        chunks.append(u' ')
                 else:
                     chunks.append(line_break)
                 
                 # This is Clark Evans's interpretation (also in the spec
                 # examples):
                 #
-                #if folded and line_break == '\n':
+                #if folded and line_break == u'\n':
                 #    if not breaks:
                 #        if self.peek() not in ' \t':
-                #            chunks.append(' ')
+                #            chunks.append(u' ')
                 #        else:
                 #            chunks.append(line_break)
                 #else:
@@ -1047,7 +1053,7 @@ class Scanner:
             chunks.extend(breaks)
 
         # We are done.
-        return ScalarToken(''.join(chunks), False, start_mark, end_mark,
+        return ScalarToken(u''.join(chunks), False, start_mark, end_mark,
                 style)
 
     def scan_block_scalar_indicators(self, start_mark):
@@ -1055,21 +1061,21 @@ class Scanner:
         chomping = None
         increment = None
         ch = self.peek()
-        if ch in '+-':
+        if ch in u'+-':
             if ch == '+':
                 chomping = True
             else:
                 chomping = False
             self.forward()
             ch = self.peek()
-            if ch in '0123456789':
+            if ch in u'0123456789':
                 increment = int(ch)
                 if increment == 0:
                     raise ScannerError("while scanning a block scalar", start_mark,
                             "expected indentation indicator in the range 1-9, but found 0",
                             self.get_mark())
                 self.forward()
-        elif ch in '0123456789':
+        elif ch in u'0123456789':
             increment = int(ch)
             if increment == 0:
                 raise ScannerError("while scanning a block scalar", start_mark,
@@ -1077,31 +1083,31 @@ class Scanner:
                         self.get_mark())
             self.forward()
             ch = self.peek()
-            if ch in '+-':
+            if ch in u'+-':
                 if ch == '+':
                     chomping = True
                 else:
                     chomping = False
                 self.forward()
         ch = self.peek()
-        if ch not in '\0 \r\n\x85\u2028\u2029':
+        if ch not in u'\0 \r\n\x85\u2028\u2029':
             raise ScannerError("while scanning a block scalar", start_mark,
                     "expected chomping or indentation indicators, but found %r"
-                    % ch, self.get_mark())
+                        % ch.encode('utf-8'), self.get_mark())
         return chomping, increment
 
     def scan_block_scalar_ignored_line(self, start_mark):
         # See the specification for details.
-        while self.peek() == ' ':
+        while self.peek() == u' ':
             self.forward()
-        if self.peek() == '#':
-            while self.peek() not in '\0\r\n\x85\u2028\u2029':
+        if self.peek() == u'#':
+            while self.peek() not in u'\0\r\n\x85\u2028\u2029':
                 self.forward()
         ch = self.peek()
-        if ch not in '\0\r\n\x85\u2028\u2029':
+        if ch not in u'\0\r\n\x85\u2028\u2029':
             raise ScannerError("while scanning a block scalar", start_mark,
-                    "expected a comment or a line break, but found %r" % ch,
-                    self.get_mark())
+                    "expected a comment or a line break, but found %r"
+                        % ch.encode('utf-8'), self.get_mark())
         self.scan_line_break()
 
     def scan_block_scalar_indentation(self):
@@ -1109,8 +1115,8 @@ class Scanner:
         chunks = []
         max_indent = 0
         end_mark = self.get_mark()
-        while self.peek() in ' \r\n\x85\u2028\u2029':
-            if self.peek() != ' ':
+        while self.peek() in u' \r\n\x85\u2028\u2029':
+            if self.peek() != u' ':
                 chunks.append(self.scan_line_break())
                 end_mark = self.get_mark()
             else:
@@ -1123,12 +1129,12 @@ class Scanner:
         # See the specification for details.
         chunks = []
         end_mark = self.get_mark()
-        while self.column < indent and self.peek() == ' ':
+        while self.column < indent and self.peek() == u' ':
             self.forward()
-        while self.peek() in '\r\n\x85\u2028\u2029':
+        while self.peek() in u'\r\n\x85\u2028\u2029':
             chunks.append(self.scan_line_break())
             end_mark = self.get_mark()
-            while self.column < indent and self.peek() == ' ':
+            while self.column < indent and self.peek() == u' ':
                 self.forward()
         return chunks, end_mark
 
@@ -1153,33 +1159,33 @@ class Scanner:
             chunks.extend(self.scan_flow_scalar_non_spaces(double, start_mark))
         self.forward()
         end_mark = self.get_mark()
-        return ScalarToken(''.join(chunks), False, start_mark, end_mark,
+        return ScalarToken(u''.join(chunks), False, start_mark, end_mark,
                 style)
 
     ESCAPE_REPLACEMENTS = {
-        '0':    '\0',
-        'a':    '\x07',
-        'b':    '\x08',
-        't':    '\x09',
-        '\t':   '\x09',
-        'n':    '\x0A',
-        'v':    '\x0B',
-        'f':    '\x0C',
-        'r':    '\x0D',
-        'e':    '\x1B',
-        ' ':    '\x20',
-        '\"':   '\"',
-        '\\':   '\\',
-        'N':    '\x85',
-        '_':    '\xA0',
-        'L':    '\u2028',
-        'P':    '\u2029',
+        u'0':   u'\0',
+        u'a':   u'\x07',
+        u'b':   u'\x08',
+        u't':   u'\x09',
+        u'\t':  u'\x09',
+        u'n':   u'\x0A',
+        u'v':   u'\x0B',
+        u'f':   u'\x0C',
+        u'r':   u'\x0D',
+        u'e':   u'\x1B',
+        u' ':   u'\x20',
+        u'\"':  u'\"',
+        u'\\':  u'\\',
+        u'N':   u'\x85',
+        u'_':   u'\xA0',
+        u'L':   u'\u2028',
+        u'P':   u'\u2029',
     }
 
     ESCAPE_CODES = {
-        'x':    2,
-        'u':    4,
-        'U':    8,
+        u'x':   2,
+        u'u':   4,
+        u'U':   8,
     }
 
     def scan_flow_scalar_non_spaces(self, double, start_mark):
@@ -1187,19 +1193,19 @@ class Scanner:
         chunks = []
         while True:
             length = 0
-            while self.peek(length) not in '\'\"\\\0 \t\r\n\x85\u2028\u2029':
+            while self.peek(length) not in u'\'\"\\\0 \t\r\n\x85\u2028\u2029':
                 length += 1
             if length:
                 chunks.append(self.prefix(length))
                 self.forward(length)
             ch = self.peek()
-            if not double and ch == '\'' and self.peek(1) == '\'':
-                chunks.append('\'')
+            if not double and ch == u'\'' and self.peek(1) == u'\'':
+                chunks.append(u'\'')
                 self.forward(2)
-            elif (double and ch == '\'') or (not double and ch in '\"\\'):
+            elif (double and ch == u'\'') or (not double and ch in u'\"\\'):
                 chunks.append(ch)
                 self.forward()
-            elif double and ch == '\\':
+            elif double and ch == u'\\':
                 self.forward()
                 ch = self.peek()
                 if ch in self.ESCAPE_REPLACEMENTS:
@@ -1209,19 +1215,19 @@ class Scanner:
                     length = self.ESCAPE_CODES[ch]
                     self.forward()
                     for k in range(length):
-                        if self.peek(k) not in '0123456789ABCDEFabcdef':
+                        if self.peek(k) not in u'0123456789ABCDEFabcdef':
                             raise ScannerError("while scanning a double-quoted scalar", start_mark,
                                     "expected escape sequence of %d hexdecimal numbers, but found %r" %
-                                        (length, self.peek(k)), self.get_mark())
+                                        (length, self.peek(k).encode('utf-8')), self.get_mark())
                     code = int(self.prefix(length), 16)
-                    chunks.append(chr(code))
+                    chunks.append(unichr(code))
                     self.forward(length)
-                elif ch in '\r\n\x85\u2028\u2029':
+                elif ch in u'\r\n\x85\u2028\u2029':
                     self.scan_line_break()
                     chunks.extend(self.scan_flow_scalar_breaks(double, start_mark))
                 else:
                     raise ScannerError("while scanning a double-quoted scalar", start_mark,
-                            "found unknown escape character %r" % ch, self.get_mark())
+                            "found unknown escape character %r" % ch.encode('utf-8'), self.get_mark())
             else:
                 return chunks
 
@@ -1229,21 +1235,21 @@ class Scanner:
         # See the specification for details.
         chunks = []
         length = 0
-        while self.peek(length) in ' \t':
+        while self.peek(length) in u' \t':
             length += 1
         whitespaces = self.prefix(length)
         self.forward(length)
         ch = self.peek()
-        if ch == '\0':
+        if ch == u'\0':
             raise ScannerError("while scanning a quoted scalar", start_mark,
                     "found unexpected end of stream", self.get_mark())
-        elif ch in '\r\n\x85\u2028\u2029':
+        elif ch in u'\r\n\x85\u2028\u2029':
             line_break = self.scan_line_break()
             breaks = self.scan_flow_scalar_breaks(double, start_mark)
-            if line_break != '\n':
+            if line_break != u'\n':
                 chunks.append(line_break)
             elif not breaks:
-                chunks.append(' ')
+                chunks.append(u' ')
             chunks.extend(breaks)
         else:
             chunks.append(whitespaces)
@@ -1256,13 +1262,13 @@ class Scanner:
             # Instead of checking indentation, we check for document
             # separators.
             prefix = self.prefix(3)
-            if (prefix == '---' or prefix == '...')   \
-                    and self.peek(3) in '\0 \t\r\n\x85\u2028\u2029':
+            if (prefix == u'---' or prefix == u'...')   \
+                    and self.peek(3) in u'\0 \t\r\n\x85\u2028\u2029':
                 raise ScannerError("while scanning a quoted scalar", start_mark,
                         "found unexpected document separator", self.get_mark())
-            while self.peek() in ' \t':
+            while self.peek() in u' \t':
                 self.forward()
-            if self.peek() in '\r\n\x85\u2028\u2029':
+            if self.peek() in u'\r\n\x85\u2028\u2029':
                 chunks.append(self.scan_line_break())
             else:
                 return chunks
@@ -1284,19 +1290,19 @@ class Scanner:
         spaces = []
         while True:
             length = 0
-            if self.peek() == '#':
+            if self.peek() == u'#':
                 break
             while True:
                 ch = self.peek(length)
-                if ch in '\0 \t\r\n\x85\u2028\u2029'    \
-                        or (not self.flow_level and ch == ':' and
-                                self.peek(length+1) in '\0 \t\r\n\x85\u2028\u2029') \
-                        or (self.flow_level and ch in ',:?[]{}'):
+                if ch in u'\0 \t\r\n\x85\u2028\u2029'   \
+                        or (not self.flow_level and ch == u':' and
+                                self.peek(length+1) in u'\0 \t\r\n\x85\u2028\u2029') \
+                        or (self.flow_level and ch in u',:?[]{}'):
                     break
                 length += 1
             # It's not clear what we should do with ':' in the flow context.
-            if (self.flow_level and ch == ':'
-                    and self.peek(length+1) not in '\0 \t\r\n\x85\u2028\u2029,[]{}'):
+            if (self.flow_level and ch == u':'
+                    and self.peek(length+1) not in u'\0 \t\r\n\x85\u2028\u2029,[]{}'):
                 self.forward(length)
                 raise ScannerError("while scanning a plain scalar", start_mark,
                     "found unexpected ':'", self.get_mark(),
@@ -1309,10 +1315,10 @@ class Scanner:
             self.forward(length)
             end_mark = self.get_mark()
             spaces = self.scan_plain_spaces(indent, start_mark)
-            if not spaces or self.peek() == '#' \
+            if not spaces or self.peek() == u'#' \
                     or (not self.flow_level and self.column < indent):
                 break
-        return ScalarToken(''.join(chunks), True, start_mark, end_mark)
+        return ScalarToken(u''.join(chunks), True, start_mark, end_mark)
 
     def scan_plain_spaces(self, indent, start_mark):
         # See the specification for details.
@@ -1320,32 +1326,32 @@ class Scanner:
         # We just forbid them completely. Do not use tabs in YAML!
         chunks = []
         length = 0
-        while self.peek(length) in ' ':
+        while self.peek(length) in u' ':
             length += 1
         whitespaces = self.prefix(length)
         self.forward(length)
         ch = self.peek()
-        if ch in '\r\n\x85\u2028\u2029':
+        if ch in u'\r\n\x85\u2028\u2029':
             line_break = self.scan_line_break()
             self.allow_simple_key = True
             prefix = self.prefix(3)
-            if (prefix == '---' or prefix == '...')   \
-                    and self.peek(3) in '\0 \t\r\n\x85\u2028\u2029':
+            if (prefix == u'---' or prefix == u'...')   \
+                    and self.peek(3) in u'\0 \t\r\n\x85\u2028\u2029':
                 return
             breaks = []
-            while self.peek() in ' \r\n\x85\u2028\u2029':
+            while self.peek() in u' \r\n\x85\u2028\u2029':
                 if self.peek() == ' ':
                     self.forward()
                 else:
                     breaks.append(self.scan_line_break())
                     prefix = self.prefix(3)
-                    if (prefix == '---' or prefix == '...')   \
-                            and self.peek(3) in '\0 \t\r\n\x85\u2028\u2029':
+                    if (prefix == u'---' or prefix == u'...')   \
+                            and self.peek(3) in u'\0 \t\r\n\x85\u2028\u2029':
                         return
-            if line_break != '\n':
+            if line_break != u'\n':
                 chunks.append(line_break)
             elif not breaks:
-                chunks.append(' ')
+                chunks.append(u' ')
             chunks.extend(breaks)
         elif whitespaces:
             chunks.append(whitespaces)
@@ -1356,20 +1362,22 @@ class Scanner:
         # For some strange reasons, the specification does not allow '_' in
         # tag handles. I have allowed it anyway.
         ch = self.peek()
-        if ch != '!':
+        if ch != u'!':
             raise ScannerError("while scanning a %s" % name, start_mark,
-                    "expected '!', but found %r" % ch, self.get_mark())
+                    "expected '!', but found %r" % ch.encode('utf-8'),
+                    self.get_mark())
         length = 1
         ch = self.peek(length)
-        if ch != ' ':
-            while '0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z'  \
-                    or ch in '-_':
+        if ch != u' ':
+            while u'0' <= ch <= u'9' or u'A' <= ch <= u'Z' or u'a' <= ch <= u'z'    \
+                    or ch in u'-_':
                 length += 1
                 ch = self.peek(length)
-            if ch != '!':
+            if ch != u'!':
                 self.forward(length)
                 raise ScannerError("while scanning a %s" % name, start_mark,
-                        "expected '!', but found %r" % ch, self.get_mark())
+                        "expected '!', but found %r" % ch.encode('utf-8'),
+                        self.get_mark())
             length += 1
         value = self.prefix(length)
         self.forward(length)
@@ -1381,9 +1389,9 @@ class Scanner:
         chunks = []
         length = 0
         ch = self.peek(length)
-        while '0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z'  \
-                or ch in '-;/?:@&=+$,_.!~*\'()[]%':
-            if ch == '%':
+        while u'0' <= ch <= u'9' or u'A' <= ch <= u'Z' or u'a' <= ch <= u'z'    \
+                or ch in u'-;/?:@&=+$,_.!~*\'()[]%':
+            if ch == u'%':
                 chunks.append(self.prefix(length))
                 self.forward(length)
                 length = 0
@@ -1397,25 +1405,26 @@ class Scanner:
             length = 0
         if not chunks:
             raise ScannerError("while parsing a %s" % name, start_mark,
-                    "expected URI, but found %r" % ch, self.get_mark())
-        return ''.join(chunks)
+                    "expected URI, but found %r" % ch.encode('utf-8'),
+                    self.get_mark())
+        return u''.join(chunks)
 
     def scan_uri_escapes(self, name, start_mark):
         # See the specification for details.
-        codes = []
+        bytes = []
         mark = self.get_mark()
-        while self.peek() == '%':
+        while self.peek() == u'%':
             self.forward()
             for k in range(2):
-                if self.peek(k) not in '0123456789ABCDEFabcdef':
+                if self.peek(k) not in u'0123456789ABCDEFabcdef':
                     raise ScannerError("while scanning a %s" % name, start_mark,
-                            "expected URI escape sequence of 2 hexdecimal numbers, but found %r"
-                            % self.peek(k), self.get_mark())
-            codes.append(int(self.prefix(2), 16))
+                            "expected URI escape sequence of 2 hexdecimal numbers, but found %r" %
+                                (self.peek(k).encode('utf-8')), self.get_mark())
+            bytes.append(chr(int(self.prefix(2), 16)))
             self.forward(2)
         try:
-            value = bytes(codes).decode('utf-8')
-        except UnicodeDecodeError as exc:
+            value = unicode(''.join(bytes), 'utf-8')
+        except UnicodeDecodeError, exc:
             raise ScannerError("while scanning a %s" % name, start_mark, str(exc), mark)
         return value
 
@@ -1429,16 +1438,16 @@ class Scanner:
         #   '\u2029     :   '\u2029'
         #   default     :   ''
         ch = self.peek()
-        if ch in '\r\n\x85':
-            if self.prefix(2) == '\r\n':
+        if ch in u'\r\n\x85':
+            if self.prefix(2) == u'\r\n':
                 self.forward(2)
             else:
                 self.forward()
-            return '\n'
-        elif ch in '\u2028\u2029':
+            return u'\n'
+        elif ch in u'\u2028\u2029':
             self.forward()
             return ch
-        return ''
+        return u''
 
 #try:
 #    import psyco

tablib/packages/yaml/serializer.py

@@ -1,16 +1,16 @@
 
 __all__ = ['Serializer', 'SerializerError']
 
-from .error import YAMLError
-from .events import *
-from .nodes import *
+from error import YAMLError
+from events import *
+from nodes import *
 
 class SerializerError(YAMLError):
     pass
 
-class Serializer:
+class Serializer(object):
 
-    ANCHOR_TEMPLATE = 'id%03d'
+    ANCHOR_TEMPLATE = u'id%03d'
 
     def __init__(self, encoding=None,
             explicit_start=None, explicit_end=None, version=None, tags=None):

tablib/packages/yaml3/__init__.py

@@ -0,0 +1,288 @@
+
+from .error import *
+
+from .tokens import *
+from .events import *
+from .nodes import *
+
+from .loader import *
+from .dumper import *
+
+__version__ = '3.09'
+
+try:
+    from cyaml import *
+    __with_libyaml__ = True
+except ImportError:
+    __with_libyaml__ = False
+
+def scan(stream, Loader=Loader):
+    """
+    Scan a YAML stream and produce scanning tokens.
+    """
+    loader = Loader(stream)
+    while loader.check_token():
+        yield loader.get_token()
+
+def parse(stream, Loader=Loader):
+    """
+    Parse a YAML stream and produce parsing events.
+    """
+    loader = Loader(stream)
+    while loader.check_event():
+        yield loader.get_event()
+
+def compose(stream, Loader=Loader):
+    """
+    Parse the first YAML document in a stream
+    and produce the corresponding representation tree.
+    """
+    loader = Loader(stream)
+    return loader.get_single_node()
+
+def compose_all(stream, Loader=Loader):
+    """
+    Parse all YAML documents in a stream
+    and produce corresponding representation trees.
+    """
+    loader = Loader(stream)
+    while loader.check_node():
+        yield loader.get_node()
+
+def load(stream, Loader=Loader):
+    """
+    Parse the first YAML document in a stream
+    and produce the corresponding Python object.
+    """
+    loader = Loader(stream)
+    return loader.get_single_data()
+
+def load_all(stream, Loader=Loader):
+    """
+    Parse all YAML documents in a stream
+    and produce corresponding Python objects.
+    """
+    loader = Loader(stream)
+    while loader.check_data():
+        yield loader.get_data()
+
+def safe_load(stream):
+    """
+    Parse the first YAML document in a stream
+    and produce the corresponding Python object.
+    Resolve only basic YAML tags.
+    """
+    return load(stream, SafeLoader)
+
+def safe_load_all(stream):
+    """
+    Parse all YAML documents in a stream
+    and produce corresponding Python objects.
+    Resolve only basic YAML tags.
+    """
+    return load_all(stream, SafeLoader)
+
+def emit(events, stream=None, Dumper=Dumper,
+        canonical=None, indent=None, width=None,
+        allow_unicode=None, line_break=None):
+    """
+    Emit YAML parsing events into a stream.
+    If stream is None, return the produced string instead.
+    """
+    getvalue = None
+    if stream is None:
+        from StringIO import StringIO
+        stream = StringIO()
+        getvalue = stream.getvalue
+    dumper = Dumper(stream, canonical=canonical, indent=indent, width=width,
+            allow_unicode=allow_unicode, line_break=line_break)
+    for event in events:
+        dumper.emit(event)
+    if getvalue:
+        return getvalue()
+
+def serialize_all(nodes, stream=None, Dumper=Dumper,
+        canonical=None, indent=None, width=None,
+        allow_unicode=None, line_break=None,
+        encoding='utf-8', explicit_start=None, explicit_end=None,
+        version=None, tags=None):
+    """
+    Serialize a sequence of representation trees into a YAML stream.
+    If stream is None, return the produced string instead.
+    """
+    getvalue = None
+    if stream is None:
+        if encoding is None:
+            from StringIO import StringIO
+        else:
+            from cStringIO import StringIO
+        stream = StringIO()
+        getvalue = stream.getvalue
+    dumper = Dumper(stream, canonical=canonical, indent=indent, width=width,
+            allow_unicode=allow_unicode, line_break=line_break,
+            encoding=encoding, version=version, tags=tags,
+            explicit_start=explicit_start, explicit_end=explicit_end)
+    dumper.open()
+    for node in nodes:
+        dumper.serialize(node)
+    dumper.close()
+    if getvalue:
+        return getvalue()
+
+def serialize(node, stream=None, Dumper=Dumper, **kwds):
+    """
+    Serialize a representation tree into a YAML stream.
+    If stream is None, return the produced string instead.
+    """
+    return serialize_all([node], stream, Dumper=Dumper, **kwds)
+
+def dump_all(documents, stream=None, Dumper=Dumper,
+        default_style=None, default_flow_style=None,
+        canonical=None, indent=None, width=None,
+        allow_unicode=None, line_break=None,
+        encoding='utf-8', explicit_start=None, explicit_end=None,
+        version=None, tags=None):
+    """
+    Serialize a sequence of Python objects into a YAML stream.
+    If stream is None, return the produced string instead.
+    """
+    getvalue = None
+    if stream is None:
+        if encoding is None:
+            from io import StringIO
+        else:
+            from io import StringIO
+        stream = StringIO()
+        getvalue = stream.getvalue
+    dumper = Dumper(stream, default_style=default_style,
+            default_flow_style=default_flow_style,
+            canonical=canonical, indent=indent, width=width,
+            allow_unicode=allow_unicode, line_break=line_break,
+            encoding=encoding, version=version, tags=tags,
+            explicit_start=explicit_start, explicit_end=explicit_end)
+    dumper.open()
+    for data in documents:
+        dumper.represent(data)
+    dumper.close()
+    if getvalue:
+        return getvalue()
+
+def dump(data, stream=None, Dumper=Dumper, **kwds):
+    """
+    Serialize a Python object into a YAML stream.
+    If stream is None, return the produced string instead.
+    """
+    return dump_all([data], stream, Dumper=Dumper, **kwds)
+
+def safe_dump_all(documents, stream=None, **kwds):
+    """
+    Serialize a sequence of Python objects into a YAML stream.
+    Produce only basic YAML tags.
+    If stream is None, return the produced string instead.
+    """
+    return dump_all(documents, stream, Dumper=SafeDumper, **kwds)
+
+def safe_dump(data, stream=None, **kwds):
+    """
+    Serialize a Python object into a YAML stream.
+    Produce only basic YAML tags.
+    If stream is None, return the produced string instead.
+    """
+    return dump_all([data], stream, Dumper=SafeDumper, **kwds)
+
+def add_implicit_resolver(tag, regexp, first=None,
+        Loader=Loader, Dumper=Dumper):
+    """
+    Add an implicit scalar detector.
+    If an implicit scalar value matches the given regexp,
+    the corresponding tag is assigned to the scalar.
+    first is a sequence of possible initial characters or None.
+    """
+    Loader.add_implicit_resolver(tag, regexp, first)
+    Dumper.add_implicit_resolver(tag, regexp, first)
+
+def add_path_resolver(tag, path, kind=None, Loader=Loader, Dumper=Dumper):
+    """
+    Add a path based resolver for the given tag.
+    A path is a list of keys that forms a path
+    to a node in the representation tree.
+    Keys can be string values, integers, or None.
+    """
+    Loader.add_path_resolver(tag, path, kind)
+    Dumper.add_path_resolver(tag, path, kind)
+
+def add_constructor(tag, constructor, Loader=Loader):
+    """
+    Add a constructor for the given tag.
+    Constructor is a function that accepts a Loader instance
+    and a node object and produces the corresponding Python object.
+    """
+    Loader.add_constructor(tag, constructor)
+
+def add_multi_constructor(tag_prefix, multi_constructor, Loader=Loader):
+    """
+    Add a multi-constructor for the given tag prefix.
+    Multi-constructor is called for a node if its tag starts with tag_prefix.
+    Multi-constructor accepts a Loader instance, a tag suffix,
+    and a node object and produces the corresponding Python object.
+    """
+    Loader.add_multi_constructor(tag_prefix, multi_constructor)
+
+def add_representer(data_type, representer, Dumper=Dumper):
+    """
+    Add a representer for the given type.
+    Representer is a function accepting a Dumper instance
+    and an instance of the given data type
+    and producing the corresponding representation node.
+    """
+    Dumper.add_representer(data_type, representer)
+
+def add_multi_representer(data_type, multi_representer, Dumper=Dumper):
+    """
+    Add a representer for the given type.
+    Multi-representer is a function accepting a Dumper instance
+    and an instance of the given data type or subtype
+    and producing the corresponding representation node.
+    """
+    Dumper.add_multi_representer(data_type, multi_representer)
+
+class YAMLObjectMetaclass(type):
+    """
+    The metaclass for YAMLObject.
+    """
+    def __init__(cls, name, bases, kwds):
+        super(YAMLObjectMetaclass, cls).__init__(name, bases, kwds)
+        if 'yaml_tag' in kwds and kwds['yaml_tag'] is not None:
+            cls.yaml_loader.add_constructor(cls.yaml_tag, cls.from_yaml)
+            cls.yaml_dumper.add_representer(cls, cls.to_yaml)
+
+class YAMLObject(object):
+    """
+    An object that can dump itself to a YAML stream
+    and load itself from a YAML stream.
+    """
+
+    __metaclass__ = YAMLObjectMetaclass
+    __slots__ = ()  # no direct instantiation, so allow immutable subclasses
+
+    yaml_loader = Loader
+    yaml_dumper = Dumper
+
+    yaml_tag = None
+    yaml_flow_style = None
+
+    def from_yaml(cls, loader, node):
+        """
+        Convert a representation node to a Python object.
+        """
+        return loader.construct_yaml_object(node, cls)
+    from_yaml = classmethod(from_yaml)
+
+    def to_yaml(cls, dumper, data):
+        """
+        Convert a Python object to a representation node.
+        """
+        return dumper.represent_yaml_object(cls.yaml_tag, data, cls,
+                flow_style=cls.yaml_flow_style)
+    to_yaml = classmethod(to_yaml)
+

tablib/packages/yaml3/composer.py

@@ -0,0 +1,139 @@
+
+__all__ = ['Composer', 'ComposerError']
+
+from .error import MarkedYAMLError
+from .events import *
+from .nodes import *
+
+class ComposerError(MarkedYAMLError):
+    pass
+
+class Composer:
+
+    def __init__(self):
+        self.anchors = {}
+
+    def check_node(self):
+        # Drop the STREAM-START event.
+        if self.check_event(StreamStartEvent):
+            self.get_event()
+
+        # If there are more documents available?
+        return not self.check_event(StreamEndEvent)
+
+    def get_node(self):
+        # Get the root node of the next document.
+        if not self.check_event(StreamEndEvent):
+            return self.compose_document()
+
+    def get_single_node(self):
+        # Drop the STREAM-START event.
+        self.get_event()
+
+        # Compose a document if the stream is not empty.
+        document = None
+        if not self.check_event(StreamEndEvent):
+            document = self.compose_document()
+
+        # Ensure that the stream contains no more documents.
+        if not self.check_event(StreamEndEvent):
+            event = self.get_event()
+            raise ComposerError("expected a single document in the stream",
+                    document.start_mark, "but found another document",
+                    event.start_mark)
+
+        # Drop the STREAM-END event.
+        self.get_event()
+
+        return document
+
+    def compose_document(self):
+        # Drop the DOCUMENT-START event.
+        self.get_event()
+
+        # Compose the root node.
+        node = self.compose_node(None, None)
+
+        # Drop the DOCUMENT-END event.
+        self.get_event()
+
+        self.anchors = {}
+        return node
+
+    def compose_node(self, parent, index):
+        if self.check_event(AliasEvent):
+            event = self.get_event()
+            anchor = event.anchor
+            if anchor not in self.anchors:
+                raise ComposerError(None, None, "found undefined alias %r"
+                        % anchor, event.start_mark)
+            return self.anchors[anchor]
+        event = self.peek_event()
+        anchor = event.anchor
+        if anchor is not None:
+            if anchor in self.anchors:
+                raise ComposerError("found duplicate anchor %r; first occurence"
+                        % anchor, self.anchors[anchor].start_mark,
+                        "second occurence", event.start_mark)
+        self.descend_resolver(parent, index)
+        if self.check_event(ScalarEvent):
+            node = self.compose_scalar_node(anchor)
+        elif self.check_event(SequenceStartEvent):
+            node = self.compose_sequence_node(anchor)
+        elif self.check_event(MappingStartEvent):
+            node = self.compose_mapping_node(anchor)
+        self.ascend_resolver()
+        return node
+
+    def compose_scalar_node(self, anchor):
+        event = self.get_event()
+        tag = event.tag
+        if tag is None or tag == '!':
+            tag = self.resolve(ScalarNode, event.value, event.implicit)
+        node = ScalarNode(tag, event.value,
+                event.start_mark, event.end_mark, style=event.style)
+        if anchor is not None:
+            self.anchors[anchor] = node
+        return node
+
+    def compose_sequence_node(self, anchor):
+        start_event = self.get_event()
+        tag = start_event.tag
+        if tag is None or tag == '!':
+            tag = self.resolve(SequenceNode, None, start_event.implicit)
+        node = SequenceNode(tag, [],
+                start_event.start_mark, None,
+                flow_style=start_event.flow_style)
+        if anchor is not None:
+            self.anchors[anchor] = node
+        index = 0
+        while not self.check_event(SequenceEndEvent):
+            node.value.append(self.compose_node(node, index))
+            index += 1
+        end_event = self.get_event()
+        node.end_mark = end_event.end_mark
+        return node
+
+    def compose_mapping_node(self, anchor):
+        start_event = self.get_event()
+        tag = start_event.tag
+        if tag is None or tag == '!':
+            tag = self.resolve(MappingNode, None, start_event.implicit)
+        node = MappingNode(tag, [],
+                start_event.start_mark, None,
+                flow_style=start_event.flow_style)
+        if anchor is not None:
+            self.anchors[anchor] = node
+        while not self.check_event(MappingEndEvent):
+            #key_event = self.peek_event()
+            item_key = self.compose_node(node, None)
+            #if item_key in node.value:
+            #    raise ComposerError("while composing a mapping", start_event.start_mark,
+            #            "found duplicate key", key_event.start_mark)
+            item_value = self.compose_node(node, item_key)
+            #node.value[item_key] = item_value
+            node.value.append((item_key, item_value))
+        end_event = self.get_event()
+        node.end_mark = end_event.end_mark
+        return node
+

tablib/packages/yaml3/constructor.py

@@ -0,0 +1,686 @@
+
+__all__ = ['BaseConstructor', 'SafeConstructor', 'Constructor',
+    'ConstructorError']
+
+from .error import *
+from .nodes import *
+
+import collections, datetime, base64, binascii, re, sys, types
+
+class ConstructorError(MarkedYAMLError):
+    pass
+
+class BaseConstructor:
+
+    yaml_constructors = {}
+    yaml_multi_constructors = {}
+
+    def __init__(self):
+        self.constructed_objects = {}
+        self.recursive_objects = {}
+        self.state_generators = []
+        self.deep_construct = False
+
+    def check_data(self):
+        # If there are more documents available?
+        return self.check_node()
+
+    def get_data(self):
+        # Construct and return the next document.
+        if self.check_node():
+            return self.construct_document(self.get_node())
+
+    def get_single_data(self):
+        # Ensure that the stream contains a single document and construct it.
+        node = self.get_single_node()
+        if node is not None:
+            return self.construct_document(node)
+        return None
+
+    def construct_document(self, node):
+        data = self.construct_object(node)
+        while self.state_generators:
+            state_generators = self.state_generators
+            self.state_generators = []
+            for generator in state_generators:
+                for dummy in generator:
+                    pass
+        self.constructed_objects = {}
+        self.recursive_objects = {}
+        self.deep_construct = False
+        return data
+
+    def construct_object(self, node, deep=False):
+        if deep:
+            old_deep = self.deep_construct
+            self.deep_construct = True
+        if node in self.constructed_objects:
+            return self.constructed_objects[node]
+        if node in self.recursive_objects:
+            raise ConstructorError(None, None,
+                    "found unconstructable recursive node", node.start_mark)
+        self.recursive_objects[node] = None
+        constructor = None
+        tag_suffix = None
+        if node.tag in self.yaml_constructors:
+            constructor = self.yaml_constructors[node.tag]
+        else:
+            for tag_prefix in self.yaml_multi_constructors:
+                if node.tag.startswith(tag_prefix):
+                    tag_suffix = node.tag[len(tag_prefix):]
+                    constructor = self.yaml_multi_constructors[tag_prefix]
+                    break
+            else:
+                if None in self.yaml_multi_constructors:
+                    tag_suffix = node.tag
+                    constructor = self.yaml_multi_constructors[None]
+                elif None in self.yaml_constructors:
+                    constructor = self.yaml_constructors[None]
+                elif isinstance(node, ScalarNode):
+                    constructor = self.__class__.construct_scalar
+                elif isinstance(node, SequenceNode):
+                    constructor = self.__class__.construct_sequence
+                elif isinstance(node, MappingNode):
+                    constructor = self.__class__.construct_mapping
+        if tag_suffix is None:
+            data = constructor(self, node)
+        else:
+            data = constructor(self, tag_suffix, node)
+        if isinstance(data, types.GeneratorType):
+            generator = data
+            data = next(generator)
+            if self.deep_construct:
+                for dummy in generator:
+                    pass
+            else:
+                self.state_generators.append(generator)
+        self.constructed_objects[node] = data
+        del self.recursive_objects[node]
+        if deep:
+            self.deep_construct = old_deep
+        return data
+
+    def construct_scalar(self, node):
+        if not isinstance(node, ScalarNode):
+            raise ConstructorError(None, None,
+                    "expected a scalar node, but found %s" % node.id,
+                    node.start_mark)
+        return node.value
+
+    def construct_sequence(self, node, deep=False):
+        if not isinstance(node, SequenceNode):
+            raise ConstructorError(None, None,
+                    "expected a sequence node, but found %s" % node.id,
+                    node.start_mark)
+        return [self.construct_object(child, deep=deep)
+                for child in node.value]
+
+    def construct_mapping(self, node, deep=False):
+        if not isinstance(node, MappingNode):
+            raise ConstructorError(None, None,
+                    "expected a mapping node, but found %s" % node.id,
+                    node.start_mark)
+        mapping = {}
+        for key_node, value_node in node.value:
+            key = self.construct_object(key_node, deep=deep)
+            if not isinstance(key, collections.Hashable):
+                raise ConstructorError("while constructing a mapping", node.start_mark,
+                        "found unhashable key", key_node.start_mark)
+            value = self.construct_object(value_node, deep=deep)
+            mapping[key] = value
+        return mapping
+
+    def construct_pairs(self, node, deep=False):
+        if not isinstance(node, MappingNode):
+            raise ConstructorError(None, None,
+                    "expected a mapping node, but found %s" % node.id,
+                    node.start_mark)
+        pairs = []
+        for key_node, value_node in node.value:
+            key = self.construct_object(key_node, deep=deep)
+            value = self.construct_object(value_node, deep=deep)
+            pairs.append((key, value))
+        return pairs
+
+    @classmethod
+    def add_constructor(cls, tag, constructor):
+        if not 'yaml_constructors' in cls.__dict__:
+            cls.yaml_constructors = cls.yaml_constructors.copy()
+        cls.yaml_constructors[tag] = constructor
+
+    @classmethod
+    def add_multi_constructor(cls, tag_prefix, multi_constructor):
+        if not 'yaml_multi_constructors' in cls.__dict__:
+            cls.yaml_multi_constructors = cls.yaml_multi_constructors.copy()
+        cls.yaml_multi_constructors[tag_prefix] = multi_constructor
+
+class SafeConstructor(BaseConstructor):
+
+    def construct_scalar(self, node):
+        if isinstance(node, MappingNode):
+            for key_node, value_node in node.value:
+                if key_node.tag == 'tag:yaml.org,2002:value':
+                    return self.construct_scalar(value_node)
+        return super().construct_scalar(node)
+
+    def flatten_mapping(self, node):
+        merge = []
+        index = 0
+        while index < len(node.value):
+            key_node, value_node = node.value[index]
+            if key_node.tag == 'tag:yaml.org,2002:merge':
+                del node.value[index]
+                if isinstance(value_node, MappingNode):
+                    self.flatten_mapping(value_node)
+                    merge.extend(value_node.value)
+                elif isinstance(value_node, SequenceNode):
+                    submerge = []
+                    for subnode in value_node.value:
+                        if not isinstance(subnode, MappingNode):
+                            raise ConstructorError("while constructing a mapping",
+                                    node.start_mark,
+                                    "expected a mapping for merging, but found %s"
+                                    % subnode.id, subnode.start_mark)
+                        self.flatten_mapping(subnode)
+                        submerge.append(subnode.value)
+                    submerge.reverse()
+                    for value in submerge:
+                        merge.extend(value)
+                else:
+                    raise ConstructorError("while constructing a mapping", node.start_mark,
+                            "expected a mapping or list of mappings for merging, but found %s"
+                            % value_node.id, value_node.start_mark)
+            elif key_node.tag == 'tag:yaml.org,2002:value':
+                key_node.tag = 'tag:yaml.org,2002:str'
+                index += 1
+            else:
+                index += 1
+        if merge:
+            node.value = merge + node.value
+
+    def construct_mapping(self, node, deep=False):
+        if isinstance(node, MappingNode):
+            self.flatten_mapping(node)
+        return super().construct_mapping(node, deep=deep)
+
+    def construct_yaml_null(self, node):
+        self.construct_scalar(node)
+        return None
+
+    bool_values = {
+        'yes':      True,
+        'no':       False,
+        'true':     True,
+        'false':    False,
+        'on':       True,
+        'off':      False,
+    }
+
+    def construct_yaml_bool(self, node):
+        value = self.construct_scalar(node)
+        return self.bool_values[value.lower()]
+
+    def construct_yaml_int(self, node):
+        value = self.construct_scalar(node)
+        value = value.replace('_', '')
+        sign = +1
+        if value[0] == '-':
+            sign = -1
+        if value[0] in '+-':
+            value = value[1:]
+        if value == '0':
+            return 0
+        elif value.startswith('0b'):
+            return sign*int(value[2:], 2)
+        elif value.startswith('0x'):
+            return sign*int(value[2:], 16)
+        elif value[0] == '0':
+            return sign*int(value, 8)
+        elif ':' in value:
+            digits = [int(part) for part in value.split(':')]
+            digits.reverse()
+            base = 1
+            value = 0
+            for digit in digits:
+                value += digit*base
+                base *= 60
+            return sign*value
+        else:
+            return sign*int(value)
+
+    inf_value = 1e300
+    while inf_value != inf_value*inf_value:
+        inf_value *= inf_value
+    nan_value = -inf_value/inf_value   # Trying to make a quiet NaN (like C99).
+
+    def construct_yaml_float(self, node):
+        value = self.construct_scalar(node)
+        value = value.replace('_', '').lower()
+        sign = +1
+        if value[0] == '-':
+            sign = -1
+        if value[0] in '+-':
+            value = value[1:]
+        if value == '.inf':
+            return sign*self.inf_value
+        elif value == '.nan':
+            return self.nan_value
+        elif ':' in value:
+            digits = [float(part) for part in value.split(':')]
+            digits.reverse()
+            base = 1
+            value = 0.0
+            for digit in digits:
+                value += digit*base
+                base *= 60
+            return sign*value
+        else:
+            return sign*float(value)
+
+    def construct_yaml_binary(self, node):
+        try:
+            value = self.construct_scalar(node).encode('ascii')
+        except UnicodeEncodeError as exc:
+            raise ConstructorError(None, None,
+                    "failed to convert base64 data into ascii: %s" % exc,
+                    node.start_mark)
+        try:
+            if hasattr(base64, 'decodebytes'):
+                return base64.decodebytes(value)
+            else:
+                return base64.decodestring(value)
+        except binascii.Error as exc:
+            raise ConstructorError(None, None,
+                    "failed to decode base64 data: %s" % exc, node.start_mark)
+
+    timestamp_regexp = re.compile(
+            r'''^(?P<year>[0-9][0-9][0-9][0-9])
+                -(?P<month>[0-9][0-9]?)
+                -(?P<day>[0-9][0-9]?)
+                (?:(?:[Tt]|[ \t]+)
+                (?P<hour>[0-9][0-9]?)
+                :(?P<minute>[0-9][0-9])
+                :(?P<second>[0-9][0-9])
+                (?:\.(?P<fraction>[0-9]*))?
+                (?:[ \t]*(?P<tz>Z|(?P<tz_sign>[-+])(?P<tz_hour>[0-9][0-9]?)
+                (?::(?P<tz_minute>[0-9][0-9]))?))?)?$''', re.X)
+
+    def construct_yaml_timestamp(self, node):
+        value = self.construct_scalar(node)
+        match = self.timestamp_regexp.match(node.value)
+        values = match.groupdict()
+        year = int(values['year'])
+        month = int(values['month'])
+        day = int(values['day'])
+        if not values['hour']:
+            return datetime.date(year, month, day)
+        hour = int(values['hour'])
+        minute = int(values['minute'])
+        second = int(values['second'])
+        fraction = 0
+        if values['fraction']:
+            fraction = values['fraction'][:6]
+            while len(fraction) < 6:
+                fraction += '0'
+            fraction = int(fraction)
+        delta = None
+        if values['tz_sign']:
+            tz_hour = int(values['tz_hour'])
+            tz_minute = int(values['tz_minute'] or 0)
+            delta = datetime.timedelta(hours=tz_hour, minutes=tz_minute)
+            if values['tz_sign'] == '-':
+                delta = -delta
+        data = datetime.datetime(year, month, day, hour, minute, second, fraction)
+        if delta:
+            data -= delta
+        return data
+
+    def construct_yaml_omap(self, node):
+        # Note: we do not check for duplicate keys, because it's too
+        # CPU-expensive.
+        omap = []
+        yield omap
+        if not isinstance(node, SequenceNode):
+            raise ConstructorError("while constructing an ordered map", node.start_mark,
+                    "expected a sequence, but found %s" % node.id, node.start_mark)
+        for subnode in node.value:
+            if not isinstance(subnode, MappingNode):
+                raise ConstructorError("while constructing an ordered map", node.start_mark,
+                        "expected a mapping of length 1, but found %s" % subnode.id,
+                        subnode.start_mark)
+            if len(subnode.value) != 1:
+                raise ConstructorError("while constructing an ordered map", node.start_mark,
+                        "expected a single mapping item, but found %d items" % len(subnode.value),
+                        subnode.start_mark)
+            key_node, value_node = subnode.value[0]
+            key = self.construct_object(key_node)
+            value = self.construct_object(value_node)
+            omap.append((key, value))
+
+    def construct_yaml_pairs(self, node):
+        # Note: the same code as `construct_yaml_omap`.
+        pairs = []
+        yield pairs
+        if not isinstance(node, SequenceNode):
+            raise ConstructorError("while constructing pairs", node.start_mark,
+                    "expected a sequence, but found %s" % node.id, node.start_mark)
+        for subnode in node.value:
+            if not isinstance(subnode, MappingNode):
+                raise ConstructorError("while constructing pairs", node.start_mark,
+                        "expected a mapping of length 1, but found %s" % subnode.id,
+                        subnode.start_mark)
+            if len(subnode.value) != 1:
+                raise ConstructorError("while constructing pairs", node.start_mark,
+                        "expected a single mapping item, but found %d items" % len(subnode.value),
+                        subnode.start_mark)
+            key_node, value_node = subnode.value[0]
+            key = self.construct_object(key_node)
+            value = self.construct_object(value_node)
+            pairs.append((key, value))
+
+    def construct_yaml_set(self, node):
+        data = set()
+        yield data
+        value = self.construct_mapping(node)
+        data.update(value)
+
+    def construct_yaml_str(self, node):
+        return self.construct_scalar(node)
+
+    def construct_yaml_seq(self, node):
+        data = []
+        yield data
+        data.extend(self.construct_sequence(node))
+
+    def construct_yaml_map(self, node):
+        data = {}
+        yield data
+        value = self.construct_mapping(node)
+        data.update(value)
+
+    def construct_yaml_object(self, node, cls):
+        data = cls.__new__(cls)
+        yield data
+        if hasattr(data, '__setstate__'):
+            state = self.construct_mapping(node, deep=True)
+            data.__setstate__(state)
+        else:
+            state = self.construct_mapping(node)
+            data.__dict__.update(state)
+
+    def construct_undefined(self, node):
+        raise ConstructorError(None, None,
+                "could not determine a constructor for the tag %r" % node.tag,
+                node.start_mark)
+
+SafeConstructor.add_constructor(
+        'tag:yaml.org,2002:null',
+        SafeConstructor.construct_yaml_null)
+
+SafeConstructor.add_constructor(
+        'tag:yaml.org,2002:bool',
+        SafeConstructor.construct_yaml_bool)
+
+SafeConstructor.add_constructor(
+        'tag:yaml.org,2002:int',
+        SafeConstructor.construct_yaml_int)
+
+SafeConstructor.add_constructor(
+        'tag:yaml.org,2002:float',
+        SafeConstructor.construct_yaml_float)
+
+SafeConstructor.add_constructor(
+        'tag:yaml.org,2002:binary',
+        SafeConstructor.construct_yaml_binary)
+
+SafeConstructor.add_constructor(
+        'tag:yaml.org,2002:timestamp',
+        SafeConstructor.construct_yaml_timestamp)
+
+SafeConstructor.add_constructor(
+        'tag:yaml.org,2002:omap',
+        SafeConstructor.construct_yaml_omap)
+
+SafeConstructor.add_constructor(
+        'tag:yaml.org,2002:pairs',
+        SafeConstructor.construct_yaml_pairs)
+
+SafeConstructor.add_constructor(
+        'tag:yaml.org,2002:set',
+        SafeConstructor.construct_yaml_set)
+
+SafeConstructor.add_constructor(
+        'tag:yaml.org,2002:str',
+        SafeConstructor.construct_yaml_str)
+
+SafeConstructor.add_constructor(
+        'tag:yaml.org,2002:seq',
+        SafeConstructor.construct_yaml_seq)
+
+SafeConstructor.add_constructor(
+        'tag:yaml.org,2002:map',
+        SafeConstructor.construct_yaml_map)
+
+SafeConstructor.add_constructor(None,
+        SafeConstructor.construct_undefined)
+
+class Constructor(SafeConstructor):
+
+    def construct_python_str(self, node):
+        return self.construct_scalar(node)
+
+    def construct_python_unicode(self, node):
+        return self.construct_scalar(node)
+
+    def construct_python_bytes(self, node):
+        try:
+            value = self.construct_scalar(node).encode('ascii')
+        except UnicodeEncodeError as exc:
+            raise ConstructorError(None, None,
+                    "failed to convert base64 data into ascii: %s" % exc,
+                    node.start_mark)
+        try:
+            if hasattr(base64, 'decodebytes'):
+                return base64.decodebytes(value)
+            else:
+                return base64.decodestring(value)
+        except binascii.Error as exc:
+            raise ConstructorError(None, None,
+                    "failed to decode base64 data: %s" % exc, node.start_mark)
+
+    def construct_python_long(self, node):
+        return self.construct_yaml_int(node)
+
+    def construct_python_complex(self, node):
+       return complex(self.construct_scalar(node))
+
+    def construct_python_tuple(self, node):
+        return tuple(self.construct_sequence(node))
+
+    def find_python_module(self, name, mark):
+        if not name:
+            raise ConstructorError("while constructing a Python module", mark,
+                    "expected non-empty name appended to the tag", mark)
+        try:
+            __import__(name)
+        except ImportError as exc:
+            raise ConstructorError("while constructing a Python module", mark,
+                    "cannot find module %r (%s)" % (name, exc), mark)
+        return sys.modules[name]
+
+    def find_python_name(self, name, mark):
+        if not name:
+            raise ConstructorError("while constructing a Python object", mark,
+                    "expected non-empty name appended to the tag", mark)
+        if '.' in name:
+            module_name, object_name = name.rsplit('.', 1)
+        else:
+            module_name = 'builtins'
+            object_name = name
+        try:
+            __import__(module_name)
+        except ImportError as exc:
+            raise ConstructorError("while constructing a Python object", mark,
+                    "cannot find module %r (%s)" % (module_name, exc), mark)
+        module = sys.modules[module_name]
+        if not hasattr(module, object_name):
+            raise ConstructorError("while constructing a Python object", mark,
+                    "cannot find %r in the module %r"
+                    % (object_name, module.__name__), mark)
+        return getattr(module, object_name)
+
+    def construct_python_name(self, suffix, node):
+        value = self.construct_scalar(node)
+        if value:
+            raise ConstructorError("while constructing a Python name", node.start_mark,
+                    "expected the empty value, but found %r" % value, node.start_mark)
+        return self.find_python_name(suffix, node.start_mark)
+
+    def construct_python_module(self, suffix, node):
+        value = self.construct_scalar(node)
+        if value:
+            raise ConstructorError("while constructing a Python module", node.start_mark,
+                    "expected the empty value, but found %r" % value, node.start_mark)
+        return self.find_python_module(suffix, node.start_mark)
+
+    def make_python_instance(self, suffix, node,
+            args=None, kwds=None, newobj=False):
+        if not args:
+            args = []
+        if not kwds:
+            kwds = {}
+        cls = self.find_python_name(suffix, node.start_mark)
+        if newobj and isinstance(cls, type):
+            return cls.__new__(cls, *args, **kwds)
+        else:
+            return cls(*args, **kwds)
+
+    def set_python_instance_state(self, instance, state):
+        if hasattr(instance, '__setstate__'):
+            instance.__setstate__(state)
+        else:
+            slotstate = {}
+            if isinstance(state, tuple) and len(state) == 2:
+                state, slotstate = state
+            if hasattr(instance, '__dict__'):
+                instance.__dict__.update(state)
+            elif state:
+                slotstate.update(state)
+            for key, value in slotstate.items():
+                setattr(object, key, value)
+
+    def construct_python_object(self, suffix, node):
+        # Format:
+        #   !!python/object:module.name { ... state ... }
+        instance = self.make_python_instance(suffix, node, newobj=True)
+        yield instance
+        deep = hasattr(instance, '__setstate__')
+        state = self.construct_mapping(node, deep=deep)
+        self.set_python_instance_state(instance, state)
+
+    def construct_python_object_apply(self, suffix, node, newobj=False):
+        # Format:
+        #   !!python/object/apply       # (or !!python/object/new)
+        #   args: [ ... arguments ... ]
+        #   kwds: { ... keywords ... }
+        #   state: ... state ...
+        #   listitems: [ ... listitems ... ]
+        #   dictitems: { ... dictitems ... }
+        # or short format:
+        #   !!python/object/apply [ ... arguments ... ]
+        # The difference between !!python/object/apply and !!python/object/new
+        # is how an object is created, check make_python_instance for details.
+        if isinstance(node, SequenceNode):
+            args = self.construct_sequence(node, deep=True)
+            kwds = {}
+            state = {}
+            listitems = []
+            dictitems = {}
+        else:
+            value = self.construct_mapping(node, deep=True)
+            args = value.get('args', [])
+            kwds = value.get('kwds', {})
+            state = value.get('state', {})
+            listitems = value.get('listitems', [])
+            dictitems = value.get('dictitems', {})
+        instance = self.make_python_instance(suffix, node, args, kwds, newobj)
+        if state:
+            self.set_python_instance_state(instance, state)
+        if listitems:
+            instance.extend(listitems)
+        if dictitems:
+            for key in dictitems:
+                instance[key] = dictitems[key]
+        return instance
+
+    def construct_python_object_new(self, suffix, node):
+        return self.construct_python_object_apply(suffix, node, newobj=True)
+
+Constructor.add_constructor(
+    'tag:yaml.org,2002:python/none',
+    Constructor.construct_yaml_null)
+
+Constructor.add_constructor(
+    'tag:yaml.org,2002:python/bool',
+    Constructor.construct_yaml_bool)
+
+Constructor.add_constructor(
+    'tag:yaml.org,2002:python/str',
+    Constructor.construct_python_str)
+
+Constructor.add_constructor(
+    'tag:yaml.org,2002:python/unicode',
+    Constructor.construct_python_unicode)
+
+Constructor.add_constructor(
+    'tag:yaml.org,2002:python/bytes',
+    Constructor.construct_python_bytes)
+
+Constructor.add_constructor(
+    'tag:yaml.org,2002:python/int',
+    Constructor.construct_yaml_int)
+
+Constructor.add_constructor(
+    'tag:yaml.org,2002:python/long',
+    Constructor.construct_python_long)
+
+Constructor.add_constructor(
+    'tag:yaml.org,2002:python/float',
+    Constructor.construct_yaml_float)
+
+Constructor.add_constructor(
+    'tag:yaml.org,2002:python/complex',
+    Constructor.construct_python_complex)
+
+Constructor.add_constructor(
+    'tag:yaml.org,2002:python/list',
+    Constructor.construct_yaml_seq)
+
+Constructor.add_constructor(
+    'tag:yaml.org,2002:python/tuple',
+    Constructor.construct_python_tuple)
+
+Constructor.add_constructor(
+    'tag:yaml.org,2002:python/dict',
+    Constructor.construct_yaml_map)
+
+Constructor.add_multi_constructor(
+    'tag:yaml.org,2002:python/name:',
+    Constructor.construct_python_name)
+
+Constructor.add_multi_constructor(
+    'tag:yaml.org,2002:python/module:',
+    Constructor.construct_python_module)
+
+Constructor.add_multi_constructor(
+    'tag:yaml.org,2002:python/object:',
+    Constructor.construct_python_object)
+
+Constructor.add_multi_constructor(
+    'tag:yaml.org,2002:python/object/apply:',
+    Constructor.construct_python_object_apply)
+
+Constructor.add_multi_constructor(
+    'tag:yaml.org,2002:python/object/new:',
+    Constructor.construct_python_object_new)
+

tablib/packages/yaml3/cyaml.py

@@ -0,0 +1,85 @@
+
+__all__ = ['CBaseLoader', 'CSafeLoader', 'CLoader',
+        'CBaseDumper', 'CSafeDumper', 'CDumper']
+
+from _yaml import CParser, CEmitter
+
+from .constructor import *
+
+from .serializer import *
+from .representer import *
+
+from .resolver import *
+
+class CBaseLoader(CParser, BaseConstructor, BaseResolver):
+
+    def __init__(self, stream):
+        CParser.__init__(self, stream)
+        BaseConstructor.__init__(self)
+        BaseResolver.__init__(self)
+
+class CSafeLoader(CParser, SafeConstructor, Resolver):
+
+    def __init__(self, stream):
+        CParser.__init__(self, stream)
+        SafeConstructor.__init__(self)
+        Resolver.__init__(self)
+
+class CLoader(CParser, Constructor, Resolver):
+
+    def __init__(self, stream):
+        CParser.__init__(self, stream)
+        Constructor.__init__(self)
+        Resolver.__init__(self)
+
+class CBaseDumper(CEmitter, BaseRepresenter, BaseResolver):
+
+    def __init__(self, stream,
+            default_style=None, default_flow_style=None,
+            canonical=None, indent=None, width=None,
+            allow_unicode=None, line_break=None,
+            encoding=None, explicit_start=None, explicit_end=None,
+            version=None, tags=None):
+        CEmitter.__init__(self, stream, canonical=canonical,
+                indent=indent, width=width, encoding=encoding,
+                allow_unicode=allow_unicode, line_break=line_break,
+                explicit_start=explicit_start, explicit_end=explicit_end,
+                version=version, tags=tags)
+        Representer.__init__(self, default_style=default_style,
+                default_flow_style=default_flow_style)
+        Resolver.__init__(self)
+
+class CSafeDumper(CEmitter, SafeRepresenter, Resolver):
+
+    def __init__(self, stream,
+            default_style=None, default_flow_style=None,
+            canonical=None, indent=None, width=None,
+            allow_unicode=None, line_break=None,
+            encoding=None, explicit_start=None, explicit_end=None,
+            version=None, tags=None):
+        CEmitter.__init__(self, stream, canonical=canonical,
+                indent=indent, width=width, encoding=encoding,
+                allow_unicode=allow_unicode, line_break=line_break,
+                explicit_start=explicit_start, explicit_end=explicit_end,
+                version=version, tags=tags)
+        SafeRepresenter.__init__(self, default_style=default_style,
+                default_flow_style=default_flow_style)
+        Resolver.__init__(self)
+
+class CDumper(CEmitter, Serializer, Representer, Resolver):
+
+    def __init__(self, stream,
+            default_style=None, default_flow_style=None,
+            canonical=None, indent=None, width=None,
+            allow_unicode=None, line_break=None,
+            encoding=None, explicit_start=None, explicit_end=None,
+            version=None, tags=None):
+        CEmitter.__init__(self, stream, canonical=canonical,
+                indent=indent, width=width, encoding=encoding,
+                allow_unicode=allow_unicode, line_break=line_break,
+                explicit_start=explicit_start, explicit_end=explicit_end,
+                version=version, tags=tags)
+        Representer.__init__(self, default_style=default_style,
+                default_flow_style=default_flow_style)
+        Resolver.__init__(self)
+

tablib/packages/yaml3/dumper.py

@@ -0,0 +1,62 @@
+
+__all__ = ['BaseDumper', 'SafeDumper', 'Dumper']
+
+from .emitter import *
+from .serializer import *
+from .representer import *
+from .resolver import *
+
+class BaseDumper(Emitter, Serializer, BaseRepresenter, BaseResolver):
+
+    def __init__(self, stream,
+            default_style=None, default_flow_style=None,
+            canonical=None, indent=None, width=None,
+            allow_unicode=None, line_break=None,
+            encoding=None, explicit_start=None, explicit_end=None,
+            version=None, tags=None):
+        Emitter.__init__(self, stream, canonical=canonical,
+                indent=indent, width=width,
+                allow_unicode=allow_unicode, line_break=line_break)
+        Serializer.__init__(self, encoding=encoding,
+                explicit_start=explicit_start, explicit_end=explicit_end,
+                version=version, tags=tags)
+        Representer.__init__(self, default_style=default_style,
+                default_flow_style=default_flow_style)
+        Resolver.__init__(self)
+
+class SafeDumper(Emitter, Serializer, SafeRepresenter, Resolver):
+
+    def __init__(self, stream,
+            default_style=None, default_flow_style=None,
+            canonical=None, indent=None, width=None,
+            allow_unicode=None, line_break=None,
+            encoding=None, explicit_start=None, explicit_end=None,
+            version=None, tags=None):
+        Emitter.__init__(self, stream, canonical=canonical,
+                indent=indent, width=width,
+                allow_unicode=allow_unicode, line_break=line_break)
+        Serializer.__init__(self, encoding=encoding,
+                explicit_start=explicit_start, explicit_end=explicit_end,
+                version=version, tags=tags)
+        SafeRepresenter.__init__(self, default_style=default_style,
+                default_flow_style=default_flow_style)
+        Resolver.__init__(self)
+
+class Dumper(Emitter, Serializer, Representer, Resolver):
+
+    def __init__(self, stream,
+            default_style=None, default_flow_style=None,
+            canonical=None, indent=None, width=None,
+            allow_unicode=None, line_break=None,
+            encoding=None, explicit_start=None, explicit_end=None,
+            version=None, tags=None):
+        Emitter.__init__(self, stream, canonical=canonical,
+                indent=indent, width=width,
+                allow_unicode=allow_unicode, line_break=line_break)
+        Serializer.__init__(self, encoding=encoding,
+                explicit_start=explicit_start, explicit_end=explicit_end,
+                version=version, tags=tags)
+        Representer.__init__(self, default_style=default_style,
+                default_flow_style=default_flow_style)
+        Resolver.__init__(self)
+

tablib/packages/yaml3/error.py

@@ -0,0 +1,75 @@
+
+__all__ = ['Mark', 'YAMLError', 'MarkedYAMLError']
+
+class Mark:
+
+    def __init__(self, name, index, line, column, buffer, pointer):
+        self.name = name
+        self.index = index
+        self.line = line
+        self.column = column
+        self.buffer = buffer
+        self.pointer = pointer
+
+    def get_snippet(self, indent=4, max_length=75):
+        if self.buffer is None:
+            return None
+        head = ''
+        start = self.pointer
+        while start > 0 and self.buffer[start-1] not in '\0\r\n\x85\u2028\u2029':
+            start -= 1
+            if self.pointer-start > max_length/2-1:
+                head = ' ... '
+                start += 5
+                break
+        tail = ''
+        end = self.pointer
+        while end < len(self.buffer) and self.buffer[end] not in '\0\r\n\x85\u2028\u2029':
+            end += 1
+            if end-self.pointer > max_length/2-1:
+                tail = ' ... '
+                end -= 5
+                break
+        snippet = self.buffer[start:end]
+        return ' '*indent + head + snippet + tail + '\n'  \
+                + ' '*(indent+self.pointer-start+len(head)) + '^'
+
+    def __str__(self):
+        snippet = self.get_snippet()
+        where = "  in \"%s\", line %d, column %d"   \
+                % (self.name, self.line+1, self.column+1)
+        if snippet is not None:
+            where += ":\n"+snippet
+        return where
+
+class YAMLError(Exception):
+    pass
+
+class MarkedYAMLError(YAMLError):
+
+    def __init__(self, context=None, context_mark=None,
+            problem=None, problem_mark=None, note=None):
+        self.context = context
+        self.context_mark = context_mark
+        self.problem = problem
+        self.problem_mark = problem_mark
+        self.note = note
+
+    def __str__(self):
+        lines = []
+        if self.context is not None:
+            lines.append(self.context)
+        if self.context_mark is not None  \
+            and (self.problem is None or self.problem_mark is None
+                    or self.context_mark.name != self.problem_mark.name
+                    or self.context_mark.line != self.problem_mark.line
+                    or self.context_mark.column != self.problem_mark.column):
+            lines.append(str(self.context_mark))
+        if self.problem is not None:
+            lines.append(self.problem)
+        if self.problem_mark is not None:
+            lines.append(str(self.problem_mark))
+        if self.note is not None:
+            lines.append(self.note)
+        return '\n'.join(lines)
+

tablib/packages/yaml3/events.py

@@ -0,0 +1,86 @@
+
+# Abstract classes.
+
+class Event(object):
+    def __init__(self, start_mark=None, end_mark=None):
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+    def __repr__(self):
+        attributes = [key for key in ['anchor', 'tag', 'implicit', 'value']
+                if hasattr(self, key)]
+        arguments = ', '.join(['%s=%r' % (key, getattr(self, key))
+                for key in attributes])
+        return '%s(%s)' % (self.__class__.__name__, arguments)
+
+class NodeEvent(Event):
+    def __init__(self, anchor, start_mark=None, end_mark=None):
+        self.anchor = anchor
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+
+class CollectionStartEvent(NodeEvent):
+    def __init__(self, anchor, tag, implicit, start_mark=None, end_mark=None,
+            flow_style=None):
+        self.anchor = anchor
+        self.tag = tag
+        self.implicit = implicit
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+        self.flow_style = flow_style
+
+class CollectionEndEvent(Event):
+    pass
+
+# Implementations.
+
+class StreamStartEvent(Event):
+    def __init__(self, start_mark=None, end_mark=None, encoding=None):
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+        self.encoding = encoding
+
+class StreamEndEvent(Event):
+    pass
+
+class DocumentStartEvent(Event):
+    def __init__(self, start_mark=None, end_mark=None,
+            explicit=None, version=None, tags=None):
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+        self.explicit = explicit
+        self.version = version
+        self.tags = tags
+
+class DocumentEndEvent(Event):
+    def __init__(self, start_mark=None, end_mark=None,
+            explicit=None):
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+        self.explicit = explicit
+
+class AliasEvent(NodeEvent):
+    pass
+
+class ScalarEvent(NodeEvent):
+    def __init__(self, anchor, tag, implicit, value,
+            start_mark=None, end_mark=None, style=None):
+        self.anchor = anchor
+        self.tag = tag
+        self.implicit = implicit
+        self.value = value
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+        self.style = style
+
+class SequenceStartEvent(CollectionStartEvent):
+    pass
+
+class SequenceEndEvent(CollectionEndEvent):
+    pass
+
+class MappingStartEvent(CollectionStartEvent):
+    pass
+
+class MappingEndEvent(CollectionEndEvent):
+    pass
+

tablib/packages/yaml3/loader.py

@@ -0,0 +1,40 @@
+
+__all__ = ['BaseLoader', 'SafeLoader', 'Loader']
+
+from .reader import *
+from .scanner import *
+from .parser import *
+from .composer import *
+from .constructor import *
+from .resolver import *
+
+class BaseLoader(Reader, Scanner, Parser, Composer, BaseConstructor, BaseResolver):
+
+    def __init__(self, stream):
+        Reader.__init__(self, stream)
+        Scanner.__init__(self)
+        Parser.__init__(self)
+        Composer.__init__(self)
+        BaseConstructor.__init__(self)
+        BaseResolver.__init__(self)
+
+class SafeLoader(Reader, Scanner, Parser, Composer, SafeConstructor, Resolver):
+
+    def __init__(self, stream):
+        Reader.__init__(self, stream)
+        Scanner.__init__(self)
+        Parser.__init__(self)
+        Composer.__init__(self)
+        SafeConstructor.__init__(self)
+        Resolver.__init__(self)
+
+class Loader(Reader, Scanner, Parser, Composer, Constructor, Resolver):
+
+    def __init__(self, stream):
+        Reader.__init__(self, stream)
+        Scanner.__init__(self)
+        Parser.__init__(self)
+        Composer.__init__(self)
+        Constructor.__init__(self)
+        Resolver.__init__(self)
+

tablib/packages/yaml3/nodes.py

@@ -0,0 +1,49 @@
+
+class Node(object):
+    def __init__(self, tag, value, start_mark, end_mark):
+        self.tag = tag
+        self.value = value
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+    def __repr__(self):
+        value = self.value
+        #if isinstance(value, list):
+        #    if len(value) == 0:
+        #        value = '<empty>'
+        #    elif len(value) == 1:
+        #        value = '<1 item>'
+        #    else:
+        #        value = '<%d items>' % len(value)
+        #else:
+        #    if len(value) > 75:
+        #        value = repr(value[:70]+u' ... ')
+        #    else:
+        #        value = repr(value)
+        value = repr(value)
+        return '%s(tag=%r, value=%s)' % (self.__class__.__name__, self.tag, value)
+
+class ScalarNode(Node):
+    id = 'scalar'
+    def __init__(self, tag, value,
+            start_mark=None, end_mark=None, style=None):
+        self.tag = tag
+        self.value = value
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+        self.style = style
+
+class CollectionNode(Node):
+    def __init__(self, tag, value,
+            start_mark=None, end_mark=None, flow_style=None):
+        self.tag = tag
+        self.value = value
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+        self.flow_style = flow_style
+
+class SequenceNode(CollectionNode):
+    id = 'sequence'
+
+class MappingNode(CollectionNode):
+    id = 'mapping'
+

tablib/packages/yaml3/parser.py

@@ -0,0 +1,584 @@
+
+# The following YAML grammar is LL(1) and is parsed by a recursive descent
+# parser.
+#
+# stream            ::= STREAM-START implicit_document? explicit_document* STREAM-END
+# implicit_document ::= block_node DOCUMENT-END*
+# explicit_document ::= DIRECTIVE* DOCUMENT-START block_node? DOCUMENT-END*
+# block_node_or_indentless_sequence ::=
+#                       ALIAS
+#                       | properties (block_content | indentless_block_sequence)?
+#                       | block_content
+#                       | indentless_block_sequence
+# block_node        ::= ALIAS
+#                       | properties block_content?
+#                       | block_content
+# flow_node         ::= ALIAS
+#                       | properties flow_content?
+#                       | flow_content
+# properties        ::= TAG ANCHOR? | ANCHOR TAG?
+# block_content     ::= block_collection | flow_collection | SCALAR
+# flow_content      ::= flow_collection | SCALAR
+# block_collection  ::= block_sequence | block_mapping
+# flow_collection   ::= flow_sequence | flow_mapping
+# block_sequence    ::= BLOCK-SEQUENCE-START (BLOCK-ENTRY block_node?)* BLOCK-END
+# indentless_sequence   ::= (BLOCK-ENTRY block_node?)+
+# block_mapping     ::= BLOCK-MAPPING_START
+#                       ((KEY block_node_or_indentless_sequence?)?
+#                       (VALUE block_node_or_indentless_sequence?)?)*
+#                       BLOCK-END
+# flow_sequence     ::= FLOW-SEQUENCE-START
+#                       (flow_sequence_entry FLOW-ENTRY)*
+#                       flow_sequence_entry?
+#                       FLOW-SEQUENCE-END
+# flow_sequence_entry   ::= flow_node | KEY flow_node? (VALUE flow_node?)?
+# flow_mapping      ::= FLOW-MAPPING-START
+#                       (flow_mapping_entry FLOW-ENTRY)*
+#                       flow_mapping_entry?
+#                       FLOW-MAPPING-END
+# flow_mapping_entry    ::= flow_node | KEY flow_node? (VALUE flow_node?)?
+#
+# FIRST sets:
+#
+# stream: { STREAM-START }
+# explicit_document: { DIRECTIVE DOCUMENT-START }
+# implicit_document: FIRST(block_node)
+# block_node: { ALIAS TAG ANCHOR SCALAR BLOCK-SEQUENCE-START BLOCK-MAPPING-START FLOW-SEQUENCE-START FLOW-MAPPING-START }
+# flow_node: { ALIAS ANCHOR TAG SCALAR FLOW-SEQUENCE-START FLOW-MAPPING-START }
+# block_content: { BLOCK-SEQUENCE-START BLOCK-MAPPING-START FLOW-SEQUENCE-START FLOW-MAPPING-START SCALAR }
+# flow_content: { FLOW-SEQUENCE-START FLOW-MAPPING-START SCALAR }
+# block_collection: { BLOCK-SEQUENCE-START BLOCK-MAPPING-START }
+# flow_collection: { FLOW-SEQUENCE-START FLOW-MAPPING-START }
+# block_sequence: { BLOCK-SEQUENCE-START }
+# block_mapping: { BLOCK-MAPPING-START }
+# block_node_or_indentless_sequence: { ALIAS ANCHOR TAG SCALAR BLOCK-SEQUENCE-START BLOCK-MAPPING-START FLOW-SEQUENCE-START FLOW-MAPPING-START BLOCK-ENTRY }
+# indentless_sequence: { ENTRY }
+# flow_collection: { FLOW-SEQUENCE-START FLOW-MAPPING-START }
+# flow_sequence: { FLOW-SEQUENCE-START }
+# flow_mapping: { FLOW-MAPPING-START }
+# flow_sequence_entry: { ALIAS ANCHOR TAG SCALAR FLOW-SEQUENCE-START FLOW-MAPPING-START KEY }
+# flow_mapping_entry: { ALIAS ANCHOR TAG SCALAR FLOW-SEQUENCE-START FLOW-MAPPING-START KEY }
+
+__all__ = ['Parser', 'ParserError']
+
+from .error import MarkedYAMLError
+from .tokens import *
+from .events import *
+from .scanner import *
+
+class ParserError(MarkedYAMLError):
+    pass
+
+class Parser:
+    # Since writing a recursive-descendant parser is a straightforward task, we
+    # do not give many comments here.
+
+    DEFAULT_TAGS = {
+        '!':   '!',
+        '!!':  'tag:yaml.org,2002:',
+    }
+
+    def __init__(self):
+        self.current_event = None
+        self.yaml_version = None
+        self.tag_handles = {}
+        self.states = []
+        self.marks = []
+        self.state = self.parse_stream_start
+
+    def check_event(self, *choices):
+        # Check the type of the next event.
+        if self.current_event is None:
+            if self.state:
+                self.current_event = self.state()
+        if self.current_event is not None:
+            if not choices:
+                return True
+            for choice in choices:
+                if isinstance(self.current_event, choice):
+                    return True
+        return False
+
+    def peek_event(self):
+        # Get the next event.
+        if self.current_event is None:
+            if self.state:
+                self.current_event = self.state()
+        return self.current_event
+
+    def get_event(self):
+        # Get the next event and proceed further.
+        if self.current_event is None:
+            if self.state:
+                self.current_event = self.state()
+        value = self.current_event
+        self.current_event = None
+        return value
+
+    # stream    ::= STREAM-START implicit_document? explicit_document* STREAM-END
+    # implicit_document ::= block_node DOCUMENT-END*
+    # explicit_document ::= DIRECTIVE* DOCUMENT-START block_node? DOCUMENT-END*
+
+    def parse_stream_start(self):
+
+        # Parse the stream start.
+        token = self.get_token()
+        event = StreamStartEvent(token.start_mark, token.end_mark,
+                encoding=token.encoding)
+
+        # Prepare the next state.
+        self.state = self.parse_implicit_document_start
+
+        return event
+
+    def parse_implicit_document_start(self):
+
+        # Parse an implicit document.
+        if not self.check_token(DirectiveToken, DocumentStartToken,
+                StreamEndToken):
+            self.tag_handles = self.DEFAULT_TAGS
+            token = self.peek_token()
+            start_mark = end_mark = token.start_mark
+            event = DocumentStartEvent(start_mark, end_mark,
+                    explicit=False)
+
+            # Prepare the next state.
+            self.states.append(self.parse_document_end)
+            self.state = self.parse_block_node
+
+            return event
+
+        else:
+            return self.parse_document_start()
+
+    def parse_document_start(self):
+
+        # Parse any extra document end indicators.
+        while self.check_token(DocumentEndToken):
+            self.get_token()
+
+        # Parse an explicit document.
+        if not self.check_token(StreamEndToken):
+            token = self.peek_token()
+            start_mark = token.start_mark
+            version, tags = self.process_directives()
+            if not self.check_token(DocumentStartToken):
+                raise ParserError(None, None,
+                        "expected '<document start>', but found %r"
+                        % self.peek_token().id,
+                        self.peek_token().start_mark)
+            token = self.get_token()
+            end_mark = token.end_mark
+            event = DocumentStartEvent(start_mark, end_mark,
+                    explicit=True, version=version, tags=tags)
+            self.states.append(self.parse_document_end)
+            self.state = self.parse_document_content
+        else:
+            # Parse the end of the stream.
+            token = self.get_token()
+            event = StreamEndEvent(token.start_mark, token.end_mark)
+            assert not self.states
+            assert not self.marks
+            self.state = None
+        return event
+
+    def parse_document_end(self):
+
+        # Parse the document end.
+        token = self.peek_token()
+        start_mark = end_mark = token.start_mark
+        explicit = False
+        if self.check_token(DocumentEndToken):
+            token = self.get_token()
+            end_mark = token.end_mark
+            explicit = True
+        event = DocumentEndEvent(start_mark, end_mark,
+                explicit=explicit)
+
+        # Prepare the next state.
+        self.state = self.parse_document_start
+
+        return event
+
+    def parse_document_content(self):
+        if self.check_token(DirectiveToken,
+                DocumentStartToken, DocumentEndToken, StreamEndToken):
+            event = self.process_empty_scalar(self.peek_token().start_mark)
+            self.state = self.states.pop()
+            return event
+        else:
+            return self.parse_block_node()
+
+    def process_directives(self):
+        self.yaml_version = None
+        self.tag_handles = {}
+        while self.check_token(DirectiveToken):
+            token = self.get_token()
+            if token.name == 'YAML':
+                if self.yaml_version is not None:
+                    raise ParserError(None, None,
+                            "found duplicate YAML directive", token.start_mark)
+                major, minor = token.value
+                if major != 1:
+                    raise ParserError(None, None,
+                            "found incompatible YAML document (version 1.* is required)",
+                            token.start_mark)
+                self.yaml_version = token.value
+            elif token.name == 'TAG':
+                handle, prefix = token.value
+                if handle in self.tag_handles:
+                    raise ParserError(None, None,
+                            "duplicate tag handle %r" % handle,
+                            token.start_mark)
+                self.tag_handles[handle] = prefix
+        if self.tag_handles:
+            value = self.yaml_version, self.tag_handles.copy()
+        else:
+            value = self.yaml_version, None
+        for key in self.DEFAULT_TAGS:
+            if key not in self.tag_handles:
+                self.tag_handles[key] = self.DEFAULT_TAGS[key]
+        return value
+
+    # block_node_or_indentless_sequence ::= ALIAS
+    #               | properties (block_content | indentless_block_sequence)?
+    #               | block_content
+    #               | indentless_block_sequence
+    # block_node    ::= ALIAS
+    #                   | properties block_content?
+    #                   | block_content
+    # flow_node     ::= ALIAS
+    #                   | properties flow_content?
+    #                   | flow_content
+    # properties    ::= TAG ANCHOR? | ANCHOR TAG?
+    # block_content     ::= block_collection | flow_collection | SCALAR
+    # flow_content      ::= flow_collection | SCALAR
+    # block_collection  ::= block_sequence | block_mapping
+    # flow_collection   ::= flow_sequence | flow_mapping
+
+    def parse_block_node(self):
+        return self.parse_node(block=True)
+
+    def parse_flow_node(self):
+        return self.parse_node()
+
+    def parse_block_node_or_indentless_sequence(self):
+        return self.parse_node(block=True, indentless_sequence=True)
+
+    def parse_node(self, block=False, indentless_sequence=False):
+        if self.check_token(AliasToken):
+            token = self.get_token()
+            event = AliasEvent(token.value, token.start_mark, token.end_mark)
+            self.state = self.states.pop()
+        else:
+            anchor = None
+            tag = None
+            start_mark = end_mark = tag_mark = None
+            if self.check_token(AnchorToken):
+                token = self.get_token()
+                start_mark = token.start_mark
+                end_mark = token.end_mark
+                anchor = token.value
+                if self.check_token(TagToken):
+                    token = self.get_token()
+                    tag_mark = token.start_mark
+                    end_mark = token.end_mark
+                    tag = token.value
+            elif self.check_token(TagToken):
+                token = self.get_token()
+                start_mark = tag_mark = token.start_mark
+                end_mark = token.end_mark
+                tag = token.value
+                if self.check_token(AnchorToken):
+                    token = self.get_token()
+                    end_mark = token.end_mark
+                    anchor = token.value
+            if tag is not None:
+                handle, suffix = tag
+                if handle is not None:
+                    if handle not in self.tag_handles:
+                        raise ParserError("while parsing a node", start_mark,
+                                "found undefined tag handle %r" % handle,
+                                tag_mark)
+                    tag = self.tag_handles[handle]+suffix
+                else:
+                    tag = suffix
+            #if tag == '!':
+            #    raise ParserError("while parsing a node", start_mark,
+            #            "found non-specific tag '!'", tag_mark,
+            #            "Please check 'http://pyyaml.org/wiki/YAMLNonSpecificTag' and share your opinion.")
+            if start_mark is None:
+                start_mark = end_mark = self.peek_token().start_mark
+            event = None
+            implicit = (tag is None or tag == '!')
+            if indentless_sequence and self.check_token(BlockEntryToken):
+                end_mark = self.peek_token().end_mark
+                event = SequenceStartEvent(anchor, tag, implicit,
+                        start_mark, end_mark)
+                self.state = self.parse_indentless_sequence_entry
+            else:
+                if self.check_token(ScalarToken):
+                    token = self.get_token()
+                    end_mark = token.end_mark
+                    if (token.plain and tag is None) or tag == '!':
+                        implicit = (True, False)
+                    elif tag is None:
+                        implicit = (False, True)
+                    else:
+                        implicit = (False, False)
+                    event = ScalarEvent(anchor, tag, implicit, token.value,
+                            start_mark, end_mark, style=token.style)
+                    self.state = self.states.pop()
+                elif self.check_token(FlowSequenceStartToken):
+                    end_mark = self.peek_token().end_mark
+                    event = SequenceStartEvent(anchor, tag, implicit,
+                            start_mark, end_mark, flow_style=True)
+                    self.state = self.parse_flow_sequence_first_entry
+                elif self.check_token(FlowMappingStartToken):
+                    end_mark = self.peek_token().end_mark
+                    event = MappingStartEvent(anchor, tag, implicit,
+                            start_mark, end_mark, flow_style=True)
+                    self.state = self.parse_flow_mapping_first_key
+                elif block and self.check_token(BlockSequenceStartToken):
+                    end_mark = self.peek_token().start_mark
+                    event = SequenceStartEvent(anchor, tag, implicit,
+                            start_mark, end_mark, flow_style=False)
+                    self.state = self.parse_block_sequence_first_entry
+                elif block and self.check_token(BlockMappingStartToken):
+                    end_mark = self.peek_token().start_mark
+                    event = MappingStartEvent(anchor, tag, implicit,
+                            start_mark, end_mark, flow_style=False)
+                    self.state = self.parse_block_mapping_first_key
+                elif anchor is not None or tag is not None:
+                    # Empty scalars are allowed even if a tag or an anchor is
+                    # specified.
+                    event = ScalarEvent(anchor, tag, (implicit, False), '',
+                            start_mark, end_mark)
+                    self.state = self.states.pop()
+                else:
+                    if block:
+                        node = 'block'
+                    else:
+                        node = 'flow'
+                    token = self.peek_token()
+                    raise ParserError("while parsing a %s node" % node, start_mark,
+                            "expected the node content, but found %r" % token.id,
+                            token.start_mark)
+        return event
+
+    # block_sequence ::= BLOCK-SEQUENCE-START (BLOCK-ENTRY block_node?)* BLOCK-END
+
+    def parse_block_sequence_first_entry(self):
+        token = self.get_token()
+        self.marks.append(token.start_mark)
+        return self.parse_block_sequence_entry()
+
+    def parse_block_sequence_entry(self):
+        if self.check_token(BlockEntryToken):
+            token = self.get_token()
+            if not self.check_token(BlockEntryToken, BlockEndToken):
+                self.states.append(self.parse_block_sequence_entry)
+                return self.parse_block_node()
+            else:
+                self.state = self.parse_block_sequence_entry
+                return self.process_empty_scalar(token.end_mark)
+        if not self.check_token(BlockEndToken):
+            token = self.peek_token()
+            raise ParserError("while parsing a block collection", self.marks[-1],
+                    "expected <block end>, but found %r" % token.id, token.start_mark)
+        token = self.get_token()
+        event = SequenceEndEvent(token.start_mark, token.end_mark)
+        self.state = self.states.pop()
+        self.marks.pop()
+        return event
+
+    # indentless_sequence ::= (BLOCK-ENTRY block_node?)+
+
+    def parse_indentless_sequence_entry(self):
+        if self.check_token(BlockEntryToken):
+            token = self.get_token()
+            if not self.check_token(BlockEntryToken,
+                    KeyToken, ValueToken, BlockEndToken):
+                self.states.append(self.parse_indentless_sequence_entry)
+                return self.parse_block_node()
+            else:
+                self.state = self.parse_indentless_sequence_entry
+                return self.process_empty_scalar(token.end_mark)
+        token = self.peek_token()
+        event = SequenceEndEvent(token.start_mark, token.start_mark)
+        self.state = self.states.pop()
+        return event
+
+    # block_mapping     ::= BLOCK-MAPPING_START
+    #                       ((KEY block_node_or_indentless_sequence?)?
+    #                       (VALUE block_node_or_indentless_sequence?)?)*
+    #                       BLOCK-END
+
+    def parse_block_mapping_first_key(self):
+        token = self.get_token()
+        self.marks.append(token.start_mark)
+        return self.parse_block_mapping_key()
+
+    def parse_block_mapping_key(self):
+        if self.check_token(KeyToken):
+            token = self.get_token()
+            if not self.check_token(KeyToken, ValueToken, BlockEndToken):
+                self.states.append(self.parse_block_mapping_value)
+                return self.parse_block_node_or_indentless_sequence()
+            else:
+                self.state = self.parse_block_mapping_value
+                return self.process_empty_scalar(token.end_mark)
+        if not self.check_token(BlockEndToken):
+            token = self.peek_token()
+            raise ParserError("while parsing a block mapping", self.marks[-1],
+                    "expected <block end>, but found %r" % token.id, token.start_mark)
+        token = self.get_token()
+        event = MappingEndEvent(token.start_mark, token.end_mark)
+        self.state = self.states.pop()
+        self.marks.pop()
+        return event
+
+    def parse_block_mapping_value(self):
+        if self.check_token(ValueToken):
+            token = self.get_token()
+            if not self.check_token(KeyToken, ValueToken, BlockEndToken):
+                self.states.append(self.parse_block_mapping_key)
+                return self.parse_block_node_or_indentless_sequence()
+            else:
+                self.state = self.parse_block_mapping_key
+                return self.process_empty_scalar(token.end_mark)
+        else:
+            self.state = self.parse_block_mapping_key
+            token = self.peek_token()
+            return self.process_empty_scalar(token.start_mark)
+
+    # flow_sequence     ::= FLOW-SEQUENCE-START
+    #                       (flow_sequence_entry FLOW-ENTRY)*
+    #                       flow_sequence_entry?
+    #                       FLOW-SEQUENCE-END
+    # flow_sequence_entry   ::= flow_node | KEY flow_node? (VALUE flow_node?)?
+    #
+    # Note that while production rules for both flow_sequence_entry and
+    # flow_mapping_entry are equal, their interpretations are different.
+    # For `flow_sequence_entry`, the part `KEY flow_node? (VALUE flow_node?)?`
+    # generate an inline mapping (set syntax).
+
+    def parse_flow_sequence_first_entry(self):
+        token = self.get_token()
+        self.marks.append(token.start_mark)
+        return self.parse_flow_sequence_entry(first=True)
+
+    def parse_flow_sequence_entry(self, first=False):
+        if not self.check_token(FlowSequenceEndToken):
+            if not first:
+                if self.check_token(FlowEntryToken):
+                    self.get_token()
+                else:
+                    token = self.peek_token()
+                    raise ParserError("while parsing a flow sequence", self.marks[-1],
+                            "expected ',' or ']', but got %r" % token.id, token.start_mark)
+            
+            if self.check_token(KeyToken):
+                token = self.peek_token()
+                event = MappingStartEvent(None, None, True,
+                        token.start_mark, token.end_mark,
+                        flow_style=True)
+                self.state = self.parse_flow_sequence_entry_mapping_key
+                return event
+            elif not self.check_token(FlowSequenceEndToken):
+                self.states.append(self.parse_flow_sequence_entry)
+                return self.parse_flow_node()
+        token = self.get_token()
+        event = SequenceEndEvent(token.start_mark, token.end_mark)
+        self.state = self.states.pop()
+        self.marks.pop()
+        return event
+
+    def parse_flow_sequence_entry_mapping_key(self):
+        token = self.get_token()
+        if not self.check_token(ValueToken,
+                FlowEntryToken, FlowSequenceEndToken):
+            self.states.append(self.parse_flow_sequence_entry_mapping_value)
+            return self.parse_flow_node()
+        else:
+            self.state = self.parse_flow_sequence_entry_mapping_value
+            return self.process_empty_scalar(token.end_mark)
+
+    def parse_flow_sequence_entry_mapping_value(self):
+        if self.check_token(ValueToken):
+            token = self.get_token()
+            if not self.check_token(FlowEntryToken, FlowSequenceEndToken):
+                self.states.append(self.parse_flow_sequence_entry_mapping_end)
+                return self.parse_flow_node()
+            else:
+                self.state = self.parse_flow_sequence_entry_mapping_end
+                return self.process_empty_scalar(token.end_mark)
+        else:
+            self.state = self.parse_flow_sequence_entry_mapping_end
+            token = self.peek_token()
+            return self.process_empty_scalar(token.start_mark)
+
+    def parse_flow_sequence_entry_mapping_end(self):
+        self.state = self.parse_flow_sequence_entry
+        token = self.peek_token()
+        return MappingEndEvent(token.start_mark, token.start_mark)
+
+    # flow_mapping  ::= FLOW-MAPPING-START
+    #                   (flow_mapping_entry FLOW-ENTRY)*
+    #                   flow_mapping_entry?
+    #                   FLOW-MAPPING-END
+    # flow_mapping_entry    ::= flow_node | KEY flow_node? (VALUE flow_node?)?
+
+    def parse_flow_mapping_first_key(self):
+        token = self.get_token()
+        self.marks.append(token.start_mark)
+        return self.parse_flow_mapping_key(first=True)
+
+    def parse_flow_mapping_key(self, first=False):
+        if not self.check_token(FlowMappingEndToken):
+            if not first:
+                if self.check_token(FlowEntryToken):
+                    self.get_token()
+                else:
+                    token = self.peek_token()
+                    raise ParserError("while parsing a flow mapping", self.marks[-1],
+                            "expected ',' or '}', but got %r" % token.id, token.start_mark)
+            if self.check_token(KeyToken):
+                token = self.get_token()
+                if not self.check_token(ValueToken,
+                        FlowEntryToken, FlowMappingEndToken):
+                    self.states.append(self.parse_flow_mapping_value)
+                    return self.parse_flow_node()
+                else:
+                    self.state = self.parse_flow_mapping_value
+                    return self.process_empty_scalar(token.end_mark)
+            elif not self.check_token(FlowMappingEndToken):
+                self.states.append(self.parse_flow_mapping_empty_value)
+                return self.parse_flow_node()
+        token = self.get_token()
+        event = MappingEndEvent(token.start_mark, token.end_mark)
+        self.state = self.states.pop()
+        self.marks.pop()
+        return event
+
+    def parse_flow_mapping_value(self):
+        if self.check_token(ValueToken):
+            token = self.get_token()
+            if not self.check_token(FlowEntryToken, FlowMappingEndToken):
+                self.states.append(self.parse_flow_mapping_key)
+                return self.parse_flow_node()
+            else:
+                self.state = self.parse_flow_mapping_key
+                return self.process_empty_scalar(token.end_mark)
+        else:
+            self.state = self.parse_flow_mapping_key
+            token = self.peek_token()
+            return self.process_empty_scalar(token.start_mark)
+
+    def parse_flow_mapping_empty_value(self):
+        self.state = self.parse_flow_mapping_key
+        return self.process_empty_scalar(self.peek_token().start_mark)
+
+    def process_empty_scalar(self, mark):
+        return ScalarEvent(None, None, (True, False), '', mark, mark)
+

tablib/packages/yaml3/reader.py

@@ -0,0 +1,192 @@
+# This module contains abstractions for the input stream. You don't have to
+# looks further, there are no pretty code.
+#
+# We define two classes here.
+#
+#   Mark(source, line, column)
+# It's just a record and its only use is producing nice error messages.
+# Parser does not use it for any other purposes.
+#
+#   Reader(source, data)
+# Reader determines the encoding of `data` and converts it to unicode.
+# Reader provides the following methods and attributes:
+#   reader.peek(length=1) - return the next `length` characters
+#   reader.forward(length=1) - move the current position to `length` characters.
+#   reader.index - the number of the current character.
+#   reader.line, stream.column - the line and the column of the current character.
+
+__all__ = ['Reader', 'ReaderError']
+
+from .error import YAMLError, Mark
+
+import codecs, re
+
+class ReaderError(YAMLError):
+
+    def __init__(self, name, position, character, encoding, reason):
+        self.name = name
+        self.character = character
+        self.position = position
+        self.encoding = encoding
+        self.reason = reason
+
+    def __str__(self):
+        if isinstance(self.character, bytes):
+            return "'%s' codec can't decode byte #x%02x: %s\n"  \
+                    "  in \"%s\", position %d"    \
+                    % (self.encoding, ord(self.character), self.reason,
+                            self.name, self.position)
+        else:
+            return "unacceptable character #x%04x: %s\n"    \
+                    "  in \"%s\", position %d"    \
+                    % (self.character, self.reason,
+                            self.name, self.position)
+
+class Reader(object):
+    # Reader:
+    # - determines the data encoding and converts it to a unicode string,
+    # - checks if characters are in allowed range,
+    # - adds '\0' to the end.
+
+    # Reader accepts
+    #  - a `bytes` object,
+    #  - a `str` object,
+    #  - a file-like object with its `read` method returning `str`,
+    #  - a file-like object with its `read` method returning `unicode`.
+
+    # Yeah, it's ugly and slow.
+
+    def __init__(self, stream):
+        self.name = None
+        self.stream = None
+        self.stream_pointer = 0
+        self.eof = True
+        self.buffer = ''
+        self.pointer = 0
+        self.raw_buffer = None
+        self.raw_decode = None
+        self.encoding = None
+        self.index = 0
+        self.line = 0
+        self.column = 0
+        if isinstance(stream, str):
+            self.name = "<unicode string>"
+            self.check_printable(stream)
+            self.buffer = stream+'\0'
+        elif isinstance(stream, bytes):
+            self.name = "<byte string>"
+            self.raw_buffer = stream
+            self.determine_encoding()
+        else:
+            self.stream = stream
+            self.name = getattr(stream, 'name', "<file>")
+            self.eof = False
+            self.raw_buffer = None
+            self.determine_encoding()
+
+    def peek(self, index=0):
+        try:
+            return self.buffer[self.pointer+index]
+        except IndexError:
+            self.update(index+1)
+            return self.buffer[self.pointer+index]
+
+    def prefix(self, length=1):
+        if self.pointer+length >= len(self.buffer):
+            self.update(length)
+        return self.buffer[self.pointer:self.pointer+length]
+
+    def forward(self, length=1):
+        if self.pointer+length+1 >= len(self.buffer):
+            self.update(length+1)
+        while length:
+            ch = self.buffer[self.pointer]
+            self.pointer += 1
+            self.index += 1
+            if ch in '\n\x85\u2028\u2029'  \
+                    or (ch == '\r' and self.buffer[self.pointer] != '\n'):
+                self.line += 1
+                self.column = 0
+            elif ch != '\uFEFF':
+                self.column += 1
+            length -= 1
+
+    def get_mark(self):
+        if self.stream is None:
+            return Mark(self.name, self.index, self.line, self.column,
+                    self.buffer, self.pointer)
+        else:
+            return Mark(self.name, self.index, self.line, self.column,
+                    None, None)
+
+    def determine_encoding(self):
+        while not self.eof and (self.raw_buffer is None or len(self.raw_buffer) < 2):
+            self.update_raw()
+        if isinstance(self.raw_buffer, bytes):
+            if self.raw_buffer.startswith(codecs.BOM_UTF16_LE):
+                self.raw_decode = codecs.utf_16_le_decode
+                self.encoding = 'utf-16-le'
+            elif self.raw_buffer.startswith(codecs.BOM_UTF16_BE):
+                self.raw_decode = codecs.utf_16_be_decode
+                self.encoding = 'utf-16-be'
+            else:
+                self.raw_decode = codecs.utf_8_decode
+                self.encoding = 'utf-8'
+        self.update(1)
+
+    NON_PRINTABLE = re.compile('[^\x09\x0A\x0D\x20-\x7E\x85\xA0-\uD7FF\uE000-\uFFFD]')
+    def check_printable(self, data):
+        match = self.NON_PRINTABLE.search(data)
+        if match:
+            character = match.group()
+            position = self.index+(len(self.buffer)-self.pointer)+match.start()
+            raise ReaderError(self.name, position, ord(character),
+                    'unicode', "special characters are not allowed")
+
+    def update(self, length):
+        if self.raw_buffer is None:
+            return
+        self.buffer = self.buffer[self.pointer:]
+        self.pointer = 0
+        while len(self.buffer) < length:
+            if not self.eof:
+                self.update_raw()
+            if self.raw_decode is not None:
+                try:
+                    data, converted = self.raw_decode(self.raw_buffer,
+                            'strict', self.eof)
+                except UnicodeDecodeError as exc:
+                    character = self.raw_buffer[exc.start]
+                    if self.stream is not None:
+                        position = self.stream_pointer-len(self.raw_buffer)+exc.start
+                    else:
+                        position = exc.start
+                    raise ReaderError(self.name, position, character,
+                            exc.encoding, exc.reason)
+            else:
+                data = self.raw_buffer
+                converted = len(data)
+            self.check_printable(data)
+            self.buffer += data
+            self.raw_buffer = self.raw_buffer[converted:]
+            if self.eof:
+                self.buffer += '\0'
+                self.raw_buffer = None
+                break
+
+    def update_raw(self, size=4096):
+        data = self.stream.read(size)
+        if self.raw_buffer is None:
+            self.raw_buffer = data
+        else:
+            self.raw_buffer += data
+        self.stream_pointer += len(data)
+        if not data:
+            self.eof = True
+
+#try:
+#    import psyco
+#    psyco.bind(Reader)
+#except ImportError:
+#    pass
+

tablib/packages/yaml3/representer.py

@@ -0,0 +1,374 @@
+
+__all__ = ['BaseRepresenter', 'SafeRepresenter', 'Representer',
+    'RepresenterError']
+
+from .error import *
+from .nodes import *
+
+import datetime, sys, copyreg, types, base64
+
+class RepresenterError(YAMLError):
+    pass
+
+class BaseRepresenter:
+
+    yaml_representers = {}
+    yaml_multi_representers = {}
+
+    def __init__(self, default_style=None, default_flow_style=None):
+        self.default_style = default_style
+        self.default_flow_style = default_flow_style
+        self.represented_objects = {}
+        self.object_keeper = []
+        self.alias_key = None
+
+    def represent(self, data):
+        node = self.represent_data(data)
+        self.serialize(node)
+        self.represented_objects = {}
+        self.object_keeper = []
+        self.alias_key = None
+
+    def represent_data(self, data):
+        if self.ignore_aliases(data):
+            self.alias_key = None
+        else:
+            self.alias_key = id(data)
+        if self.alias_key is not None:
+            if self.alias_key in self.represented_objects:
+                node = self.represented_objects[self.alias_key]
+                #if node is None:
+                #    raise RepresenterError("recursive objects are not allowed: %r" % data)
+                return node
+            #self.represented_objects[alias_key] = None
+            self.object_keeper.append(data)
+        data_types = type(data).__mro__
+        if data_types[0] in self.yaml_representers:
+            node = self.yaml_representers[data_types[0]](self, data)
+        else:
+            for data_type in data_types:
+                if data_type in self.yaml_multi_representers:
+                    node = self.yaml_multi_representers[data_type](self, data)
+                    break
+            else:
+                if None in self.yaml_multi_representers:
+                    node = self.yaml_multi_representers[None](self, data)
+                elif None in self.yaml_representers:
+                    node = self.yaml_representers[None](self, data)
+                else:
+                    node = ScalarNode(None, str(data))
+        #if alias_key is not None:
+        #    self.represented_objects[alias_key] = node
+        return node
+
+    @classmethod
+    def add_representer(cls, data_type, representer):
+        if not 'yaml_representers' in cls.__dict__:
+            cls.yaml_representers = cls.yaml_representers.copy()
+        cls.yaml_representers[data_type] = representer
+
+    @classmethod
+    def add_multi_representer(cls, data_type, representer):
+        if not 'yaml_multi_representers' in cls.__dict__:
+            cls.yaml_multi_representers = cls.yaml_multi_representers.copy()
+        cls.yaml_multi_representers[data_type] = representer
+
+    def represent_scalar(self, tag, value, style=None):
+        if style is None:
+            style = self.default_style
+        node = ScalarNode(tag, value, style=style)
+        if self.alias_key is not None:
+            self.represented_objects[self.alias_key] = node
+        return node
+
+    def represent_sequence(self, tag, sequence, flow_style=None):
+        value = []
+        node = SequenceNode(tag, value, flow_style=flow_style)
+        if self.alias_key is not None:
+            self.represented_objects[self.alias_key] = node
+        best_style = True
+        for item in sequence:
+            node_item = self.represent_data(item)
+            if not (isinstance(node_item, ScalarNode) and not node_item.style):
+                best_style = False
+            value.append(node_item)
+        if flow_style is None:
+            if self.default_flow_style is not None:
+                node.flow_style = self.default_flow_style
+            else:
+                node.flow_style = best_style
+        return node
+
+    def represent_mapping(self, tag, mapping, flow_style=None):
+        value = []
+        node = MappingNode(tag, value, flow_style=flow_style)
+        if self.alias_key is not None:
+            self.represented_objects[self.alias_key] = node
+        best_style = True
+        if hasattr(mapping, 'items'):
+            mapping = list(mapping.items())
+            try:
+                mapping = sorted(mapping)
+            except TypeError:
+                pass
+        for item_key, item_value in mapping:
+            node_key = self.represent_data(item_key)
+            node_value = self.represent_data(item_value)
+            if not (isinstance(node_key, ScalarNode) and not node_key.style):
+                best_style = False
+            if not (isinstance(node_value, ScalarNode) and not node_value.style):
+                best_style = False
+            value.append((node_key, node_value))
+        if flow_style is None:
+            if self.default_flow_style is not None:
+                node.flow_style = self.default_flow_style
+            else:
+                node.flow_style = best_style
+        return node
+
+    def ignore_aliases(self, data):
+        return False
+
+class SafeRepresenter(BaseRepresenter):
+
+    def ignore_aliases(self, data):
+        if data in [None, ()]:
+            return True
+        if isinstance(data, (str, bytes, bool, int, float)):
+            return True
+
+    def represent_none(self, data):
+        return self.represent_scalar('tag:yaml.org,2002:null', 'null')
+
+    def represent_str(self, data):
+        return self.represent_scalar('tag:yaml.org,2002:str', data)
+
+    def represent_binary(self, data):
+        if hasattr(base64, 'encodebytes'):
+            data = base64.encodebytes(data).decode('ascii')
+        else:
+            data = base64.encodestring(data).decode('ascii')
+        return self.represent_scalar('tag:yaml.org,2002:binary', data, style='|')
+
+    def represent_bool(self, data):
+        if data:
+            value = 'true'
+        else:
+            value = 'false'
+        return self.represent_scalar('tag:yaml.org,2002:bool', value)
+
+    def represent_int(self, data):
+        return self.represent_scalar('tag:yaml.org,2002:int', str(data))
+
+    inf_value = 1e300
+    while repr(inf_value) != repr(inf_value*inf_value):
+        inf_value *= inf_value
+
+    def represent_float(self, data):
+        if data != data or (data == 0.0 and data == 1.0):
+            value = '.nan'
+        elif data == self.inf_value:
+            value = '.inf'
+        elif data == -self.inf_value:
+            value = '-.inf'
+        else:
+            value = repr(data).lower()
+            # Note that in some cases `repr(data)` represents a float number
+            # without the decimal parts.  For instance:
+            #   >>> repr(1e17)
+            #   '1e17'
+            # Unfortunately, this is not a valid float representation according
+            # to the definition of the `!!float` tag.  We fix this by adding
+            # '.0' before the 'e' symbol.
+            if '.' not in value and 'e' in value:
+                value = value.replace('e', '.0e', 1)
+        return self.represent_scalar('tag:yaml.org,2002:float', value)
+
+    def represent_list(self, data):
+        #pairs = (len(data) > 0 and isinstance(data, list))
+        #if pairs:
+        #    for item in data:
+        #        if not isinstance(item, tuple) or len(item) != 2:
+        #            pairs = False
+        #            break
+        #if not pairs:
+            return self.represent_sequence('tag:yaml.org,2002:seq', data)
+        #value = []
+        #for item_key, item_value in data:
+        #    value.append(self.represent_mapping(u'tag:yaml.org,2002:map',
+        #        [(item_key, item_value)]))
+        #return SequenceNode(u'tag:yaml.org,2002:pairs', value)
+
+    def represent_dict(self, data):
+        return self.represent_mapping('tag:yaml.org,2002:map', data)
+
+    def represent_set(self, data):
+        value = {}
+        for key in data:
+            value[key] = None
+        return self.represent_mapping('tag:yaml.org,2002:set', value)
+
+    def represent_date(self, data):
+        value = data.isoformat()
+        return self.represent_scalar('tag:yaml.org,2002:timestamp', value)
+
+    def represent_datetime(self, data):
+        value = data.isoformat(' ')
+        return self.represent_scalar('tag:yaml.org,2002:timestamp', value)
+
+    def represent_yaml_object(self, tag, data, cls, flow_style=None):
+        if hasattr(data, '__getstate__'):
+            state = data.__getstate__()
+        else:
+            state = data.__dict__.copy()
+        return self.represent_mapping(tag, state, flow_style=flow_style)
+
+    def represent_undefined(self, data):
+        raise RepresenterError("cannot represent an object: %s" % data)
+
+SafeRepresenter.add_representer(type(None),
+        SafeRepresenter.represent_none)
+
+SafeRepresenter.add_representer(str,
+        SafeRepresenter.represent_str)
+
+SafeRepresenter.add_representer(bytes,
+        SafeRepresenter.represent_binary)
+
+SafeRepresenter.add_representer(bool,
+        SafeRepresenter.represent_bool)
+
+SafeRepresenter.add_representer(int,
+        SafeRepresenter.represent_int)
+
+SafeRepresenter.add_representer(float,
+        SafeRepresenter.represent_float)
+
+SafeRepresenter.add_representer(list,
+        SafeRepresenter.represent_list)
+
+SafeRepresenter.add_representer(tuple,
+        SafeRepresenter.represent_list)
+
+SafeRepresenter.add_representer(dict,
+        SafeRepresenter.represent_dict)
+
+SafeRepresenter.add_representer(set,
+        SafeRepresenter.represent_set)
+
+SafeRepresenter.add_representer(datetime.date,
+        SafeRepresenter.represent_date)
+
+SafeRepresenter.add_representer(datetime.datetime,
+        SafeRepresenter.represent_datetime)
+
+SafeRepresenter.add_representer(None,
+        SafeRepresenter.represent_undefined)
+
+class Representer(SafeRepresenter):
+
+    def represent_complex(self, data):
+        if data.imag == 0.0:
+            data = '%r' % data.real
+        elif data.real == 0.0:
+            data = '%rj' % data.imag
+        elif data.imag > 0:
+            data = '%r+%rj' % (data.real, data.imag)
+        else:
+            data = '%r%rj' % (data.real, data.imag)
+        return self.represent_scalar('tag:yaml.org,2002:python/complex', data)
+
+    def represent_tuple(self, data):
+        return self.represent_sequence('tag:yaml.org,2002:python/tuple', data)
+
+    def represent_name(self, data):
+        name = '%s.%s' % (data.__module__, data.__name__)
+        return self.represent_scalar('tag:yaml.org,2002:python/name:'+name, '')
+
+    def represent_module(self, data):
+        return self.represent_scalar(
+                'tag:yaml.org,2002:python/module:'+data.__name__, '')
+
+    def represent_object(self, data):
+        # We use __reduce__ API to save the data. data.__reduce__ returns
+        # a tuple of length 2-5:
+        #   (function, args, state, listitems, dictitems)
+
+        # For reconstructing, we calls function(*args), then set its state,
+        # listitems, and dictitems if they are not None.
+
+        # A special case is when function.__name__ == '__newobj__'. In this
+        # case we create the object with args[0].__new__(*args).
+
+        # Another special case is when __reduce__ returns a string - we don't
+        # support it.
+
+        # We produce a !!python/object, !!python/object/new or
+        # !!python/object/apply node.
+
+        cls = type(data)
+        if cls in copyreg.dispatch_table:
+            reduce = copyreg.dispatch_table[cls](data)
+        elif hasattr(data, '__reduce_ex__'):
+            reduce = data.__reduce_ex__(2)
+        elif hasattr(data, '__reduce__'):
+            reduce = data.__reduce__()
+        else:
+            raise RepresenterError("cannot represent object: %r" % data)
+        reduce = (list(reduce)+[None]*5)[:5]
+        function, args, state, listitems, dictitems = reduce
+        args = list(args)
+        if state is None:
+            state = {}
+        if listitems is not None:
+            listitems = list(listitems)
+        if dictitems is not None:
+            dictitems = dict(dictitems)
+        if function.__name__ == '__newobj__':
+            function = args[0]
+            args = args[1:]
+            tag = 'tag:yaml.org,2002:python/object/new:'
+            newobj = True
+        else:
+            tag = 'tag:yaml.org,2002:python/object/apply:'
+            newobj = False
+        function_name = '%s.%s' % (function.__module__, function.__name__)
+        if not args and not listitems and not dictitems \
+                and isinstance(state, dict) and newobj:
+            return self.represent_mapping(
+                    'tag:yaml.org,2002:python/object:'+function_name, state)
+        if not listitems and not dictitems  \
+                and isinstance(state, dict) and not state:
+            return self.represent_sequence(tag+function_name, args)
+        value = {}
+        if args:
+            value['args'] = args
+        if state or not isinstance(state, dict):
+            value['state'] = state
+        if listitems:
+            value['listitems'] = listitems
+        if dictitems:
+            value['dictitems'] = dictitems
+        return self.represent_mapping(tag+function_name, value)
+
+Representer.add_representer(complex,
+        Representer.represent_complex)
+
+Representer.add_representer(tuple,
+        Representer.represent_tuple)
+
+Representer.add_representer(type,
+        Representer.represent_name)
+
+Representer.add_representer(types.FunctionType,
+        Representer.represent_name)
+
+Representer.add_representer(types.BuiltinFunctionType,
+        Representer.represent_name)
+
+Representer.add_representer(types.ModuleType,
+        Representer.represent_module)
+
+Representer.add_multi_representer(object,
+        Representer.represent_object)
+

tablib/packages/yaml3/resolver.py

@@ -0,0 +1,224 @@
+
+__all__ = ['BaseResolver', 'Resolver']
+
+from .error import *
+from .nodes import *
+
+import re
+
+class ResolverError(YAMLError):
+    pass
+
+class BaseResolver:
+
+    DEFAULT_SCALAR_TAG = 'tag:yaml.org,2002:str'
+    DEFAULT_SEQUENCE_TAG = 'tag:yaml.org,2002:seq'
+    DEFAULT_MAPPING_TAG = 'tag:yaml.org,2002:map'
+
+    yaml_implicit_resolvers = {}
+    yaml_path_resolvers = {}
+
+    def __init__(self):
+        self.resolver_exact_paths = []
+        self.resolver_prefix_paths = []
+
+    @classmethod
+    def add_implicit_resolver(cls, tag, regexp, first):
+        if not 'yaml_implicit_resolvers' in cls.__dict__:
+            cls.yaml_implicit_resolvers = cls.yaml_implicit_resolvers.copy()
+        if first is None:
+            first = [None]
+        for ch in first:
+            cls.yaml_implicit_resolvers.setdefault(ch, []).append((tag, regexp))
+
+    @classmethod
+    def add_path_resolver(cls, tag, path, kind=None):
+        # Note: `add_path_resolver` is experimental.  The API could be changed.
+        # `new_path` is a pattern that is matched against the path from the
+        # root to the node that is being considered.  `node_path` elements are
+        # tuples `(node_check, index_check)`.  `node_check` is a node class:
+        # `ScalarNode`, `SequenceNode`, `MappingNode` or `None`.  `None`
+        # matches any kind of a node.  `index_check` could be `None`, a boolean
+        # value, a string value, or a number.  `None` and `False` match against
+        # any _value_ of sequence and mapping nodes.  `True` matches against
+        # any _key_ of a mapping node.  A string `index_check` matches against
+        # a mapping value that corresponds to a scalar key which content is
+        # equal to the `index_check` value.  An integer `index_check` matches
+        # against a sequence value with the index equal to `index_check`.
+        if not 'yaml_path_resolvers' in cls.__dict__:
+            cls.yaml_path_resolvers = cls.yaml_path_resolvers.copy()
+        new_path = []
+        for element in path:
+            if isinstance(element, (list, tuple)):
+                if len(element) == 2:
+                    node_check, index_check = element
+                elif len(element) == 1:
+                    node_check = element[0]
+                    index_check = True
+                else:
+                    raise ResolverError("Invalid path element: %s" % element)
+            else:
+                node_check = None
+                index_check = element
+            if node_check is str:
+                node_check = ScalarNode
+            elif node_check is list:
+                node_check = SequenceNode
+            elif node_check is dict:
+                node_check = MappingNode
+            elif node_check not in [ScalarNode, SequenceNode, MappingNode]  \
+                    and not isinstance(node_check, str) \
+                    and node_check is not None:
+                raise ResolverError("Invalid node checker: %s" % node_check)
+            if not isinstance(index_check, (str, int))  \
+                    and index_check is not None:
+                raise ResolverError("Invalid index checker: %s" % index_check)
+            new_path.append((node_check, index_check))
+        if kind is str:
+            kind = ScalarNode
+        elif kind is list:
+            kind = SequenceNode
+        elif kind is dict:
+            kind = MappingNode
+        elif kind not in [ScalarNode, SequenceNode, MappingNode]    \
+                and kind is not None:
+            raise ResolverError("Invalid node kind: %s" % kind)
+        cls.yaml_path_resolvers[tuple(new_path), kind] = tag
+
+    def descend_resolver(self, current_node, current_index):
+        if not self.yaml_path_resolvers:
+            return
+        exact_paths = {}
+        prefix_paths = []
+        if current_node:
+            depth = len(self.resolver_prefix_paths)
+            for path, kind in self.resolver_prefix_paths[-1]:
+                if self.check_resolver_prefix(depth, path, kind,
+                        current_node, current_index):
+                    if len(path) > depth:
+                        prefix_paths.append((path, kind))
+                    else:
+                        exact_paths[kind] = self.yaml_path_resolvers[path, kind]
+        else:
+            for path, kind in self.yaml_path_resolvers:
+                if not path:
+                    exact_paths[kind] = self.yaml_path_resolvers[path, kind]
+                else:
+                    prefix_paths.append((path, kind))
+        self.resolver_exact_paths.append(exact_paths)
+        self.resolver_prefix_paths.append(prefix_paths)
+
+    def ascend_resolver(self):
+        if not self.yaml_path_resolvers:
+            return
+        self.resolver_exact_paths.pop()
+        self.resolver_prefix_paths.pop()
+
+    def check_resolver_prefix(self, depth, path, kind,
+            current_node, current_index):
+        node_check, index_check = path[depth-1]
+        if isinstance(node_check, str):
+            if current_node.tag != node_check:
+                return
+        elif node_check is not None:
+            if not isinstance(current_node, node_check):
+                return
+        if index_check is True and current_index is not None:
+            return
+        if (index_check is False or index_check is None)    \
+                and current_index is None:
+            return
+        if isinstance(index_check, str):
+            if not (isinstance(current_index, ScalarNode)
+                    and index_check == current_index.value):
+                return
+        elif isinstance(index_check, int) and not isinstance(index_check, bool):
+            if index_check != current_index:
+                return
+        return True
+
+    def resolve(self, kind, value, implicit):
+        if kind is ScalarNode and implicit[0]:
+            if value == '':
+                resolvers = self.yaml_implicit_resolvers.get('', [])
+            else:
+                resolvers = self.yaml_implicit_resolvers.get(value[0], [])
+            resolvers += self.yaml_implicit_resolvers.get(None, [])
+            for tag, regexp in resolvers:
+                if regexp.match(value):
+                    return tag
+            implicit = implicit[1]
+        if self.yaml_path_resolvers:
+            exact_paths = self.resolver_exact_paths[-1]
+            if kind in exact_paths:
+                return exact_paths[kind]
+            if None in exact_paths:
+                return exact_paths[None]
+        if kind is ScalarNode:
+            return self.DEFAULT_SCALAR_TAG
+        elif kind is SequenceNode:
+            return self.DEFAULT_SEQUENCE_TAG
+        elif kind is MappingNode:
+            return self.DEFAULT_MAPPING_TAG
+
+class Resolver(BaseResolver):
+    pass
+
+Resolver.add_implicit_resolver(
+        'tag:yaml.org,2002:bool',
+        re.compile(r'''^(?:yes|Yes|YES|no|No|NO
+                    |true|True|TRUE|false|False|FALSE
+                    |on|On|ON|off|Off|OFF)$''', re.X),
+        list('yYnNtTfFoO'))
+
+Resolver.add_implicit_resolver(
+        'tag:yaml.org,2002:float',
+        re.compile(r'''^(?:[-+]?(?:[0-9][0-9_]*)\.[0-9_]*(?:[eE][-+][0-9]+)?
+                    |\.[0-9_]+(?:[eE][-+][0-9]+)?
+                    |[-+]?[0-9][0-9_]*(?::[0-5]?[0-9])+\.[0-9_]*
+                    |[-+]?\.(?:inf|Inf|INF)
+                    |\.(?:nan|NaN|NAN))$''', re.X),
+        list('-+0123456789.'))
+
+Resolver.add_implicit_resolver(
+        'tag:yaml.org,2002:int',
+        re.compile(r'''^(?:[-+]?0b[0-1_]+
+                    |[-+]?0[0-7_]+
+                    |[-+]?(?:0|[1-9][0-9_]*)
+                    |[-+]?0x[0-9a-fA-F_]+
+                    |[-+]?[1-9][0-9_]*(?::[0-5]?[0-9])+)$''', re.X),
+        list('-+0123456789'))
+
+Resolver.add_implicit_resolver(
+        'tag:yaml.org,2002:merge',
+        re.compile(r'^(?:<<)$'),
+        ['<'])
+
+Resolver.add_implicit_resolver(
+        'tag:yaml.org,2002:null',
+        re.compile(r'''^(?: ~
+                    |null|Null|NULL
+                    | )$''', re.X),
+        ['~', 'n', 'N', ''])
+
+Resolver.add_implicit_resolver(
+        'tag:yaml.org,2002:timestamp',
+        re.compile(r'''^(?:[0-9][0-9][0-9][0-9]-[0-9][0-9]-[0-9][0-9]
+                    |[0-9][0-9][0-9][0-9] -[0-9][0-9]? -[0-9][0-9]?
+                     (?:[Tt]|[ \t]+)[0-9][0-9]?
+                     :[0-9][0-9] :[0-9][0-9] (?:\.[0-9]*)?
+                     (?:[ \t]*(?:Z|[-+][0-9][0-9]?(?::[0-9][0-9])?))?)$''', re.X),
+        list('0123456789'))
+
+Resolver.add_implicit_resolver(
+        'tag:yaml.org,2002:value',
+        re.compile(r'^(?:=)$'),
+        ['='])
+
+# The following resolver is only for documentation purposes. It cannot work
+# because plain scalars cannot start with '!', '&', or '*'.
+Resolver.add_implicit_resolver(
+        'tag:yaml.org,2002:yaml',
+        re.compile(r'^(?:!|&|\*)$'),
+        list('!&*'))
+

tablib/packages/yaml3/serializer.py

@@ -0,0 +1,111 @@
+
+__all__ = ['Serializer', 'SerializerError']
+
+from .error import YAMLError
+from .events import *
+from .nodes import *
+
+class SerializerError(YAMLError):
+    pass
+
+class Serializer:
+
+    ANCHOR_TEMPLATE = 'id%03d'
+
+    def __init__(self, encoding=None,
+            explicit_start=None, explicit_end=None, version=None, tags=None):
+        self.use_encoding = encoding
+        self.use_explicit_start = explicit_start
+        self.use_explicit_end = explicit_end
+        self.use_version = version
+        self.use_tags = tags
+        self.serialized_nodes = {}
+        self.anchors = {}
+        self.last_anchor_id = 0
+        self.closed = None
+
+    def open(self):
+        if self.closed is None:
+            self.emit(StreamStartEvent(encoding=self.use_encoding))
+            self.closed = False
+        elif self.closed:
+            raise SerializerError("serializer is closed")
+        else:
+            raise SerializerError("serializer is already opened")
+
+    def close(self):
+        if self.closed is None:
+            raise SerializerError("serializer is not opened")
+        elif not self.closed:
+            self.emit(StreamEndEvent())
+            self.closed = True
+
+    #def __del__(self):
+    #    self.close()
+
+    def serialize(self, node):
+        if self.closed is None:
+            raise SerializerError("serializer is not opened")
+        elif self.closed:
+            raise SerializerError("serializer is closed")
+        self.emit(DocumentStartEvent(explicit=self.use_explicit_start,
+            version=self.use_version, tags=self.use_tags))
+        self.anchor_node(node)
+        self.serialize_node(node, None, None)
+        self.emit(DocumentEndEvent(explicit=self.use_explicit_end))
+        self.serialized_nodes = {}
+        self.anchors = {}
+        self.last_anchor_id = 0
+
+    def anchor_node(self, node):
+        if node in self.anchors:
+            if self.anchors[node] is None:
+                self.anchors[node] = self.generate_anchor(node)
+        else:
+            self.anchors[node] = None
+            if isinstance(node, SequenceNode):
+                for item in node.value:
+                    self.anchor_node(item)
+            elif isinstance(node, MappingNode):
+                for key, value in node.value:
+                    self.anchor_node(key)
+                    self.anchor_node(value)
+
+    def generate_anchor(self, node):
+        self.last_anchor_id += 1
+        return self.ANCHOR_TEMPLATE % self.last_anchor_id
+
+    def serialize_node(self, node, parent, index):
+        alias = self.anchors[node]
+        if node in self.serialized_nodes:
+            self.emit(AliasEvent(alias))
+        else:
+            self.serialized_nodes[node] = True
+            self.descend_resolver(parent, index)
+            if isinstance(node, ScalarNode):
+                detected_tag = self.resolve(ScalarNode, node.value, (True, False))
+                default_tag = self.resolve(ScalarNode, node.value, (False, True))
+                implicit = (node.tag == detected_tag), (node.tag == default_tag)
+                self.emit(ScalarEvent(alias, node.tag, implicit, node.value,
+                    style=node.style))
+            elif isinstance(node, SequenceNode):
+                implicit = (node.tag
+                            == self.resolve(SequenceNode, node.value, True))
+                self.emit(SequenceStartEvent(alias, node.tag, implicit,
+                    flow_style=node.flow_style))
+                index = 0
+                for item in node.value:
+                    self.serialize_node(item, node, index)
+                    index += 1
+                self.emit(SequenceEndEvent())
+            elif isinstance(node, MappingNode):
+                implicit = (node.tag
+                            == self.resolve(MappingNode, node.value, True))
+                self.emit(MappingStartEvent(alias, node.tag, implicit,
+                    flow_style=node.flow_style))
+                for key, value in node.value:
+                    self.serialize_node(key, node, None)
+                    self.serialize_node(value, node, key)
+                self.emit(MappingEndEvent())
+            self.ascend_resolver()
+

tablib/packages/yaml3/tokens.py

@@ -0,0 +1,104 @@
+
+class Token(object):
+    def __init__(self, start_mark, end_mark):
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+    def __repr__(self):
+        attributes = [key for key in self.__dict__
+                if not key.endswith('_mark')]
+        attributes.sort()
+        arguments = ', '.join(['%s=%r' % (key, getattr(self, key))
+                for key in attributes])
+        return '%s(%s)' % (self.__class__.__name__, arguments)
+
+#class BOMToken(Token):
+#    id = '<byte order mark>'
+
+class DirectiveToken(Token):
+    id = '<directive>'
+    def __init__(self, name, value, start_mark, end_mark):
+        self.name = name
+        self.value = value
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+
+class DocumentStartToken(Token):
+    id = '<document start>'
+
+class DocumentEndToken(Token):
+    id = '<document end>'
+
+class StreamStartToken(Token):
+    id = '<stream start>'
+    def __init__(self, start_mark=None, end_mark=None,
+            encoding=None):
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+        self.encoding = encoding
+
+class StreamEndToken(Token):
+    id = '<stream end>'
+
+class BlockSequenceStartToken(Token):
+    id = '<block sequence start>'
+
+class BlockMappingStartToken(Token):
+    id = '<block mapping start>'
+
+class BlockEndToken(Token):
+    id = '<block end>'
+
+class FlowSequenceStartToken(Token):
+    id = '['
+
+class FlowMappingStartToken(Token):
+    id = '{'
+
+class FlowSequenceEndToken(Token):
+    id = ']'
+
+class FlowMappingEndToken(Token):
+    id = '}'
+
+class KeyToken(Token):
+    id = '?'
+
+class ValueToken(Token):
+    id = ':'
+
+class BlockEntryToken(Token):
+    id = '-'
+
+class FlowEntryToken(Token):
+    id = ','
+
+class AliasToken(Token):
+    id = '<alias>'
+    def __init__(self, value, start_mark, end_mark):
+        self.value = value
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+
+class AnchorToken(Token):
+    id = '<anchor>'
+    def __init__(self, value, start_mark, end_mark):
+        self.value = value
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+
+class TagToken(Token):
+    id = '<tag>'
+    def __init__(self, value, start_mark, end_mark):
+        self.value = value
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+
+class ScalarToken(Token):
+    id = '<scalar>'
+    def __init__(self, value, plain, start_mark, end_mark, style=None):
+        self.value = value
+        self.plain = plain
+        self.start_mark = start_mark
+        self.end_mark = end_mark
+        self.style = style
+