xlwt3 package

This commit is contained in:
Kenneth Reitz
2011-03-23 01:47:01 -04:00
parent b07512071e
commit 0e56db632a
18 changed files with 11308 additions and 0 deletions
File diff suppressed because it is too large Load Diff
+258
View File
@@ -0,0 +1,258 @@
# 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
+233
View File
@@ -0,0 +1,233 @@
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
+34
View File
@@ -0,0 +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()
+516
View File
@@ -0,0 +1,516 @@
import sys
import struct
ENCODING = 'utf-16-le'
class Reader:
def __init__(self, filename, dump = False):
self.dump = dump
self.STREAMS = {}
doc = file(filename, 'rb').read()
self.header, self.data = doc[0:512], doc[512:]
del doc
self.__build_header()
self.__build_MSAT()
self.__build_SAT()
self.__build_directory()
self.__build_short_sectors_data()
if len(self.short_sectors_data) > 0:
self.__build_SSAT()
else:
if self.dump and (self.total_ssat_sectors != 0 or self.ssat_start_sid != -2):
print('NOTE: header says that must be', self.total_ssat_sectors, 'short sectors')
print('NOTE: starting at', self.ssat_start_sid, 'sector')
print('NOTE: but file does not contains data in short sectors')
self.ssat_start_sid = -2
self.total_ssat_sectors = 0
self.SSAT = [-2]
for dentry in self.dir_entry_list[1:]:
(did,
sz, name,
t, c,
did_left, did_right, did_root,
dentry_start_sid,
stream_size
) = dentry
stream_data = b''
if stream_size > 0:
if stream_size >= self.min_stream_size:
args = (self.data, self.SAT, dentry_start_sid, self.sect_size)
else:
args = (self.short_sectors_data, self.SSAT, dentry_start_sid, self.short_sect_size)
stream_data = self.get_stream_data(*args)
if name != b'':
# BAD IDEA: names may be equal. NEED use full paths...
self.STREAMS[name] = stream_data
def __build_header(self):
self.doc_magic = self.header[0:8]
if self.doc_magic != b'\xD0\xCF\x11\xE0\xA1\xB1\x1A\xE1':
raise Exception('Not an OLE file.')
self.file_uid = self.header[8:24]
self.rev_num = self.header[24:26]
self.ver_num = self.header[26:28]
self.byte_order = self.header[28:30]
self.log2_sect_size, = struct.unpack('<H', self.header[30:32])
self.log2_short_sect_size, = struct.unpack('<H', self.header[32:34])
self.total_sat_sectors, = struct.unpack('<L', self.header[44:48])
self.dir_start_sid, = struct.unpack('<l', self.header[48:52])
self.min_stream_size, = struct.unpack('<L', self.header[56:60])
self.ssat_start_sid, = struct.unpack('<l', self.header[60:64])
self.total_ssat_sectors, = struct.unpack('<L', self.header[64:68])
self.msat_start_sid, = struct.unpack('<l', self.header[68:72])
self.total_msat_sectors, = struct.unpack('<L', self.header[72:76])
self.sect_size = 1 << self.log2_sect_size
self.short_sect_size = 1 << self.log2_short_sect_size
if self.dump:
print('file magic: ')
print_bin_data(self.doc_magic)
print('file uid: ')
print_bin_data(self.file_uid)
print('revision number: ')
print_bin_data(self.rev_num)
print('version number: ')
print_bin_data(self.ver_num)
print('byte order: ')
print_bin_data(self.byte_order)
print('sector size :', hex(self.sect_size), self.sect_size)
#print 'total sectors in file :', hex(self.total_sectors), self.total_sectors
print('short sector size :', hex(self.short_sect_size), self.short_sect_size)
print('Total number of sectors used for the SAT :', hex(self.total_sat_sectors), self.total_sat_sectors)
print('SID of first sector of the directory stream:', hex(self.dir_start_sid), self.dir_start_sid)
print('Minimum size of a standard stream :', hex(self.min_stream_size), self.min_stream_size)
print('SID of first sector of the SSAT :', hex(self.ssat_start_sid), self.ssat_start_sid)
print('Total number of sectors used for the SSAT :', hex(self.total_ssat_sectors), self.total_ssat_sectors)
print('SID of first additional sector of the MSAT :', hex(self.msat_start_sid), self.msat_start_sid)
print('Total number of sectors used for the MSAT :', hex(self.total_msat_sectors), self.total_msat_sectors)
def __build_MSAT(self):
self.MSAT = list(struct.unpack('<109l', self.header[76:]))
next = self.msat_start_sid
while next > 0:
msat_sector = struct.unpack('<128l', self.data[next*self.sect_size:(next+1)*self.sect_size])
self.MSAT.extend(msat_sector[:127])
next = msat_sector[-1]
if self.dump:
print('MSAT (header part): \n', self.MSAT[:109])
print('additional MSAT sectors: \n', self.MSAT[109:])
def __build_SAT(self):
sat_stream = b''.join([self.data[i*self.sect_size:(i+1)*self.sect_size] for i in self.MSAT if i >= 0])
sat_sids_count = len(sat_stream) >> 2
self.SAT = struct.unpack('<%dl' % sat_sids_count, sat_stream) # SIDs tuple
if self.dump:
print('SAT sid count:\n', sat_sids_count)
print('SAT content:\n', self.SAT)
def __build_SSAT(self):
ssat_stream = self.get_stream_data(self.data, self.SAT, self.ssat_start_sid, self.sect_size)
ssids_count = len(ssat_stream) >> 2
self.SSAT = struct.unpack('<%dl' % ssids_count, ssat_stream)
if self.dump:
print('SSID count:', ssids_count)
print('SSAT content:\n', self.SSAT)
def __build_directory(self):
dir_stream = self.get_stream_data(self.data, self.SAT, self.dir_start_sid, self.sect_size)
self.dir_entry_list = []
i = 0
while i < len(dir_stream):
dentry = dir_stream[i:i+128] # 128 -- dir entry size
i += 128
did = len(self.dir_entry_list)
sz, = struct.unpack('<H', dentry[64:66])
if sz > 0 :
name = dentry[0:sz-2].decode('utf_16_le', 'replace')
else:
name = b''
t, = struct.unpack('B', dentry[66])
c, = struct.unpack('B', dentry[67])
did_left , = struct.unpack('<l', dentry[68:72])
did_right , = struct.unpack('<l', dentry[72:76])
did_root , = struct.unpack('<l', dentry[76:80])
dentry_start_sid , = struct.unpack('<l', dentry[116:120])
stream_size , = struct.unpack('<L', dentry[120:124])
self.dir_entry_list.extend([(did, sz, name, t, c,
did_left, did_right, did_root,
dentry_start_sid, stream_size)])
if self.dump:
dentry_types = {
0x00: 'Empty',
0x01: 'User storage',
0x02: 'User stream',
0x03: 'LockBytes',
0x04: 'Property',
0x05: 'Root storage'
}
node_colours = {
0x00: 'Red',
0x01: 'Black'
}
print('total directory entries:', len(self.dir_entry_list))
for dentry in self.dir_entry_list:
(did, sz, name, t, c,
did_left, did_right, did_root,
dentry_start_sid, stream_size) = dentry
print('DID', did)
print('Size of the used area of the character buffer of the name:', sz)
print('dir entry name:', repr(name))
print('type of entry:', t, dentry_types[t])
print('entry colour:', c, node_colours[c])
print('left child DID :', did_left)
print('right child DID:', did_right)
print('root DID :', did_root)
print('start SID :', dentry_start_sid)
print('stream size :', stream_size)
if stream_size == 0:
print('stream is empty')
elif stream_size >= self.min_stream_size:
print('stream stored as normal stream')
else:
print('stream stored as short-stream')
def __build_short_sectors_data(self):
(did, sz, name, t, c,
did_left, did_right, did_root,
dentry_start_sid, stream_size) = self.dir_entry_list[0]
assert t == 0x05 # Short-Stream Container Stream (SSCS) resides in Root Storage
if stream_size == 0:
self.short_sectors_data = b''
else:
self.short_sectors_data = self.get_stream_data(self.data, self.SAT, dentry_start_sid, self.sect_size)
def get_stream_data(self, data, SAT, start_sid, sect_size):
sid = start_sid
chunks = [(sid, sid)]
stream_data = b''
while SAT[sid] >= 0:
next_in_chain = SAT[sid]
last_chunk_start, last_chunk_finish = chunks[-1]
if next_in_chain == last_chunk_finish + 1:
chunks[-1] = last_chunk_start, next_in_chain
else:
chunks.extend([(next_in_chain, next_in_chain)])
sid = next_in_chain
for s, f in chunks:
stream_data += data[s*sect_size:(f+1)*sect_size]
#print chunks
return stream_data
def print_bin_data(data):
i = 0
while i < len(data):
j = 0
while (i < len(data)) and (j < 16):
c = b'0x%02X' % ord(data[i])
sys.stdout.write(c)
sys.stdout.write(' ')
i += 1
j += 1
print()
if i == 0:
print('<NO DATA>')
# This implementation writes only 'Root Entry', 'Workbook' streams
# and 2 empty streams for aligning directory stream on sector boundary
#
# LAYOUT:
# 0 header
# 76 MSAT (1st part: 109 SID)
# 512 workbook stream
# ... additional MSAT sectors if streams' size > about 7 Mb == (109*512 * 128)
# ... SAT
# ... directory stream
#
# NOTE: this layout is "ad hoc". It can be more general. RTFM
class XlsDoc:
SECTOR_SIZE = 0x0200
MIN_LIMIT = 0x1000
SID_FREE_SECTOR = -1
SID_END_OF_CHAIN = -2
SID_USED_BY_SAT = -3
SID_USED_BY_MSAT = -4
def __init__(self):
self.book_stream_sect = []
self.dir_stream = b''
self.dir_stream_sect = []
self.packed_SAT = b''
self.SAT_sect = []
self.packed_MSAT_1st = b''
self.packed_MSAT_2nd = b''
self.MSAT_sect_2nd = []
self.header = b''
def __build_directory(self): # align on sector boundary
self.dir_stream = b''
#(to_py3): replaced = b'\x00'.join(b'Root Entry\x00') + b'\x00'
dentry_name = 'Root Entry\x00'.encode(ENCODING)
dentry_name_sz = len(dentry_name)
dentry_name_pad = b'\x00'*(64 - dentry_name_sz)
dentry_type = 0x05 # root storage
dentry_colour = 0x01 # black
dentry_did_left = -1
dentry_did_right = -1
dentry_did_root = 1
dentry_start_sid = -2
dentry_stream_sz = 0
self.dir_stream += struct.pack('<64s H 2B 3l 9L l L L',
dentry_name + dentry_name_pad,
dentry_name_sz,
dentry_type,
dentry_colour,
dentry_did_left,
dentry_did_right,
dentry_did_root,
0, 0, 0, 0, 0, 0, 0, 0, 0,
dentry_start_sid,
dentry_stream_sz,
0
)
# (to_py3): replaced = b'\x00'.join(b'Workbook\x00') + b'\x00'
dentry_name = 'Workbook\x00'.encode(ENCODING)
dentry_name_sz = len(dentry_name)
dentry_name_pad = b'\x00'*(64 - dentry_name_sz)
dentry_type = 0x02 # user stream
dentry_colour = 0x01 # black
dentry_did_left = -1
dentry_did_right = -1
dentry_did_root = -1
dentry_start_sid = 0
dentry_stream_sz = self.book_stream_len
self.dir_stream += struct.pack('<64s H 2B 3l 9L l L L',
dentry_name + dentry_name_pad,
dentry_name_sz,
dentry_type,
dentry_colour,
dentry_did_left,
dentry_did_right,
dentry_did_root,
0, 0, 0, 0, 0, 0, 0, 0, 0,
dentry_start_sid,
dentry_stream_sz,
0
)
# padding
dentry_name = b''
dentry_name_sz = len(dentry_name)
dentry_name_pad = b'\x00'*(64 - dentry_name_sz)
dentry_type = 0x00 # empty
dentry_colour = 0x01 # black
dentry_did_left = -1
dentry_did_right = -1
dentry_did_root = -1
dentry_start_sid = -2
dentry_stream_sz = 0
self.dir_stream += struct.pack('<64s H 2B 3l 9L l L L',
dentry_name + dentry_name_pad,
dentry_name_sz,
dentry_type,
dentry_colour,
dentry_did_left,
dentry_did_right,
dentry_did_root,
0, 0, 0, 0, 0, 0, 0, 0, 0,
dentry_start_sid,
dentry_stream_sz,
0
) * 2
def __build_sat(self):
# Build SAT
book_sect_count = self.book_stream_len >> 9
dir_sect_count = len(self.dir_stream) >> 9
total_sect_count = book_sect_count + dir_sect_count
SAT_sect_count = 0
MSAT_sect_count = 0
SAT_sect_count_limit = 109
while total_sect_count > 128*SAT_sect_count or SAT_sect_count > SAT_sect_count_limit:
SAT_sect_count += 1
total_sect_count += 1
if SAT_sect_count > SAT_sect_count_limit:
MSAT_sect_count += 1
total_sect_count += 1
SAT_sect_count_limit += 127
SAT = [self.SID_FREE_SECTOR]*128*SAT_sect_count
sect = 0
while sect < book_sect_count - 1:
self.book_stream_sect.append(sect)
SAT[sect] = sect + 1
sect += 1
self.book_stream_sect.append(sect)
SAT[sect] = self.SID_END_OF_CHAIN
sect += 1
while sect < book_sect_count + MSAT_sect_count:
self.MSAT_sect_2nd.append(sect)
SAT[sect] = self.SID_USED_BY_MSAT
sect += 1
while sect < book_sect_count + MSAT_sect_count + SAT_sect_count:
self.SAT_sect.append(sect)
SAT[sect] = self.SID_USED_BY_SAT
sect += 1
while sect < book_sect_count + MSAT_sect_count + SAT_sect_count + dir_sect_count - 1:
self.dir_stream_sect.append(sect)
SAT[sect] = sect + 1
sect += 1
self.dir_stream_sect.append(sect)
SAT[sect] = self.SID_END_OF_CHAIN
sect += 1
self.packed_SAT = struct.pack('<%dl' % (SAT_sect_count*128), *SAT)
MSAT_1st = [self.SID_FREE_SECTOR]*109
for i, SAT_sect_num in zip(list(range(0, 109)), self.SAT_sect):
MSAT_1st[i] = SAT_sect_num
self.packed_MSAT_1st = struct.pack('<109l', *MSAT_1st)
MSAT_2nd = [self.SID_FREE_SECTOR]*128*MSAT_sect_count
if MSAT_sect_count > 0:
MSAT_2nd[- 1] = self.SID_END_OF_CHAIN
i = 109
msat_sect = 0
sid_num = 0
while i < SAT_sect_count:
if (sid_num + 1) % 128 == 0:
#print 'link: ',
msat_sect += 1
if msat_sect < len(self.MSAT_sect_2nd):
MSAT_2nd[sid_num] = self.MSAT_sect_2nd[msat_sect]
else:
#print 'sid: ',
MSAT_2nd[sid_num] = self.SAT_sect[i]
i += 1
#print sid_num, MSAT_2nd[sid_num]
sid_num += 1
self.packed_MSAT_2nd = struct.pack('<%dl' % (MSAT_sect_count*128), *MSAT_2nd)
#print vars()
#print zip(range(0, sect), SAT)
#print self.book_stream_sect
#print self.MSAT_sect_2nd
#print MSAT_2nd
#print self.SAT_sect
#print self.dir_stream_sect
def __build_header(self):
doc_magic = b'\xD0\xCF\x11\xE0\xA1\xB1\x1A\xE1'
file_uid = b'\x00'*16
rev_num = b'\x3E\x00'
ver_num = b'\x03\x00'
byte_order = b'\xFE\xFF'
log_sect_size = struct.pack('<H', 9)
log_short_sect_size = struct.pack('<H', 6)
not_used0 = b'\x00'*10
total_sat_sectors = struct.pack('<L', len(self.SAT_sect))
dir_start_sid = struct.pack('<l', self.dir_stream_sect[0])
not_used1 = b'\x00'*4
min_stream_size = struct.pack('<L', 0x1000)
ssat_start_sid = struct.pack('<l', -2)
total_ssat_sectors = struct.pack('<L', 0)
if len(self.MSAT_sect_2nd) == 0:
msat_start_sid = struct.pack('<l', -2)
else:
msat_start_sid = struct.pack('<l', self.MSAT_sect_2nd[0])
total_msat_sectors = struct.pack('<L', len(self.MSAT_sect_2nd))
self.header = b''.join([ doc_magic,
file_uid,
rev_num,
ver_num,
byte_order,
log_sect_size,
log_short_sect_size,
not_used0,
total_sat_sectors,
dir_start_sid,
not_used1,
min_stream_size,
ssat_start_sid,
total_ssat_sectors,
msat_start_sid,
total_msat_sectors
])
def save(self, file_name_or_filelike_obj, stream):
# 1. Align stream on 0x1000 boundary (and therefore on sector boundary)
padding = b'\x00' * (0x1000 - (len(stream) % 0x1000))
self.book_stream_len = len(stream) + len(padding)
self.__build_directory()
self.__build_sat()
self.__build_header()
f = file_name_or_filelike_obj
we_own_it = not hasattr(f, 'write')
if we_own_it:
f = open(file_name_or_filelike_obj, 'wb')
f.write(self.header)
f.write(self.packed_MSAT_1st)
f.write(stream)
f.write(padding)
f.write(self.packed_MSAT_2nd)
f.write(self.packed_SAT)
f.write(self.dir_stream)
if we_own_it:
f.close()
+41
View File
@@ -0,0 +1,41 @@
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
+126
View File
@@ -0,0 +1,126 @@
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)
+659
View File
@@ -0,0 +1,659 @@
### $ANTLR 2.7.7 (20060930): "xlwt/excel-formula.g" -> "ExcelFormulaParser.py"$
### import antlr and other modules ..
import sys
from . import antlr
### header action >>>
import struct
from . 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 action <<<
### preamble action>>>
### preamble action <<<
### import antlr.Token
from .antlr import Token
### >>>The Known Token Types <<<
SKIP = antlr.SKIP
INVALID_TYPE = antlr.INVALID_TYPE
EOF_TYPE = antlr.EOF_TYPE
EOF = antlr.EOF
NULL_TREE_LOOKAHEAD = antlr.NULL_TREE_LOOKAHEAD
MIN_USER_TYPE = antlr.MIN_USER_TYPE
TRUE_CONST = 4
FALSE_CONST = 5
STR_CONST = 6
NUM_CONST = 7
INT_CONST = 8
FUNC_IF = 9
FUNC_CHOOSE = 10
NAME = 11
QUOTENAME = 12
EQ = 13
NE = 14
GT = 15
LT = 16
GE = 17
LE = 18
ADD = 19
SUB = 20
MUL = 21
DIV = 22
POWER = 23
PERCENT = 24
LP = 25
RP = 26
LB = 27
RB = 28
COLON = 29
COMMA = 30
SEMICOLON = 31
REF2D = 32
REF2D_R1C1 = 33
BANG = 34
CONCAT = 35
class Parser(antlr.LLkParser):
### user action >>>
### user action <<<
def __init__(self, *args, **kwargs):
antlr.LLkParser.__init__(self, *args, **kwargs)
self.tokenNames = _tokenNames
### __init__ header action >>>
self.rpn = b""
self.sheet_references = []
self.xcall_references = []
### __init__ header action <<<
def formula(self):
self.expr("V")
def expr(self, arg_type):
self.prec0_expr(arg_type)
while True:
if ((self.LA(1) >= EQ and self.LA(1) <= LE)):
pass
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [EQ]:
pass
self.match(EQ)
op = struct.pack('B', ptgEQ)
elif la1 and la1 in [NE]:
pass
self.match(NE)
op = struct.pack('B', ptgNE)
elif la1 and la1 in [GT]:
pass
self.match(GT)
op = struct.pack('B', ptgGT)
elif la1 and la1 in [LT]:
pass
self.match(LT)
op = struct.pack('B', ptgLT)
elif la1 and la1 in [GE]:
pass
self.match(GE)
op = struct.pack('B', ptgGE)
elif la1 and la1 in [LE]:
pass
self.match(LE)
op = struct.pack('B', ptgLE)
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
self.prec0_expr(arg_type)
self.rpn += op
else:
break
def prec0_expr(self,
arg_type
):
pass
self.prec1_expr(arg_type)
while True:
if (self.LA(1)==CONCAT):
pass
pass
self.match(CONCAT)
op = struct.pack('B', ptgConcat)
self.prec1_expr(arg_type)
self.rpn += op
else:
break
def prec1_expr(self,
arg_type
):
pass
self.prec2_expr(arg_type)
while True:
if (self.LA(1)==ADD or self.LA(1)==SUB):
pass
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [ADD]:
pass
self.match(ADD)
op = struct.pack('B', ptgAdd)
elif la1 and la1 in [SUB]:
pass
self.match(SUB)
op = struct.pack('B', ptgSub)
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
self.prec2_expr(arg_type)
self.rpn += op;
# print "**prec1_expr4 %s" % arg_type
else:
break
def prec2_expr(self,
arg_type
):
pass
self.prec3_expr(arg_type)
while True:
if (self.LA(1)==MUL or self.LA(1)==DIV):
pass
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [MUL]:
pass
self.match(MUL)
op = struct.pack('B', ptgMul)
elif la1 and la1 in [DIV]:
pass
self.match(DIV)
op = struct.pack('B', ptgDiv)
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
self.prec3_expr(arg_type)
self.rpn += op
else:
break
def prec3_expr(self,
arg_type
):
pass
self.prec4_expr(arg_type)
while True:
if (self.LA(1)==POWER):
pass
pass
self.match(POWER)
op = struct.pack('B', ptgPower)
self.prec4_expr(arg_type)
self.rpn += op
else:
break
def prec4_expr(self,
arg_type
):
pass
self.prec5_expr(arg_type)
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [PERCENT]:
pass
self.match(PERCENT)
self.rpn += struct.pack('B', ptgPercent)
elif la1 and la1 in [EOF,EQ,NE,GT,LT,GE,LE,ADD,SUB,MUL,DIV,POWER,RP,COMMA,SEMICOLON,CONCAT]:
pass
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
def prec5_expr(self,
arg_type
):
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [TRUE_CONST,FALSE_CONST,STR_CONST,NUM_CONST,INT_CONST,FUNC_IF,FUNC_CHOOSE,NAME,QUOTENAME,LP,REF2D]:
pass
self.primary(arg_type)
elif la1 and la1 in [SUB]:
pass
self.match(SUB)
self.primary(arg_type)
self.rpn += struct.pack('B', ptgUminus)
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
def primary(self,
arg_type
):
str_tok = None
int_tok = None
num_tok = None
ref2d_tok = None
ref2d1_tok = None
ref2d2_tok = None
ref3d_ref2d = None
ref3d_ref2d2 = None
name_tok = None
func_tok = None
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [TRUE_CONST]:
pass
self.match(TRUE_CONST)
self.rpn += struct.pack("2B", ptgBool, 1)
elif la1 and la1 in [FALSE_CONST]:
pass
self.match(FALSE_CONST)
self.rpn += struct.pack("2B", ptgBool, 0)
elif la1 and la1 in [STR_CONST]:
pass
str_tok = self.LT(1)
self.match(STR_CONST)
self.rpn += struct.pack("B", ptgStr) + upack1(str_tok.text[1:-1].replace("\"\"", "\""))
elif la1 and la1 in [NUM_CONST]:
pass
num_tok = self.LT(1)
self.match(NUM_CONST)
self.rpn += struct.pack("<Bd", ptgNum, float(num_tok.text))
elif la1 and la1 in [FUNC_IF]:
pass
self.match(FUNC_IF)
self.match(LP)
self.expr("V")
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [SEMICOLON]:
pass
self.match(SEMICOLON)
elif la1 and la1 in [COMMA]:
pass
self.match(COMMA)
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
self.rpn += struct.pack("<BBH", ptgAttr, 0x02, 0) # tAttrIf
pos0 = len(self.rpn) - 2
self.expr(arg_type)
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [SEMICOLON]:
pass
self.match(SEMICOLON)
elif la1 and la1 in [COMMA]:
pass
self.match(COMMA)
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
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:]
self.expr(arg_type)
self.match(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:]
elif la1 and la1 in [FUNC_CHOOSE]:
pass
self.match(FUNC_CHOOSE)
arg_type = b"R"
rpn_chunks = []
self.match(LP)
self.expr("V")
rpn_start = len(self.rpn)
ref_markers = [len(self.sheet_references)]
while True:
if (self.LA(1)==COMMA or self.LA(1)==SEMICOLON):
pass
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [SEMICOLON]:
pass
self.match(SEMICOLON)
elif la1 and la1 in [COMMA]:
pass
self.match(COMMA)
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
mark = len(self.rpn)
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [TRUE_CONST,FALSE_CONST,STR_CONST,NUM_CONST,INT_CONST,FUNC_IF,FUNC_CHOOSE,NAME,QUOTENAME,SUB,LP,REF2D]:
pass
self.expr(arg_type)
elif la1 and la1 in [RP,COMMA,SEMICOLON]:
pass
self.rpn += struct.pack("B", ptgMissArg)
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
rpn_chunks.append(self.rpn[mark:])
ref_markers.append(len(self.sheet_references))
else:
break
self.match(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 range(nc-1, 0, -1):
skiplens[ic-1] = skiplens[ic] + chunklens[ic] + 4
jump_pos = [2 * nc + 2]
for ic in range(nc):
jump_pos.append(jump_pos[-1] + chunklens[ic] + 4)
chunk_shift = 2 * nc + 6 # size of tAttrChoose
for ic in range(nc):
for refx in range(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 range(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 += b"".join(choose_rpn)
elif la1 and la1 in [LP]:
pass
self.match(LP)
self.expr(arg_type)
self.match(RP)
self.rpn += struct.pack("B", ptgParen)
else:
if (self.LA(1)==INT_CONST) and (_tokenSet_0.member(self.LA(2))):
pass
int_tok = self.LT(1)
self.match(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))
elif (self.LA(1)==REF2D) and (_tokenSet_0.member(self.LA(2))):
pass
ref2d_tok = self.LT(1)
self.match(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)
elif (self.LA(1)==REF2D) and (self.LA(2)==COLON):
pass
ref2d1_tok = self.LT(1)
self.match(REF2D)
self.match(COLON)
ref2d2_tok = self.LT(1)
self.match(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)
elif (self.LA(1)==INT_CONST or self.LA(1)==NAME or self.LA(1)==QUOTENAME) and (self.LA(2)==COLON or self.LA(2)==BANG):
pass
sheet1=self.sheet()
sheet2 = sheet1
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [COLON]:
pass
self.match(COLON)
sheet2=self.sheet()
elif la1 and la1 in [BANG]:
pass
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
self.match(BANG)
ref3d_ref2d = self.LT(1)
self.match(REF2D)
ptg = ptgRef3dR + _RVAdeltaRef[arg_type]
rpn_ref2d = b""
r1, c1 = Utils.cell_to_packed_rowcol(ref3d_ref2d.text)
rpn_ref2d = struct.pack("<3H", 0x0000, r1, c1)
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [COLON]:
pass
self.match(COLON)
ref3d_ref2d2 = self.LT(1)
self.match(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)
elif la1 and la1 in [EOF,EQ,NE,GT,LT,GE,LE,ADD,SUB,MUL,DIV,POWER,PERCENT,RP,COMMA,SEMICOLON,CONCAT]:
pass
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
self.rpn += struct.pack("<B", ptg)
self.sheet_references.append((sheet1, sheet2, len(self.rpn)))
self.rpn += rpn_ref2d
elif (self.LA(1)==NAME) and (_tokenSet_0.member(self.LA(2))):
name_tok = self.LT(1)
self.match(NAME)
raise Exception("[formula] found unexpected NAME token (%r)" % name_tok.txt)
# #### TODO: handle references to defined names here
elif (self.LA(1)==NAME) and (self.LA(2)==LP):
func_tok = self.LT(1)
self.match(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
self.match(LP)
arg_count=self.expr_list(arg_type_list, min_argc, max_argc)
self.match(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)
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
def sheet(self):
ref = None
sheet_ref_name = None
sheet_ref_int = None
sheet_ref_quote = None
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [NAME]:
sheet_ref_name = self.LT(1)
self.match(NAME)
ref = sheet_ref_name.text
elif la1 and la1 in [INT_CONST]:
sheet_ref_int = self.LT(1)
self.match(INT_CONST)
ref = sheet_ref_int.text
elif la1 and la1 in [QUOTENAME]:
sheet_ref_quote = self.LT(1)
self.match(QUOTENAME)
ref = sheet_ref_quote.text[1:-1].replace("''", "'")
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
return ref
def expr_list(self,
arg_type_list, min_argc, max_argc
):
arg_cnt = None
arg_cnt = 0
arg_type = arg_type_list[arg_cnt]
# print "**expr_list1[%d] req=%s" % (arg_cnt, arg_type)
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [TRUE_CONST,FALSE_CONST,STR_CONST,NUM_CONST,INT_CONST,FUNC_IF,FUNC_CHOOSE,NAME,QUOTENAME,SUB,LP,REF2D]:
pass
self.expr(arg_type)
arg_cnt += 1
while True:
if (self.LA(1)==COMMA or self.LA(1)==SEMICOLON):
pass
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)
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [SEMICOLON]:
pass
self.match(SEMICOLON)
elif la1 and la1 in [COMMA]:
pass
self.match(COMMA)
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
la1 = self.LA(1)
if False:
pass
elif la1 and la1 in [TRUE_CONST,FALSE_CONST,STR_CONST,NUM_CONST,INT_CONST,FUNC_IF,FUNC_CHOOSE,NAME,QUOTENAME,SUB,LP,REF2D]:
pass
self.expr(arg_type)
elif la1 and la1 in [RP,COMMA,SEMICOLON]:
pass
self.rpn += struct.pack("B", ptgMissArg)
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
arg_cnt += 1
else:
break
elif la1 and la1 in [RP]:
pass
else:
raise antlr.NoViableAltException(self.LT(1), self.getFilename())
return arg_cnt
_tokenNames = [
"<0>",
"EOF",
"<2>",
"NULL_TREE_LOOKAHEAD",
"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",
"CONCAT"
]
### generate bit set
def mk_tokenSet_0():
### var1
data = [ 37681618946, 0]
return data
_tokenSet_0 = antlr.BitSet(mk_tokenSet_0())
+859
View File
@@ -0,0 +1,859 @@
""" lots of Excel Magic Numbers """
# Boundaries BIFF8+
MAX_ROW = 65536
MAX_COL = 256
biff_records = {
0x0000: "DIMENSIONS",
0x0001: "BLANK",
0x0002: "INTEGER",
0x0003: "NUMBER",
0x0004: "LABEL",
0x0005: "BOOLERR",
0x0006: "FORMULA",
0x0007: "STRING",
0x0008: "ROW",
0x0009: "BOF",
0x000A: "EOF",
0x000B: "INDEX",
0x000C: "CALCCOUNT",
0x000D: "CALCMODE",
0x000E: "PRECISION",
0x000F: "REFMODE",
0x0010: "DELTA",
0x0011: "ITERATION",
0x0012: "PROTECT",
0x0013: "PASSWORD",
0x0014: "HEADER",
0x0015: "FOOTER",
0x0016: "EXTERNCOUNT",
0x0017: "EXTERNSHEET",
0x0018: "NAME",
0x0019: "WINDOWPROTECT",
0x001A: "VERTICALPAGEBREAKS",
0x001B: "HORIZONTALPAGEBREAKS",
0x001C: "NOTE",
0x001D: "SELECTION",
0x001E: "FORMAT",
0x001F: "FORMATCOUNT",
0x0020: "COLUMNDEFAULT",
0x0021: "ARRAY",
0x0022: "1904",
0x0023: "EXTERNNAME",
0x0024: "COLWIDTH",
0x0025: "DEFAULTROWHEIGHT",
0x0026: "LEFTMARGIN",
0x0027: "RIGHTMARGIN",
0x0028: "TOPMARGIN",
0x0029: "BOTTOMMARGIN",
0x002A: "PRINTHEADERS",
0x002B: "PRINTGRIDLINES",
0x002F: "FILEPASS",
0x0031: "FONT",
0x0036: "TABLE",
0x003C: "CONTINUE",
0x003D: "WINDOW1",
0x003E: "WINDOW2",
0x0040: "BACKUP",
0x0041: "PANE",
0x0042: "CODEPAGE",
0x0043: "XF",
0x0044: "IXFE",
0x0045: "EFONT",
0x004D: "PLS",
0x0050: "DCON",
0x0051: "DCONREF",
0x0053: "DCONNAME",
0x0055: "DEFCOLWIDTH",
0x0056: "BUILTINFMTCNT",
0x0059: "XCT",
0x005A: "CRN",
0x005B: "FILESHARING",
0x005C: "WRITEACCESS",
0x005D: "OBJ",
0x005E: "UNCALCED",
0x005F: "SAFERECALC",
0x0060: "TEMPLATE",
0x0063: "OBJPROTECT",
0x007D: "COLINFO",
0x007E: "RK",
0x007F: "IMDATA",
0x0080: "GUTS",
0x0081: "WSBOOL",
0x0082: "GRIDSET",
0x0083: "HCENTER",
0x0084: "VCENTER",
0x0085: "BOUNDSHEET",
0x0086: "WRITEPROT",
0x0087: "ADDIN",
0x0088: "EDG",
0x0089: "PUB",
0x008C: "COUNTRY",
0x008D: "HIDEOBJ",
0x008E: "BUNDLESOFFSET",
0x008F: "BUNDLEHEADER",
0x0090: "SORT",
0x0091: "SUB",
0x0092: "PALETTE",
0x0093: "STYLE",
0x0094: "LHRECORD",
0x0095: "LHNGRAPH",
0x0096: "SOUND",
0x0098: "LPR",
0x0099: "STANDARDWIDTH",
0x009A: "FNGROUPNAME",
0x009B: "FILTERMODE",
0x009C: "FNGROUPCOUNT",
0x009D: "AUTOFILTERINFO",
0x009E: "AUTOFILTER",
0x00A0: "SCL",
0x00A1: "SETUP",
0x00A9: "COORDLIST",
0x00AB: "GCW",
0x00AE: "SCENMAN",
0x00AF: "SCENARIO",
0x00B0: "SXVIEW",
0x00B1: "SXVD",
0x00B2: "SXVI",
0x00B4: "SXIVD",
0x00B5: "SXLI",
0x00B6: "SXPI",
0x00B8: "DOCROUTE",
0x00B9: "RECIPNAME",
0x00BC: "SHRFMLA",
0x00BD: "MULRK",
0x00BE: "MULBLANK",
0x00C1: "MMS",
0x00C2: "ADDMENU",
0x00C3: "DELMENU",
0x00C5: "SXDI",
0x00C6: "SXDB",
0x00C7: "SXFIELD",
0x00C8: "SXINDEXLIST",
0x00C9: "SXDOUBLE",
0x00CD: "SXSTRING",
0x00CE: "SXDATETIME",
0x00D0: "SXTBL",
0x00D1: "SXTBRGITEM",
0x00D2: "SXTBPG",
0x00D3: "OBPROJ",
0x00D5: "SXIDSTM",
0x00D6: "RSTRING",
0x00D7: "DBCELL",
0x00DA: "BOOKBOOL",
0x00DC: "SXEXT|PARAMQRY",
0x00DD: "SCENPROTECT",
0x00DE: "OLESIZE",
0x00DF: "UDDESC",
0x00E0: "XF",
0x00E1: "INTERFACEHDR",
0x00E2: "INTERFACEEND",
0x00E3: "SXVS",
0x00E5: "MERGEDCELLS",
0x00E9: "BITMAP",
0x00EB: "MSODRAWINGGROUP",
0x00EC: "MSODRAWING",
0x00ED: "MSODRAWINGSELECTION",
0x00F0: "SXRULE",
0x00F1: "SXEX",
0x00F2: "SXFILT",
0x00F6: "SXNAME",
0x00F7: "SXSELECT",
0x00F8: "SXPAIR",
0x00F9: "SXFMLA",
0x00FB: "SXFORMAT",
0x00FC: "SST",
0x00FD: "LABELSST",
0x00FF: "EXTSST",
0x0100: "SXVDEX",
0x0103: "SXFORMULA",
0x0122: "SXDBEX",
0x0137: "CHTRINSERT",
0x0138: "CHTRINFO",
0x013B: "CHTRCELLCONTENT",
0x013D: "TABID",
0x0140: "CHTRMOVERANGE",
0x014D: "CHTRINSERTTAB",
0x015F: "LABELRANGES",
0x0160: "USESELFS",
0x0161: "DSF",
0x0162: "XL5MODIFY",
0x0196: "CHTRHEADER",
0x01A9: "USERBVIEW",
0x01AA: "USERSVIEWBEGIN",
0x01AB: "USERSVIEWEND",
0x01AD: "QSI",
0x01AE: "SUPBOOK",
0x01AF: "PROT4REV",
0x01B0: "CONDFMT",
0x01B1: "CF",
0x01B2: "DVAL",
0x01B5: "DCONBIN",
0x01B6: "TXO",
0x01B7: "REFRESHALL",
0x01B8: "HLINK",
0x01BA: "CODENAME",
0x01BB: "SXFDBTYPE",
0x01BC: "PROT4REVPASS",
0x01BE: "DV",
0x01C0: "XL9FILE",
0x01C1: "RECALCID",
0x0200: "DIMENSIONS",
0x0201: "BLANK",
0x0203: "NUMBER",
0x0204: "LABEL",
0x0205: "BOOLERR",
0x0206: "FORMULA",
0x0207: "STRING",
0x0208: "ROW",
0x0209: "BOF",
0x020B: "INDEX",
0x0218: "NAME",
0x0221: "ARRAY",
0x0223: "EXTERNNAME",
0x0225: "DEFAULTROWHEIGHT",
0x0231: "FONT",
0x0236: "TABLE",
0x023E: "WINDOW2",
0x0243: "XF",
0x027E: "RK",
0x0293: "STYLE",
0x0406: "FORMULA",
0x0409: "BOF",
0x041E: "FORMAT",
0x0443: "XF",
0x04BC: "SHRFMLA",
0x0800: "SCREENTIP",
0x0803: "WEBQRYSETTINGS",
0x0804: "WEBQRYTABLES",
0x0809: "BOF",
0x0862: "SHEETLAYOUT",
0x0867: "SHEETPROTECTION",
0x1001: "UNITS",
0x1002: "ChartChart",
0x1003: "ChartSeries",
0x1006: "ChartDataformat",
0x1007: "ChartLineformat",
0x1009: "ChartMarkerformat",
0x100A: "ChartAreaformat",
0x100B: "ChartPieformat",
0x100C: "ChartAttachedlabel",
0x100D: "ChartSeriestext",
0x1014: "ChartChartformat",
0x1015: "ChartLegend",
0x1016: "ChartSerieslist",
0x1017: "ChartBar",
0x1018: "ChartLine",
0x1019: "ChartPie",
0x101A: "ChartArea",
0x101B: "ChartScatter",
0x101C: "ChartChartline",
0x101D: "ChartAxis",
0x101E: "ChartTick",
0x101F: "ChartValuerange",
0x1020: "ChartCatserrange",
0x1021: "ChartAxislineformat",
0x1022: "ChartFormatlink",
0x1024: "ChartDefaulttext",
0x1025: "ChartText",
0x1026: "ChartFontx",
0x1027: "ChartObjectLink",
0x1032: "ChartFrame",
0x1033: "BEGIN",
0x1034: "END",
0x1035: "ChartPlotarea",
0x103A: "Chart3D",
0x103C: "ChartPicf",
0x103D: "ChartDropbar",
0x103E: "ChartRadar",
0x103F: "ChartSurface",
0x1040: "ChartRadararea",
0x1041: "ChartAxisparent",
0x1043: "ChartLegendxn",
0x1044: "ChartShtprops",
0x1045: "ChartSertocrt",
0x1046: "ChartAxesused",
0x1048: "ChartSbaseref",
0x104A: "ChartSerparent",
0x104B: "ChartSerauxtrend",
0x104E: "ChartIfmt",
0x104F: "ChartPos",
0x1050: "ChartAlruns",
0x1051: "ChartAI",
0x105B: "ChartSerauxerrbar",
0x105D: "ChartSerfmt",
0x105F: "Chart3DDataFormat",
0x1060: "ChartFbi",
0x1061: "ChartBoppop",
0x1062: "ChartAxcext",
0x1063: "ChartDat",
0x1064: "ChartPlotgrowth",
0x1065: "ChartSiindex",
0x1066: "ChartGelframe",
0x1067: "ChartBoppcustom",
0xFFFF: ""
}
all_funcs_by_name = {
# Includes Analysis ToolPak aka ATP aka add-in aka xcall functions,
# distinguished by -ve opcode.
# name: (opcode, min # args, max # args, func return type, func arg types)
# + in func arg types means more of the same.
'ABS' : ( 24, 1, 1, 'V', 'V'),
'ACCRINT' : ( -1, 6, 7, 'V', 'VVVVVVV'),
'ACCRINTM' : ( -1, 3, 5, 'V', 'VVVVV'),
'ACOS' : ( 99, 1, 1, 'V', 'V'),
'ACOSH' : (233, 1, 1, 'V', 'V'),
'ADDRESS' : (219, 2, 5, 'V', 'VVVVV'),
'AMORDEGRC' : ( -1, 7, 7, 'V', 'VVVVVVV'),
'AMORLINC' : ( -1, 7, 7, 'V', 'VVVVVVV'),
'AND' : ( 36, 1, 30, 'V', 'D+'),
'AREAS' : ( 75, 1, 1, 'V', 'R'),
'ASC' : (214, 1, 1, 'V', 'V'),
'ASIN' : ( 98, 1, 1, 'V', 'V'),
'ASINH' : (232, 1, 1, 'V', 'V'),
'ATAN' : ( 18, 1, 1, 'V', 'V'),
'ATAN2' : ( 97, 2, 2, 'V', 'VV'),
'ATANH' : (234, 1, 1, 'V', 'V'),
'AVEDEV' : (269, 1, 30, 'V', 'D+'),
'AVERAGE' : ( 5, 1, 30, 'V', 'D+'),
'AVERAGEA' : (361, 1, 30, 'V', 'D+'),
'BAHTTEXT' : (368, 1, 1, 'V', 'V'),
'BESSELI' : ( -1, 2, 2, 'V', 'VV'),
'BESSELJ' : ( -1, 2, 2, 'V', 'VV'),
'BESSELK' : ( -1, 2, 2, 'V', 'VV'),
'BESSELY' : ( -1, 2, 2, 'V', 'VV'),
'BETADIST' : (270, 3, 5, 'V', 'VVVVV'),
'BETAINV' : (272, 3, 5, 'V', 'VVVVV'),
'BIN2DEC' : ( -1, 1, 1, 'V', 'V'),
'BIN2HEX' : ( -1, 1, 2, 'V', 'VV'),
'BIN2OCT' : ( -1, 1, 2, 'V', 'VV'),
'BINOMDIST' : (273, 4, 4, 'V', 'VVVV'),
'CEILING' : (288, 2, 2, 'V', 'VV'),
'CELL' : (125, 1, 2, 'V', 'VR'),
'CHAR' : (111, 1, 1, 'V', 'V'),
'CHIDIST' : (274, 2, 2, 'V', 'VV'),
'CHIINV' : (275, 2, 2, 'V', 'VV'),
'CHITEST' : (306, 2, 2, 'V', 'AA'),
'CHOOSE' : (100, 2, 30, 'R', 'VR+'),
'CLEAN' : (162, 1, 1, 'V', 'V'),
'CODE' : (121, 1, 1, 'V', 'V'),
'COLUMN' : ( 9, 0, 1, 'V', 'R'),
'COLUMNS' : ( 77, 1, 1, 'V', 'R'),
'COMBIN' : (276, 2, 2, 'V', 'VV'),
'COMPLEX' : ( -1, 2, 3, 'V', 'VVV'),
'CONCATENATE' : (336, 1, 30, 'V', 'V+'),
'CONFIDENCE' : (277, 3, 3, 'V', 'VVV'),
'CONVERT' : ( -1, 3, 3, 'V', 'VVV'),
'CORREL' : (307, 2, 2, 'V', 'AA'),
'COS' : ( 16, 1, 1, 'V', 'V'),
'COSH' : (230, 1, 1, 'V', 'V'),
'COUNT' : ( 0, 1, 30, 'V', 'D+'),
'COUNTA' : (169, 1, 30, 'V', 'D+'),
'COUNTBLANK' : (347, 1, 1, 'V', 'R'),
'COUNTIF' : (346, 2, 2, 'V', 'RV'),
'COUPDAYBS' : ( -1, 3, 5, 'V', 'VVVVV'),
'COUPDAYS' : ( -1, 3, 5, 'V', 'VVVVV'),
'COUPDAYSNC' : ( -1, 3, 5, 'V', 'VVVVV'),
'COUPNCD' : ( -1, 3, 5, 'V', 'VVVVV'),
'COUPNUM' : ( -1, 3, 5, 'V', 'VVVVV'),
'COUPPCD' : ( -1, 3, 5, 'V', 'VVVVV'),
'COVAR' : (308, 2, 2, 'V', 'AA'),
'CRITBINOM' : (278, 3, 3, 'V', 'VVV'),
'CUMIPMT' : ( -1, 6, 6, 'V', 'VVVVVV'),
'CUMPRINC' : ( -1, 6, 6, 'V', 'VVVVVV'),
'DATE' : ( 65, 3, 3, 'V', 'VVV'),
'DATEDIF' : (351, 3, 3, 'V', 'VVV'),
'DATEVALUE' : (140, 1, 1, 'V', 'V'),
'DAVERAGE' : ( 42, 3, 3, 'V', 'RRR'),
'DAY' : ( 67, 1, 1, 'V', 'V'),
'DAYS360' : (220, 2, 3, 'V', 'VVV'),
'DB' : (247, 4, 5, 'V', 'VVVVV'),
'DBCS' : (215, 1, 1, 'V', 'V'),
'DCOUNT' : ( 40, 3, 3, 'V', 'RRR'),
'DCOUNTA' : (199, 3, 3, 'V', 'RRR'),
'DDB' : (144, 4, 5, 'V', 'VVVVV'),
'DEC2BIN' : ( -1, 1, 2, 'V', 'VV'),
'DEC2HEX' : ( -1, 1, 2, 'V', 'VV'),
'DEC2OCT' : ( -1, 1, 2, 'V', 'VV'),
'DEGREES' : (343, 1, 1, 'V', 'V'),
'DELTA' : ( -1, 1, 2, 'V', 'VV'),
'DEVSQ' : (318, 1, 30, 'V', 'D+'),
'DGET' : (235, 3, 3, 'V', 'RRR'),
'DISC' : ( -1, 4, 5, 'V', 'VVVVV'),
'DMAX' : ( 44, 3, 3, 'V', 'RRR'),
'DMIN' : ( 43, 3, 3, 'V', 'RRR'),
'DOLLAR' : ( 13, 1, 2, 'V', 'VV'),
'DOLLARDE' : ( -1, 2, 2, 'V', 'VV'),
'DOLLARFR' : ( -1, 2, 2, 'V', 'VV'),
'DPRODUCT' : (189, 3, 3, 'V', 'RRR'),
'DSTDEV' : ( 45, 3, 3, 'V', 'RRR'),
'DSTDEVP' : (195, 3, 3, 'V', 'RRR'),
'DSUM' : ( 41, 3, 3, 'V', 'RRR'),
'DURATION' : ( -1, 5, 6, 'V', 'VVVVVV'),
'DVAR' : ( 47, 3, 3, 'V', 'RRR'),
'DVARP' : (196, 3, 3, 'V', 'RRR'),
'EDATE' : ( -1, 2, 2, 'V', 'VV'),
'EFFECT' : ( -1, 2, 2, 'V', 'VV'),
'EOMONTH' : ( -1, 1, 2, 'V', 'VV'),
'ERF' : ( -1, 1, 2, 'V', 'VV'),
'ERFC' : ( -1, 1, 1, 'V', 'V'),
'ERROR.TYPE' : (261, 1, 1, 'V', 'V'),
'EVEN' : (279, 1, 1, 'V', 'V'),
'EXACT' : (117, 2, 2, 'V', 'VV'),
'EXP' : ( 21, 1, 1, 'V', 'V'),
'EXPONDIST' : (280, 3, 3, 'V', 'VVV'),
'FACT' : (184, 1, 1, 'V', 'V'),
'FACTDOUBLE' : ( -1, 1, 1, 'V', 'V'),
'FALSE' : ( 35, 0, 0, 'V', '-'),
'FDIST' : (281, 3, 3, 'V', 'VVV'),
'FIND' : (124, 2, 3, 'V', 'VVV'),
'FINDB' : (205, 2, 3, 'V', 'VVV'),
'FINV' : (282, 3, 3, 'V', 'VVV'),
'FISHER' : (283, 1, 1, 'V', 'V'),
'FISHERINV' : (284, 1, 1, 'V', 'V'),
'FIXED' : ( 14, 2, 3, 'V', 'VVV'),
'FLOOR' : (285, 2, 2, 'V', 'VV'),
'FORECAST' : (309, 3, 3, 'V', 'VAA'),
'FREQUENCY' : (252, 2, 2, 'A', 'RR'),
'FTEST' : (310, 2, 2, 'V', 'AA'),
'FV' : ( 57, 3, 5, 'V', 'VVVVV'),
'FVSCHEDULE' : ( -1, 2, 2, 'V', 'VA'),
'GAMMADIST' : (286, 4, 4, 'V', 'VVVV'),
'GAMMAINV' : (287, 3, 3, 'V', 'VVV'),
'GAMMALN' : (271, 1, 1, 'V', 'V'),
'GCD' : ( -1, 1, 29, 'V', 'V+'),
'GEOMEAN' : (319, 1, 30, 'V', 'D+'),
'GESTEP' : ( -1, 1, 2, 'V', 'VV'),
'GETPIVOTDATA': (358, 2, 30, 'A', 'VAV+'),
'GROWTH' : ( 52, 1, 4, 'A', 'RRRV'),
'HARMEAN' : (320, 1, 30, 'V', 'D+'),
'HEX2BIN' : ( -1, 1, 2, 'V', 'VV'),
'HEX2DEC' : ( -1, 1, 1, 'V', 'V'),
'HEX2OCT' : ( -1, 1, 2, 'V', 'VV'),
'HLOOKUP' : (101, 3, 4, 'V', 'VRRV'),
'HOUR' : ( 71, 1, 1, 'V', 'V'),
'HYPERLINK' : (359, 1, 2, 'V', 'VV'),
'HYPGEOMDIST' : (289, 4, 4, 'V', 'VVVV'),
'IF' : ( 1, 2, 3, 'R', 'VRR'),
'IMABS' : ( -1, 1, 1, 'V', 'V'),
'IMAGINARY' : ( -1, 1, 1, 'V', 'V'),
'IMARGUMENT' : ( -1, 1, 1, 'V', 'V'),
'IMCONJUGATE' : ( -1, 1, 1, 'V', 'V'),
'IMCOS' : ( -1, 1, 1, 'V', 'V'),
'IMDIV' : ( -1, 2, 2, 'V', 'VV'),
'IMEXP' : ( -1, 1, 1, 'V', 'V'),
'IMLN' : ( -1, 1, 1, 'V', 'V'),
'IMLOG10' : ( -1, 1, 1, 'V', 'V'),
'IMLOG2' : ( -1, 1, 1, 'V', 'V'),
'IMPOWER' : ( -1, 2, 2, 'V', 'VV'),
'IMPRODUCT' : ( -1, 2, 2, 'V', 'VV'),
'IMREAL' : ( -1, 1, 1, 'V', 'V'),
'IMSIN' : ( -1, 1, 1, 'V', 'V'),
'IMSQRT' : ( -1, 1, 1, 'V', 'V'),
'IMSUB' : ( -1, 2, 2, 'V', 'VV'),
'IMSUM' : ( -1, 1, 29, 'V', 'V+'),
'INDEX' : ( 29, 2, 4, 'R', 'RVVV'),
'INDIRECT' : (148, 1, 2, 'R', 'VV'),
'INFO' : (244, 1, 1, 'V', 'V'),
'INT' : ( 25, 1, 1, 'V', 'V'),
'INTERCEPT' : (311, 2, 2, 'V', 'AA'),
'INTRATE' : ( -1, 4, 5, 'V', 'VVVVV'),
'IPMT' : (167, 4, 6, 'V', 'VVVVVV'),
'IRR' : ( 62, 1, 2, 'V', 'RV'),
'ISBLANK' : (129, 1, 1, 'V', 'V'),
'ISERR' : (126, 1, 1, 'V', 'V'),
'ISERROR' : ( 3, 1, 1, 'V', 'V'),
'ISEVEN' : ( -1, 1, 1, 'V', 'V'),
'ISLOGICAL' : (198, 1, 1, 'V', 'V'),
'ISNA' : ( 2, 1, 1, 'V', 'V'),
'ISNONTEXT' : (190, 1, 1, 'V', 'V'),
'ISNUMBER' : (128, 1, 1, 'V', 'V'),
'ISODD' : ( -1, 1, 1, 'V', 'V'),
'ISPMT' : (350, 4, 4, 'V', 'VVVV'),
'ISREF' : (105, 1, 1, 'V', 'R'),
'ISTEXT' : (127, 1, 1, 'V', 'V'),
'KURT' : (322, 1, 30, 'V', 'D+'),
'LARGE' : (325, 2, 2, 'V', 'RV'),
'LCM' : ( -1, 1, 29, 'V', 'V+'),
'LEFT' : (115, 1, 2, 'V', 'VV'),
'LEFTB' : (208, 1, 2, 'V', 'VV'),
'LEN' : ( 32, 1, 1, 'V', 'V'),
'LENB' : (211, 1, 1, 'V', 'V'),
'LINEST' : ( 49, 1, 4, 'A', 'RRVV'),
'LN' : ( 22, 1, 1, 'V', 'V'),
'LOG' : (109, 1, 2, 'V', 'VV'),
'LOG10' : ( 23, 1, 1, 'V', 'V'),
'LOGEST' : ( 51, 1, 4, 'A', 'RRVV'),
'LOGINV' : (291, 3, 3, 'V', 'VVV'),
'LOGNORMDIST' : (290, 3, 3, 'V', 'VVV'),
'LOOKUP' : ( 28, 2, 3, 'V', 'VRR'),
'LOWER' : (112, 1, 1, 'V', 'V'),
'MATCH' : ( 64, 2, 3, 'V', 'VRR'),
'MAX' : ( 7, 1, 30, 'V', 'D+'),
'MAXA' : (362, 1, 30, 'V', 'D+'),
'MDETERM' : (163, 1, 1, 'V', 'A'),
'MDURATION' : ( -1, 5, 6, 'V', 'VVVVVV'),
'MEDIAN' : (227, 1, 30, 'V', 'D+'),
'MID' : ( 31, 3, 3, 'V', 'VVV'),
'MIDB' : (210, 3, 3, 'V', 'VVV'),
'MIN' : ( 6, 1, 30, 'V', 'D+'),
'MINA' : (363, 1, 30, 'V', 'D+'),
'MINUTE' : ( 72, 1, 1, 'V', 'V'),
'MINVERSE' : (164, 1, 1, 'A', 'A'),
'MIRR' : ( 61, 3, 3, 'V', 'RVV'),
'MMULT' : (165, 2, 2, 'A', 'AA'),
'MOD' : ( 39, 2, 2, 'V', 'VV'),
'MODE' : (330, 1, 30, 'V', 'A+'), ################ weird #################
'MONTH' : ( 68, 1, 1, 'V', 'V'),
'MROUND' : ( -1, 2, 2, 'V', 'VV'),
'MULTINOMIAL' : ( -1, 1, 29, 'V', 'V+'),
'N' : (131, 1, 1, 'V', 'R'),
'NA' : ( 10, 0, 0, 'V', '-'),
'NEGBINOMDIST': (292, 3, 3, 'V', 'VVV'),
'NETWORKDAYS' : ( -1, 2, 3, 'V', 'VVR'),
'NOMINAL' : ( -1, 2, 2, 'V', 'VV'),
'NORMDIST' : (293, 4, 4, 'V', 'VVVV'),
'NORMINV' : (295, 3, 3, 'V', 'VVV'),
'NORMSDIST' : (294, 1, 1, 'V', 'V'),
'NORMSINV' : (296, 1, 1, 'V', 'V'),
'NOT' : ( 38, 1, 1, 'V', 'V'),
'NOW' : ( 74, 0, 0, 'V', '-'),
'NPER' : ( 58, 3, 5, 'V', 'VVVVV'),
'NPV' : ( 11, 2, 30, 'V', 'VD+'),
'OCT2BIN' : ( -1, 1, 2, 'V', 'VV'),
'OCT2DEC' : ( -1, 1, 1, 'V', 'V'),
'OCT2HEX' : ( -1, 1, 2, 'V', 'VV'),
'ODD' : (298, 1, 1, 'V', 'V'),
'ODDFPRICE' : ( -1, 9, 9, 'V', 'VVVVVVVVV'),
'ODDFYIELD' : ( -1, 9, 9, 'V', 'VVVVVVVVV'),
'ODDLPRICE' : ( -1, 8, 8, 'V', 'VVVVVVVV'),
'ODDLYIELD' : ( -1, 8, 8, 'V', 'VVVVVVVV'),
'OFFSET' : ( 78, 3, 5, 'R', 'RVVVV'),
'OR' : ( 37, 1, 30, 'V', 'D+'),
'PEARSON' : (312, 2, 2, 'V', 'AA'),
'PERCENTILE' : (328, 2, 2, 'V', 'RV'),
'PERCENTRANK' : (329, 2, 3, 'V', 'RVV'),
'PERMUT' : (299, 2, 2, 'V', 'VV'),
'PHONETIC' : (360, 1, 1, 'V', 'R'),
'PI' : ( 19, 0, 0, 'V', '-'),
'PMT' : ( 59, 3, 5, 'V', 'VVVVV'),
'POISSON' : (300, 3, 3, 'V', 'VVV'),
'POWER' : (337, 2, 2, 'V', 'VV'),
'PPMT' : (168, 4, 6, 'V', 'VVVVVV'),
'PRICE' : ( -1, 6, 7, 'V', 'VVVVVVV'),
'PRICEDISC' : ( -1, 4, 5, 'V', 'VVVVV'),
'PRICEMAT' : ( -1, 5, 6, 'V', 'VVVVVV'),
'PROB' : (317, 3, 4, 'V', 'AAVV'),
'PRODUCT' : (183, 1, 30, 'V', 'D+'),
'PROPER' : (114, 1, 1, 'V', 'V'),
'PV' : ( 56, 3, 5, 'V', 'VVVVV'),
'QUARTILE' : (327, 2, 2, 'V', 'RV'),
'QUOTIENT' : ( -1, 2, 2, 'V', 'VV'),
'RADIANS' : (342, 1, 1, 'V', 'V'),
'RAND' : ( 63, 0, 0, 'V', '-'),
'RANDBETWEEN' : ( -1, 2, 2, 'V', 'VV'),
'RANK' : (216, 2, 3, 'V', 'VRV'),
'RATE' : ( 60, 3, 6, 'V', 'VVVVVV'),
'RECEIVED' : ( -1, 4, 5, 'V', 'VVVVV'),
'REPLACE' : (119, 4, 4, 'V', 'VVVV'),
'REPLACEB' : (207, 4, 4, 'V', 'VVVV'),
'REPT' : ( 30, 2, 2, 'V', 'VV'),
'RIGHT' : (116, 1, 2, 'V', 'VV'),
'RIGHTB' : (209, 1, 2, 'V', 'VV'),
'ROMAN' : (354, 1, 2, 'V', 'VV'),
'ROUND' : ( 27, 2, 2, 'V', 'VV'),
'ROUNDDOWN' : (213, 2, 2, 'V', 'VV'),
'ROUNDUP' : (212, 2, 2, 'V', 'VV'),
'ROW' : ( 8, 0, 1, 'V', 'R'),
'ROWS' : ( 76, 1, 1, 'V', 'R'),
'RSQ' : (313, 2, 2, 'V', 'AA'),
'RTD' : (379, 3, 30, 'A', 'VVV+'),
'SEARCH' : ( 82, 2, 3, 'V', 'VVV'),
'SEARCHB' : (206, 2, 3, 'V', 'VVV'),
'SECOND' : ( 73, 1, 1, 'V', 'V'),
'SERIESSUM' : ( -1, 4, 4, 'V', 'VVVA'),
'SIGN' : ( 26, 1, 1, 'V', 'V'),
'SIN' : ( 15, 1, 1, 'V', 'V'),
'SINH' : (229, 1, 1, 'V', 'V'),
'SKEW' : (323, 1, 30, 'V', 'D+'),
'SLN' : (142, 3, 3, 'V', 'VVV'),
'SLOPE' : (315, 2, 2, 'V', 'AA'),
'SMALL' : (326, 2, 2, 'V', 'RV'),
'SQRT' : ( 20, 1, 1, 'V', 'V'),
'SQRTPI' : ( -1, 1, 1, 'V', 'V'),
'STANDARDIZE' : (297, 3, 3, 'V', 'VVV'),
'STDEV' : ( 12, 1, 30, 'V', 'D+'),
'STDEVA' : (366, 1, 30, 'V', 'D+'),
'STDEVP' : (193, 1, 30, 'V', 'D+'),
'STDEVPA' : (364, 1, 30, 'V', 'D+'),
'STEYX' : (314, 2, 2, 'V', 'AA'),
'SUBSTITUTE' : (120, 3, 4, 'V', 'VVVV'),
'SUBTOTAL' : (344, 2, 30, 'V', 'VR+'),
'SUM' : ( 4, 1, 30, 'V', 'D+'),
'SUMIF' : (345, 2, 3, 'V', 'RVR'),
'SUMPRODUCT' : (228, 1, 30, 'V', 'A+'),
'SUMSQ' : (321, 1, 30, 'V', 'D+'),
'SUMX2MY2' : (304, 2, 2, 'V', 'AA'),
'SUMX2PY2' : (305, 2, 2, 'V', 'AA'),
'SUMXMY2' : (303, 2, 2, 'V', 'AA'),
'SYD' : (143, 4, 4, 'V', 'VVVV'),
'T' : (130, 1, 1, 'V', 'R'),
'TAN' : ( 17, 1, 1, 'V', 'V'),
'TANH' : (231, 1, 1, 'V', 'V'),
'TBILLEQ' : ( -1, 3, 3, 'V', 'VVV'),
'TBILLPRICE' : ( -1, 3, 3, 'V', 'VVV'),
'TBILLYIELD' : ( -1, 3, 3, 'V', 'VVV'),
'TDIST' : (301, 3, 3, 'V', 'VVV'),
'TEXT' : ( 48, 2, 2, 'V', 'VV'),
'TIME' : ( 66, 3, 3, 'V', 'VVV'),
'TIMEVALUE' : (141, 1, 1, 'V', 'V'),
'TINV' : (332, 2, 2, 'V', 'VV'),
'TODAY' : (221, 0, 0, 'V', '-'),
'TRANSPOSE' : ( 83, 1, 1, 'A', 'A'),
'TREND' : ( 50, 1, 4, 'A', 'RRRV'),
'TRIM' : (118, 1, 1, 'V', 'V'),
'TRIMMEAN' : (331, 2, 2, 'V', 'RV'),
'TRUE' : ( 34, 0, 0, 'V', '-'),
'TRUNC' : (197, 1, 2, 'V', 'VV'),
'TTEST' : (316, 4, 4, 'V', 'AAVV'),
'TYPE' : ( 86, 1, 1, 'V', 'V'),
'UPPER' : (113, 1, 1, 'V', 'V'),
'USDOLLAR' : (204, 1, 2, 'V', 'VV'),
'VALUE' : ( 33, 1, 1, 'V', 'V'),
'VAR' : ( 46, 1, 30, 'V', 'D+'),
'VARA' : (367, 1, 30, 'V', 'D+'),
'VARP' : (194, 1, 30, 'V', 'D+'),
'VARPA' : (365, 1, 30, 'V', 'D+'),
'VDB' : (222, 5, 7, 'V', 'VVVVVVV'),
'VLOOKUP' : (102, 3, 4, 'V', 'VRRV'),
'WEEKDAY' : ( 70, 1, 2, 'V', 'VV'),
'WEEKNUM' : ( -1, 1, 2, 'V', 'VV'),
'WEIBULL' : (302, 4, 4, 'V', 'VVVV'),
'WORKDAY' : ( -1, 2, 3, 'V', 'VVR'),
'XIRR' : ( -1, 2, 3, 'V', 'AAV'),
'XNPV' : ( -1, 3, 3, 'V', 'VAA'),
'YEAR' : ( 69, 1, 1, 'V', 'V'),
'YEARFRAC' : ( -1, 2, 3, 'V', 'VVV'),
'YIELD' : ( -1, 6, 7, 'V', 'VVVVVVV'),
'YIELDDISC' : ( -1, 4, 5, 'V', 'VVVVV'),
'YIELDMAT' : ( -1, 5, 6, 'V', 'VVVVVV'),
'ZTEST' : (324, 2, 3, 'V', 'RVV'),
}
# Formulas Parse things
ptgExp = 0x01
ptgTbl = 0x02
ptgAdd = 0x03
ptgSub = 0x04
ptgMul = 0x05
ptgDiv = 0x06
ptgPower = 0x07
ptgConcat = 0x08
ptgLT = 0x09
ptgLE = 0x0a
ptgEQ = 0x0b
ptgGE = 0x0c
ptgGT = 0x0d
ptgNE = 0x0e
ptgIsect = 0x0f
ptgUnion = 0x10
ptgRange = 0x11
ptgUplus = 0x12
ptgUminus = 0x13
ptgPercent = 0x14
ptgParen = 0x15
ptgMissArg = 0x16
ptgStr = 0x17
ptgExtend = 0x18
ptgAttr = 0x19
ptgSheet = 0x1a
ptgEndSheet = 0x1b
ptgErr = 0x1c
ptgBool = 0x1d
ptgInt = 0x1e
ptgNum = 0x1f
ptgArrayR = 0x20
ptgFuncR = 0x21
ptgFuncVarR = 0x22
ptgNameR = 0x23
ptgRefR = 0x24
ptgAreaR = 0x25
ptgMemAreaR = 0x26
ptgMemErrR = 0x27
ptgMemNoMemR = 0x28
ptgMemFuncR = 0x29
ptgRefErrR = 0x2a
ptgAreaErrR = 0x2b
ptgRefNR = 0x2c
ptgAreaNR = 0x2d
ptgMemAreaNR = 0x2e
ptgMemNoMemNR = 0x2f
ptgNameXR = 0x39
ptgRef3dR = 0x3a
ptgArea3dR = 0x3b
ptgRefErr3dR = 0x3c
ptgAreaErr3dR = 0x3d
ptgArrayV = 0x40
ptgFuncV = 0x41
ptgFuncVarV = 0x42
ptgNameV = 0x43
ptgRefV = 0x44
ptgAreaV = 0x45
ptgMemAreaV = 0x46
ptgMemErrV = 0x47
ptgMemNoMemV = 0x48
ptgMemFuncV = 0x49
ptgRefErrV = 0x4a
ptgAreaErrV = 0x4b
ptgRefNV = 0x4c
ptgAreaNV = 0x4d
ptgMemAreaNV = 0x4e
ptgMemNoMemNV = 0x4f
ptgFuncCEV = 0x58
ptgNameXV = 0x59
ptgRef3dV = 0x5a
ptgArea3dV = 0x5b
ptgRefErr3dV = 0x5c
ptgAreaErr3dV = 0x5d
ptgArrayA = 0x60
ptgFuncA = 0x61
ptgFuncVarA = 0x62
ptgNameA = 0x63
ptgRefA = 0x64
ptgAreaA = 0x65
ptgMemAreaA = 0x66
ptgMemErrA = 0x67
ptgMemNoMemA = 0x68
ptgMemFuncA = 0x69
ptgRefErrA = 0x6a
ptgAreaErrA = 0x6b
ptgRefNA = 0x6c
ptgAreaNA = 0x6d
ptgMemAreaNA = 0x6e
ptgMemNoMemNA = 0x6f
ptgFuncCEA = 0x78
ptgNameXA = 0x79
ptgRef3dA = 0x7a
ptgArea3dA = 0x7b
ptgRefErr3dA = 0x7c
ptgAreaErr3dA = 0x7d
PtgNames = {
ptgExp : "ptgExp",
ptgTbl : "ptgTbl",
ptgAdd : "ptgAdd",
ptgSub : "ptgSub",
ptgMul : "ptgMul",
ptgDiv : "ptgDiv",
ptgPower : "ptgPower",
ptgConcat : "ptgConcat",
ptgLT : "ptgLT",
ptgLE : "ptgLE",
ptgEQ : "ptgEQ",
ptgGE : "ptgGE",
ptgGT : "ptgGT",
ptgNE : "ptgNE",
ptgIsect : "ptgIsect",
ptgUnion : "ptgUnion",
ptgRange : "ptgRange",
ptgUplus : "ptgUplus",
ptgUminus : "ptgUminus",
ptgPercent : "ptgPercent",
ptgParen : "ptgParen",
ptgMissArg : "ptgMissArg",
ptgStr : "ptgStr",
ptgExtend : "ptgExtend",
ptgAttr : "ptgAttr",
ptgSheet : "ptgSheet",
ptgEndSheet : "ptgEndSheet",
ptgErr : "ptgErr",
ptgBool : "ptgBool",
ptgInt : "ptgInt",
ptgNum : "ptgNum",
ptgArrayR : "ptgArrayR",
ptgFuncR : "ptgFuncR",
ptgFuncVarR : "ptgFuncVarR",
ptgNameR : "ptgNameR",
ptgRefR : "ptgRefR",
ptgAreaR : "ptgAreaR",
ptgMemAreaR : "ptgMemAreaR",
ptgMemErrR : "ptgMemErrR",
ptgMemNoMemR : "ptgMemNoMemR",
ptgMemFuncR : "ptgMemFuncR",
ptgRefErrR : "ptgRefErrR",
ptgAreaErrR : "ptgAreaErrR",
ptgRefNR : "ptgRefNR",
ptgAreaNR : "ptgAreaNR",
ptgMemAreaNR : "ptgMemAreaNR",
ptgMemNoMemNR : "ptgMemNoMemNR",
ptgNameXR : "ptgNameXR",
ptgRef3dR : "ptgRef3dR",
ptgArea3dR : "ptgArea3dR",
ptgRefErr3dR : "ptgRefErr3dR",
ptgAreaErr3dR : "ptgAreaErr3dR",
ptgArrayV : "ptgArrayV",
ptgFuncV : "ptgFuncV",
ptgFuncVarV : "ptgFuncVarV",
ptgNameV : "ptgNameV",
ptgRefV : "ptgRefV",
ptgAreaV : "ptgAreaV",
ptgMemAreaV : "ptgMemAreaV",
ptgMemErrV : "ptgMemErrV",
ptgMemNoMemV : "ptgMemNoMemV",
ptgMemFuncV : "ptgMemFuncV",
ptgRefErrV : "ptgRefErrV",
ptgAreaErrV : "ptgAreaErrV",
ptgRefNV : "ptgRefNV",
ptgAreaNV : "ptgAreaNV",
ptgMemAreaNV : "ptgMemAreaNV",
ptgMemNoMemNV : "ptgMemNoMemNV",
ptgFuncCEV : "ptgFuncCEV",
ptgNameXV : "ptgNameXV",
ptgRef3dV : "ptgRef3dV",
ptgArea3dV : "ptgArea3dV",
ptgRefErr3dV : "ptgRefErr3dV",
ptgAreaErr3dV : "ptgAreaErr3dV",
ptgArrayA : "ptgArrayA",
ptgFuncA : "ptgFuncA",
ptgFuncVarA : "ptgFuncVarA",
ptgNameA : "ptgNameA",
ptgRefA : "ptgRefA",
ptgAreaA : "ptgAreaA",
ptgMemAreaA : "ptgMemAreaA",
ptgMemErrA : "ptgMemErrA",
ptgMemNoMemA : "ptgMemNoMemA",
ptgMemFuncA : "ptgMemFuncA",
ptgRefErrA : "ptgRefErrA",
ptgAreaErrA : "ptgAreaErrA",
ptgRefNA : "ptgRefNA",
ptgAreaNA : "ptgAreaNA",
ptgMemAreaNA : "ptgMemAreaNA",
ptgMemNoMemNA : "ptgMemNoMemNA",
ptgFuncCEA : "ptgFuncCEA",
ptgNameXA : "ptgNameXA",
ptgRef3dA : "ptgRef3dA",
ptgArea3dA : "ptgArea3dA",
ptgRefErr3dA : "ptgRefErr3dA",
ptgAreaErr3dA : "ptgAreaErr3dA"
}
error_msg_by_code = {
0x00: "#NULL!", # intersection of two cell ranges is empty
0x07: "#DIV/0!", # division by zero
0x0F: "#VALUE!", # wrong type of operand
0x17: "#REF!", # illegal or deleted cell reference
0x1D: "#NAME?", # wrong function or range name
0x24: "#NUM!", # value range overflow
0x2A: "#N/A!" # argument or function not available
}
+261
View File
@@ -0,0 +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,
)
+253
View File
@@ -0,0 +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
+593
View File
@@ -0,0 +1,593 @@
# -*- coding: windows-1252 -*-
from . import Formatting
from .BIFFRecords import *
import collections
FIRST_USER_DEFINED_NUM_FORMAT_IDX = 164
class XFStyle(object):
def __init__(self):
self.num_format_str = 'General'
self.font = Formatting.Font()
self.alignment = Formatting.Alignment()
self.borders = Formatting.Borders()
self.pattern = Formatting.Pattern()
self.protection = Formatting.Protection()
default_style = XFStyle()
class StyleCollection(object):
_std_num_fmt_list = [
'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',
'@'
]
def __init__(self, style_compression=0):
self.style_compression = style_compression
self.stats = [0, 0, 0, 0, 0, 0]
self._font_id2x = {}
self._font_x2id = {}
self._font_val2x = {}
for x in (0, 1, 2, 3, 5): # The font with index 4 is omitted in all BIFF versions
font = Formatting.Font()
search_key = font._search_key()
self._font_id2x[font] = x
self._font_x2id[x] = font
self._font_val2x[search_key] = x
self._xf_id2x = {}
self._xf_x2id = {}
self._xf_val2x = {}
self._num_formats = {}
for fmtidx, fmtstr in zip(list(range(0, 23)), StyleCollection._std_num_fmt_list[0:23]):
self._num_formats[fmtstr] = fmtidx
for fmtidx, fmtstr in zip(list(range(37, 50)), StyleCollection._std_num_fmt_list[23:]):
self._num_formats[fmtstr] = fmtidx
self.default_style = XFStyle()
self._default_xf = self._add_style(self.default_style)[0]
def add(self, style):
if style == None:
return 0x10
return self._add_style(style)[1]
def _add_style(self, style):
num_format_str = style.num_format_str
if num_format_str in self._num_formats:
num_format_idx = self._num_formats[num_format_str]
else:
num_format_idx = (
FIRST_USER_DEFINED_NUM_FORMAT_IDX
+ len(self._num_formats)
- len(StyleCollection._std_num_fmt_list)
)
self._num_formats[num_format_str] = num_format_idx
font = style.font
if font in self._font_id2x:
font_idx = self._font_id2x[font]
self.stats[0] += 1
elif self.style_compression:
search_key = font._search_key()
font_idx = self._font_val2x.get(search_key)
if font_idx is not None:
self._font_id2x[font] = font_idx
self.stats[1] += 1
else:
font_idx = len(self._font_x2id) + 1 # Why plus 1? Font 4 is missing
self._font_id2x[font] = font_idx
self._font_val2x[search_key] = font_idx
self._font_x2id[font_idx] = font
self.stats[2] += 1
else:
font_idx = len(self._font_id2x) + 1
self._font_id2x[font] = font_idx
self.stats[2] += 1
gof = (style.alignment, style.borders, style.pattern, style.protection)
xf = (font_idx, num_format_idx) + gof
if xf in self._xf_id2x:
xf_index = self._xf_id2x[xf]
self.stats[3] += 1
elif self.style_compression == 2:
xf_key = (font_idx, num_format_idx) + tuple([obj._search_key() for obj in gof])
xf_index = self._xf_val2x.get(xf_key)
if xf_index is not None:
self._xf_id2x[xf] = xf_index
self.stats[4] += 1
else:
xf_index = 0x10 + len(self._xf_x2id)
self._xf_id2x[xf] = xf_index
self._xf_val2x[xf_key] = xf_index
self._xf_x2id[xf_index] = xf
self.stats[5] += 1
else:
xf_index = 0x10 + len(self._xf_id2x)
self._xf_id2x[xf] = xf_index
self.stats[5] += 1
if xf_index >= 0xFFF:
# 12 bits allowed, 0xFFF is a sentinel value
raise ValueError("More than 4094 XFs (styles)")
return xf, xf_index
def get_biff_data(self):
result = b''
result += self._all_fonts()
result += self._all_num_formats()
result += self._all_cell_styles()
result += self._all_styles()
return result
def _all_fonts(self):
result = b''
if self.style_compression:
alist = list(self._font_x2id.items())
else:
alist = [(x, o) for o, x in list(self._font_id2x.items())]
alist.sort()
for font_idx, font in alist:
result += font.get_biff_record().get()
return result
def _all_num_formats(self):
result = b''
alist = [
(v, k)
for k, v in list(self._num_formats.items())
if v >= FIRST_USER_DEFINED_NUM_FORMAT_IDX
]
alist.sort()
for fmtidx, fmtstr in alist:
result += NumberFormatRecord(fmtidx, fmtstr).get()
return result
def _all_cell_styles(self):
result = b''
for i in range(0, 16):
result += XFRecord(self._default_xf, 'style').get()
if self.style_compression == 2:
alist = list(self._xf_x2id.items())
else:
alist = [(x, o) for o, x in list(self._xf_id2x.items())]
alist.sort()
for xf_idx, xf in alist:
result += XFRecord(xf).get()
return result
def _all_styles(self):
return StyleRecord().get()
# easyxf and its supporting objects ###################################
class EasyXFException(Exception):
pass
class EasyXFCallerError(EasyXFException):
pass
class EasyXFAuthorError(EasyXFException):
pass
class IntULim(object):
# If astring represents a valid unsigned integer ('123', '0xabcd', etc)
# and it is <= limit, return the int value; otherwise return None.
def __init__(self, limit):
self.limit = limit
def __call__(self, astring):
try:
value = int(astring, 0)
except ValueError:
return None
if not 0 <= value <= self.limit:
return None
return value
bool_map = {
# Text values for all Boolean attributes
'1': 1, 'yes': 1, 'true': 1, 'on': 1,
'0': 0, 'no': 0, 'false': 0, 'off': 0,
}
border_line_map = {
# Text values for these borders attributes:
# left, right, top, bottom and diag
'no_line': 0x00,
'thin': 0x01,
'medium': 0x02,
'dashed': 0x03,
'dotted': 0x04,
'thick': 0x05,
'double': 0x06,
'hair': 0x07,
'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,
}
charset_map = {
# Text values for font.charset
'ansi_latin': 0x00,
'sys_default': 0x01,
'symbol': 0x02,
'apple_roman': 0x4d,
'ansi_jap_shift_jis': 0x80,
'ansi_kor_hangul': 0x81,
'ansi_kor_johab': 0x82,
'ansi_chinese_gbk': 0x86,
'ansi_chinese_big5': 0x88,
'ansi_greek': 0xa1,
'ansi_turkish': 0xa2,
'ansi_vietnamese': 0xa3,
'ansi_hebrew': 0xb1,
'ansi_arabic': 0xb2,
'ansi_baltic': 0xba,
'ansi_cyrillic': 0xcc,
'ansi_thai': 0xde,
'ansi_latin_ii': 0xee,
'oem_latin_i': 0xff,
}
# Text values for colour indices. "grey" is a synonym of "gray".
# The names are those given by Microsoft Excel 2003 to the colours
# in the default palette. There is no great correspondence with
# any W3C name-to-RGB mapping.
_colour_map_text = """\
aqua 0x31
black 0x08
blue 0x0C
blue_gray 0x36
bright_green 0x0B
brown 0x3C
coral 0x1D
cyan_ega 0x0F
dark_blue 0x12
dark_blue_ega 0x12
dark_green 0x3A
dark_green_ega 0x11
dark_purple 0x1C
dark_red 0x10
dark_red_ega 0x10
dark_teal 0x38
dark_yellow 0x13
gold 0x33
gray_ega 0x17
gray25 0x16
gray40 0x37
gray50 0x17
gray80 0x3F
green 0x11
ice_blue 0x1F
indigo 0x3E
ivory 0x1A
lavender 0x2E
light_blue 0x30
light_green 0x2A
light_orange 0x34
light_turquoise 0x29
light_yellow 0x2B
lime 0x32
magenta_ega 0x0E
ocean_blue 0x1E
olive_ega 0x13
olive_green 0x3B
orange 0x35
pale_blue 0x2C
periwinkle 0x18
pink 0x0E
plum 0x3D
purple_ega 0x14
red 0x0A
rose 0x2D
sea_green 0x39
silver_ega 0x16
sky_blue 0x28
tan 0x2F
teal 0x15
teal_ega 0x15
turquoise 0x0F
violet 0x14
white 0x09
yellow 0x0D"""
colour_map = {}
for _line in _colour_map_text.splitlines():
_name, _num = _line.split()
_num = int(_num, 0)
colour_map[_name] = _num
if 'gray' in _name:
colour_map[_name.replace('gray', 'grey')] = _num
del _colour_map_text, _line, _name, _num
pattern_map = {
# Text values for pattern.pattern
# xlwt/doc/pattern_examples.xls showcases all of these patterns.
'no_fill': 0,
'none': 0,
'solid': 1,
'solid_fill': 1,
'solid_pattern': 1,
'fine_dots': 2,
'alt_bars': 3,
'sparse_dots': 4,
'thick_horz_bands': 5,
'thick_vert_bands': 6,
'thick_backward_diag': 7,
'thick_forward_diag': 8,
'big_spots': 9,
'bricks': 10,
'thin_horz_bands': 11,
'thin_vert_bands': 12,
'thin_backward_diag': 13,
'thin_forward_diag': 14,
'squares': 15,
'diamonds': 16,
}
def any_str_func(s):
return s.strip()
def colour_index_func(s, maxval=0x7F):
try:
value = int(s, 0)
except ValueError:
return None
if not (0 <= value <= maxval):
return None
return value
colour_index_func_7 = colour_index_func
def colour_index_func_15(s):
return colour_index_func(s, maxval=0x7FFF)
def rotation_func(s):
try:
value = int(s, 0)
except ValueError:
return None
if not (-90 <= value <= 90):
raise EasyXFCallerError("rotation %d: should be -90 to +90 degrees" % value)
if value < 0:
value = 90 - value # encode as 91 to 180 (clockwise)
return value
xf_dict = {
'align': 'alignment', # synonym
'alignment': {
'dire': {
'general': 0,
'lr': 1,
'rl': 2,
},
'direction': 'dire',
'horiz': 'horz',
'horizontal': 'horz',
'horz': {
'general': 0,
'left': 1,
'center': 2,
'centre': 2, # "align: horiz centre" means xf.alignment.horz is set to 2
'right': 3,
'filled': 4,
'justified': 5,
'center_across_selection': 6,
'centre_across_selection': 6,
'distributed': 7,
},
'inde': IntULim(15), # restriction: 0 <= value <= 15
'indent': 'inde',
'rota': [{'stacked': 255, 'none': 0, }, rotation_func],
'rotation': 'rota',
'shri': bool_map,
'shrink': 'shri',
'shrink_to_fit': 'shri',
'vert': {
'top': 0,
'center': 1,
'centre': 1,
'bottom': 2,
'justified': 3,
'distributed': 4,
},
'vertical': 'vert',
'wrap': bool_map,
},
'border': 'borders',
'borders': {
'left': [border_line_map, IntULim(0x0d)],
'right': [border_line_map, IntULim(0x0d)],
'top': [border_line_map, IntULim(0x0d)],
'bottom': [border_line_map, IntULim(0x0d)],
'diag': [border_line_map, IntULim(0x0d)],
'top_colour': [colour_map, colour_index_func_7],
'bottom_colour': [colour_map, colour_index_func_7],
'left_colour': [colour_map, colour_index_func_7],
'right_colour': [colour_map, colour_index_func_7],
'diag_colour': [colour_map, colour_index_func_7],
'top_color': 'top_colour',
'bottom_color': 'bottom_colour',
'left_color': 'left_colour',
'right_color': 'right_colour',
'diag_color': 'diag-colour',
'need_diag_1': bool_map,
'need_diag_2': bool_map,
},
'font': {
'bold': bool_map,
'charset': charset_map,
'color': 'colour_index',
'color_index': 'colour_index',
'colour': 'colour_index',
'colour_index': [colour_map, colour_index_func_15],
'escapement': {'none': 0, 'superscript': 1, 'subscript': 2},
'family': {'none': 0, 'roman': 1, 'swiss': 2, 'modern': 3, 'script': 4, 'decorative': 5, },
'height': IntULim(0xFFFF), # practical limits are much narrower e.g. 160 to 1440 (8pt to 72pt)
'italic': bool_map,
'name': any_str_func,
'outline': bool_map,
'shadow': bool_map,
'struck_out': bool_map,
'underline': [bool_map, {'none': 0, 'single': 1, 'single_acc': 0x21, 'double': 2, 'double_acc': 0x22, }],
},
'pattern': {
'back_color': 'pattern_back_colour',
'back_colour': 'pattern_back_colour',
'fore_color': 'pattern_fore_colour',
'fore_colour': 'pattern_fore_colour',
'pattern': [pattern_map, IntULim(16)],
'pattern_back_color': 'pattern_back_colour',
'pattern_back_colour': [colour_map, colour_index_func_7],
'pattern_fore_color': 'pattern_fore_colour',
'pattern_fore_colour': [colour_map, colour_index_func_7],
},
'protection': {
'cell_locked' : bool_map,
'formula_hidden': bool_map,
},
}
def _esplit(s, split_char, esc_char="\\"):
escaped = False
olist = ['']
for c in s:
if escaped:
olist[-1] += c
escaped = False
elif c == esc_char:
escaped = True
elif c == split_char:
olist.append('')
else:
olist[-1] += c
return olist
def _parse_strg_to_obj(strg, obj, parse_dict,
field_sep=",", line_sep=";", intro_sep=":", esc_char="\\", debug=False):
for line in _esplit(strg, line_sep, esc_char):
line = line.strip()
if not line:
break
split_line = _esplit(line, intro_sep, esc_char)
if len(split_line) != 2:
raise EasyXFCallerError('line %r should have exactly 1 "%c"' % (line, intro_sep))
section, item_str = split_line
section = section.strip().lower()
for counter in range(2):
result = parse_dict.get(section)
if result is None:
raise EasyXFCallerError('section %r is unknown' % section)
if isinstance(result, dict):
break
if not isinstance(result, str):
raise EasyXFAuthorError(
'section %r should map to dict or str object; found %r' % (section, type(result)))
# synonym
old_section = section
section = result
else:
raise EasyXFAuthorError('Attempt to define synonym of synonym (%r: %r)' % (old_section, result))
section_dict = result
section_obj = getattr(obj, section, None)
if section_obj is None:
raise EasyXFAuthorError('instance of %s class has no attribute named %s' % (obj.__class__.__name__, section))
for kv_str in _esplit(item_str, field_sep, esc_char):
guff = kv_str.split()
if not guff:
continue
k = guff[0].lower().replace('-', '_')
v = ' '.join(guff[1:])
if not v:
raise EasyXFCallerError("no value supplied for %s.%s" % (section, k))
for counter in range(2):
result = section_dict.get(k)
if result is None:
raise EasyXFCallerError('%s.%s is not a known attribute' % (section, k))
if not isinstance(result, str):
break
# synonym
old_k = k
k = result
else:
raise EasyXFAuthorError('Attempt to define synonym of synonym (%r: %r)' % (old_k, result))
value_info = result
if not isinstance(value_info, list):
value_info = [value_info]
for value_rule in value_info:
if isinstance(value_rule, dict):
# dict maps strings to integer field values
vl = v.lower().replace('-', '_')
if vl in value_rule:
value = value_rule[vl]
break
elif isinstance(value_rule, collections.Callable):
value = value_rule(v)
if value is not None:
break
else:
raise EasyXFAuthorError("unknown value rule for attribute %r: %r" % (k, value_rule))
else:
raise EasyXFCallerError("unexpected value %r for %s.%s" % (v, section, k))
try:
orig = getattr(section_obj, k)
except AttributeError:
raise EasyXFAuthorError('%s.%s in dictionary but not in supplied object' % (section, k))
if debug: print("+++ %s.%s = %r # %s; was %r" % (section, k, value, v, orig))
setattr(section_obj, k, value)
def easyxf(strg_to_parse="", num_format_str=None,
field_sep=",", line_sep=";", intro_sep=":", esc_char="\\", debug=False):
xfobj = XFStyle()
if num_format_str is not None:
xfobj.num_format_str = num_format_str
if strg_to_parse:
_parse_strg_to_obj(strg_to_parse, xfobj, xf_dict,
field_sep=field_sep, line_sep=line_sep, intro_sep=intro_sep, esc_char=esc_char, debug=debug)
return xfobj
+79
View File
@@ -0,0 +1,79 @@
'''
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
+196
View File
@@ -0,0 +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("'", "''") + "'"
+635
View File
@@ -0,0 +1,635 @@
'''
Record Order in BIFF8
Workbook Globals Substream
BOF Type = workbook globals
Interface Header
MMS
Interface End
WRITEACCESS
CODEPAGE
DSF
TABID
FNGROUPCOUNT
Workbook Protection Block
WINDOWPROTECT
PROTECT
PASSWORD
PROT4REV
PROT4REVPASS
BACKUP
HIDEOBJ
WINDOW1
DATEMODE
PRECISION
REFRESHALL
BOOKBOOL
FONT +
FORMAT *
XF +
STYLE +
? PALETTE
USESELFS
BOUNDSHEET +
COUNTRY
? Link Table
SST
ExtSST
EOF
'''
from . import BIFFRecords
from . import Style
class Workbook(object):
#################################################################
## Constructor
#################################################################
def __init__(self, encoding='ascii', style_compression=0):
self.encoding = encoding
self.__owner = 'None'
self.__country_code = None # 0x07 is Russia :-)
self.__wnd_protect = 0
self.__obj_protect = 0
self.__protect = 0
self.__backup_on_save = 0
# for WINDOW1 record
self.__hpos_twips = 0x01E0
self.__vpos_twips = 0x005A
self.__width_twips = 0x3FCF
self.__height_twips = 0x2A4E
self.__active_sheet = 0
self.__first_tab_index = 0
self.__selected_tabs = 0x01
self.__tab_width_twips = 0x0258
self.__wnd_hidden = 0
self.__wnd_mini = 0
self.__hscroll_visible = 1
self.__vscroll_visible = 1
self.__tabs_visible = 1
self.__styles = Style.StyleCollection(style_compression)
self.__dates_1904 = 0
self.__use_cell_values = 1
self.__sst = BIFFRecords.SharedStringTable(self.encoding)
self.__worksheets = []
self.__worksheet_idx_from_name = {}
self.__sheet_refs = {}
self._supbook_xref = {}
self._xcall_xref = {}
self._ownbook_supbookx = None
self._ownbook_supbook_ref = None
self._xcall_supbookx = None
self._xcall_supbook_ref = None
#################################################################
## Properties, "getters", "setters"
#################################################################
def get_style_stats(self):
return self.__styles.stats[:]
def set_owner(self, value):
self.__owner = value
def get_owner(self):
return self.__owner
owner = property(get_owner, set_owner)
#################################################################
def set_country_code(self, value):
self.__country_code = value
def get_country_code(self):
return self.__country_code
country_code = property(get_country_code, set_country_code)
#################################################################
def set_wnd_protect(self, value):
self.__wnd_protect = int(value)
def get_wnd_protect(self):
return bool(self.__wnd_protect)
wnd_protect = property(get_wnd_protect, set_wnd_protect)
#################################################################
def set_obj_protect(self, value):
self.__obj_protect = int(value)
def get_obj_protect(self):
return bool(self.__obj_protect)
obj_protect = property(get_obj_protect, set_obj_protect)
#################################################################
def set_protect(self, value):
self.__protect = int(value)
def get_protect(self):
return bool(self.__protect)
protect = property(get_protect, set_protect)
#################################################################
def set_backup_on_save(self, value):
self.__backup_on_save = int(value)
def get_backup_on_save(self):
return bool(self.__backup_on_save)
backup_on_save = property(get_backup_on_save, set_backup_on_save)
#################################################################
def set_hpos(self, value):
self.__hpos_twips = value & 0xFFFF
def get_hpos(self):
return self.__hpos_twips
hpos = property(get_hpos, set_hpos)
#################################################################
def set_vpos(self, value):
self.__vpos_twips = value & 0xFFFF
def get_vpos(self):
return self.__vpos_twips
vpos = property(get_vpos, set_vpos)
#################################################################
def set_width(self, value):
self.__width_twips = value & 0xFFFF
def get_width(self):
return self.__width_twips
width = property(get_width, set_width)
#################################################################
def set_height(self, value):
self.__height_twips = value & 0xFFFF
def get_height(self):
return self.__height_twips
height = property(get_height, set_height)
#################################################################
def set_active_sheet(self, value):
self.__active_sheet = value & 0xFFFF
self.__first_tab_index = self.__active_sheet
def get_active_sheet(self):
return self.__active_sheet
active_sheet = property(get_active_sheet, set_active_sheet)
#################################################################
def set_tab_width(self, value):
self.__tab_width_twips = value & 0xFFFF
def get_tab_width(self):
return self.__tab_width_twips
tab_width = property(get_tab_width, set_tab_width)
#################################################################
def set_wnd_visible(self, value):
self.__wnd_hidden = int(not value)
def get_wnd_visible(self):
return not bool(self.__wnd_hidden)
wnd_visible = property(get_wnd_visible, set_wnd_visible)
#################################################################
def set_wnd_mini(self, value):
self.__wnd_mini = int(value)
def get_wnd_mini(self):
return bool(self.__wnd_mini)
wnd_mini = property(get_wnd_mini, set_wnd_mini)
#################################################################
def set_hscroll_visible(self, value):
self.__hscroll_visible = int(value)
def get_hscroll_visible(self):
return bool(self.__hscroll_visible)
hscroll_visible = property(get_hscroll_visible, set_hscroll_visible)
#################################################################
def set_vscroll_visible(self, value):
self.__vscroll_visible = int(value)
def get_vscroll_visible(self):
return bool(self.__vscroll_visible)
vscroll_visible = property(get_vscroll_visible, set_vscroll_visible)
#################################################################
def set_tabs_visible(self, value):
self.__tabs_visible = int(value)
def get_tabs_visible(self):
return bool(self.__tabs_visible)
tabs_visible = property(get_tabs_visible, set_tabs_visible)
#################################################################
def set_dates_1904(self, value):
self.__dates_1904 = int(value)
def get_dates_1904(self):
return bool(self.__dates_1904)
dates_1904 = property(get_dates_1904, set_dates_1904)
#################################################################
def set_use_cell_values(self, value):
self.__use_cell_values = int(value)
def get_use_cell_values(self):
return bool(self.__use_cell_values)
use_cell_values = property(get_use_cell_values, set_use_cell_values)
#################################################################
def get_default_style(self):
return self.__styles.default_style
default_style = property(get_default_style)
##################################################################
## Methods
##################################################################
def add_style(self, style):
return self.__styles.add(style)
def add_str(self, s):
return self.__sst.add_str(s)
def del_str(self, sst_idx):
self.__sst.del_str(sst_idx)
def str_index(self, s):
return self.__sst.str_index(s)
def add_sheet(self, sheetname, cell_overwrite_ok=False):
from . import Worksheet, Utils
if not isinstance(sheetname, str):
sheetname = sheetname.decode(self.encoding)
if not Utils.valid_sheet_name(sheetname):
raise Exception("invalid worksheet name %r" % sheetname)
lower_name = sheetname.lower()
if lower_name in self.__worksheet_idx_from_name:
raise Exception("duplicate worksheet name %r" % sheetname)
self.__worksheet_idx_from_name[lower_name] = len(self.__worksheets)
self.__worksheets.append(Worksheet(sheetname, self, cell_overwrite_ok))
return self.__worksheets[-1]
def get_sheet(self, sheetnum):
return self.__worksheets[sheetnum]
def raise_bad_sheetname(self, sheetname):
raise Exception("Formula: unknown sheet name %s" % sheetname)
def convert_sheetindex(self, strg_ref, n_sheets):
idx = int(strg_ref)
if 0 <= idx < n_sheets:
return idx
msg = "Formula: sheet index (%s) >= number of sheets (%d)" % (strg_ref, n_sheets)
raise Exception(msg)
def _get_supbook_index(self, tag):
if tag in self._supbook_xref:
return self._supbook_xref[tag]
self._supbook_xref[tag] = idx = len(self._supbook_xref)
return idx
def setup_ownbook(self):
self._ownbook_supbookx = self._get_supbook_index(('ownbook', 0))
self._ownbook_supbook_ref = None
reference = (self._ownbook_supbookx, 0xFFFE, 0xFFFE)
if reference in self.__sheet_refs:
raise Exception("can't happen")
self.__sheet_refs[reference] = self._ownbook_supbook_ref = len(self.__sheet_refs)
def setup_xcall(self):
self._xcall_supbookx = self._get_supbook_index(('xcall', 0))
self._xcall_supbook_ref = None
reference = (self._xcall_supbookx, 0xFFFE, 0xFFFE)
if reference in self.__sheet_refs:
raise Exception("can't happen")
self.__sheet_refs[reference] = self._xcall_supbook_ref = len(self.__sheet_refs)
def add_sheet_reference(self, formula):
patches = []
n_sheets = len(self.__worksheets)
sheet_refs, xcall_refs = formula.get_references()
for ref0, ref1, offset in sheet_refs:
if not ref0.isdigit():
try:
ref0n = self.__worksheet_idx_from_name[ref0.lower()]
except KeyError:
self.raise_bad_sheetname(ref0)
else:
ref0n = self.convert_sheetindex(ref0, n_sheets)
if ref1 == ref0:
ref1n = ref0n
elif not ref1.isdigit():
try:
ref1n = self.__worksheet_idx_from_name[ref1.lower()]
except KeyError:
self.raise_bad_sheetname(ref1)
else:
ref1n = self.convert_sheetindex(ref1, n_sheets)
if ref1n < ref0n:
msg = "Formula: sheets out of order; %r:%r -> (%d, %d)" \
% (ref0, ref1, ref0n, ref1n)
raise Exception(msg)
if self._ownbook_supbookx is None:
self.setup_ownbook()
reference = (self._ownbook_supbookx, ref0n, ref1n)
if reference in self.__sheet_refs:
patches.append((offset, self.__sheet_refs[reference]))
else:
nrefs = len(self.__sheet_refs)
if nrefs > 65535:
raise Exception('More than 65536 inter-sheet references')
self.__sheet_refs[reference] = nrefs
patches.append((offset, nrefs))
for funcname, offset in xcall_refs:
if self._ownbook_supbookx is None:
self.setup_ownbook()
if self._xcall_supbookx is None:
self.setup_xcall()
# print funcname, self._supbook_xref
patches.append((offset, self._xcall_supbook_ref))
if not isinstance(funcname, str):
funcname = funcname.decode(self.encoding)
if funcname in self._xcall_xref:
idx = self._xcall_xref[funcname]
else:
self._xcall_xref[funcname] = idx = len(self._xcall_xref)
patches.append((offset + 2, idx + 1))
formula.patch_references(patches)
##################################################################
## BIFF records generation
##################################################################
def __bof_rec(self):
return BIFFRecords.Biff8BOFRecord(BIFFRecords.Biff8BOFRecord.BOOK_GLOBAL).get()
def __eof_rec(self):
return BIFFRecords.EOFRecord().get()
def __intf_hdr_rec(self):
return BIFFRecords.InteraceHdrRecord().get()
def __intf_end_rec(self):
return BIFFRecords.InteraceEndRecord().get()
def __intf_mms_rec(self):
return BIFFRecords.MMSRecord().get()
def __write_access_rec(self):
return BIFFRecords.WriteAccessRecord(self.__owner).get()
def __wnd_protect_rec(self):
return BIFFRecords.WindowProtectRecord(self.__wnd_protect).get()
def __obj_protect_rec(self):
return BIFFRecords.ObjectProtectRecord(self.__obj_protect).get()
def __protect_rec(self):
return BIFFRecords.ProtectRecord(self.__protect).get()
def __password_rec(self):
return BIFFRecords.PasswordRecord().get()
def __prot4rev_rec(self):
return BIFFRecords.Prot4RevRecord().get()
def __prot4rev_pass_rec(self):
return BIFFRecords.Prot4RevPassRecord().get()
def __backup_rec(self):
return BIFFRecords.BackupRecord(self.__backup_on_save).get()
def __hide_obj_rec(self):
return BIFFRecords.HideObjRecord().get()
def __window1_rec(self):
flags = 0
flags |= (self.__wnd_hidden) << 0
flags |= (self.__wnd_mini) << 1
flags |= (self.__hscroll_visible) << 3
flags |= (self.__vscroll_visible) << 4
flags |= (self.__tabs_visible) << 5
return BIFFRecords.Window1Record(self.__hpos_twips, self.__vpos_twips,
self.__width_twips, self.__height_twips,
flags,
self.__active_sheet, self.__first_tab_index,
self.__selected_tabs, self.__tab_width_twips).get()
def __codepage_rec(self):
return BIFFRecords.CodepageBiff8Record().get()
def __country_rec(self):
if not self.__country_code:
return b''
return BIFFRecords.CountryRecord(self.__country_code, self.__country_code).get()
def __dsf_rec(self):
return BIFFRecords.DSFRecord().get()
def __tabid_rec(self):
return BIFFRecords.TabIDRecord(len(self.__worksheets)).get()
def __fngroupcount_rec(self):
return BIFFRecords.FnGroupCountRecord().get()
def __datemode_rec(self):
return BIFFRecords.DateModeRecord(self.__dates_1904).get()
def __precision_rec(self):
return BIFFRecords.PrecisionRecord(self.__use_cell_values).get()
def __refresh_all_rec(self):
return BIFFRecords.RefreshAllRecord().get()
def __bookbool_rec(self):
return BIFFRecords.BookBoolRecord().get()
def __all_fonts_num_formats_xf_styles_rec(self):
return self.__styles.get_biff_data()
def __palette_rec(self):
result = b''
return result
def __useselfs_rec(self):
return BIFFRecords.UseSelfsRecord().get()
def __boundsheets_rec(self, data_len_before, data_len_after, sheet_biff_lens):
# .................................
# BOUNDSEHEET0
# BOUNDSEHEET1
# BOUNDSEHEET2
# ..................................
# WORKSHEET0
# WORKSHEET1
# WORKSHEET2
boundsheets_len = 0
for sheet in self.__worksheets:
boundsheets_len += len(BIFFRecords.BoundSheetRecord(
0x00, sheet.visibility, sheet.name, self.encoding
).get())
start = data_len_before + boundsheets_len + data_len_after
result = b''
for sheet_biff_len, sheet in zip(sheet_biff_lens, self.__worksheets):
result += BIFFRecords.BoundSheetRecord(
start, sheet.visibility, sheet.name, self.encoding
).get()
start += sheet_biff_len
return result
def __all_links_rec(self):
pieces = []
temp = [(idx, tag) for tag, idx in list(self._supbook_xref.items())]
temp.sort()
for idx, tag in temp:
stype, snum = tag
if stype == 'ownbook':
rec = BIFFRecords.InternalReferenceSupBookRecord(len(self.__worksheets)).get()
pieces.append(rec)
elif stype == 'xcall':
rec = BIFFRecords.XcallSupBookRecord().get()
pieces.append(rec)
temp = [(idx, name) for name, idx in list(self._xcall_xref.items())]
temp.sort()
for idx, name in temp:
rec = BIFFRecords.ExternnameRecord(
options=0, index=0, name=name, fmla='\x02\x00\x1c\x17').get()
pieces.append(rec)
else:
raise Exception('unknown supbook stype %r' % stype)
if len(self.__sheet_refs) > 0:
# get references in index order
temp = [(idx, ref) for ref, idx in list(self.__sheet_refs.items())]
temp.sort()
temp = [ref for idx, ref in temp]
externsheet_record = BIFFRecords.ExternSheetRecord(temp).get()
pieces.append(externsheet_record)
return b''.join(pieces)
def __sst_rec(self):
return self.__sst.get_biff_record()
def __ext_sst_rec(self, abs_stream_pos):
return b''
#return BIFFRecords.ExtSSTRecord(abs_stream_pos, self.sst_record.str_placement,
#self.sst_record.portions_len).get()
def get_biff_data(self):
before = b''
before += self.__bof_rec()
before += self.__intf_hdr_rec()
before += self.__intf_mms_rec()
before += self.__intf_end_rec()
before += self.__write_access_rec()
before += self.__codepage_rec()
before += self.__dsf_rec()
before += self.__tabid_rec()
before += self.__fngroupcount_rec()
before += self.__wnd_protect_rec()
before += self.__protect_rec()
before += self.__obj_protect_rec()
before += self.__password_rec()
before += self.__prot4rev_rec()
before += self.__prot4rev_pass_rec()
before += self.__backup_rec()
before += self.__hide_obj_rec()
before += self.__window1_rec()
before += self.__datemode_rec()
before += self.__precision_rec()
before += self.__refresh_all_rec()
before += self.__bookbool_rec()
before += self.__all_fonts_num_formats_xf_styles_rec()
before += self.__palette_rec()
before += self.__useselfs_rec()
country = self.__country_rec()
all_links = self.__all_links_rec()
shared_str_table = self.__sst_rec()
after = country + all_links + shared_str_table
ext_sst = self.__ext_sst_rec(0) # need fake cause we need calc stream pos
eof = self.__eof_rec()
self.__worksheets[self.__active_sheet].selected = True
sheets = b''
sheet_biff_lens = []
for sheet in self.__worksheets:
data = sheet.get_biff_data()
sheets += data
sheet_biff_lens.append(len(data))
bundlesheets = self.__boundsheets_rec(len(before), len(after)+len(ext_sst)+len(eof), sheet_biff_lens)
sst_stream_pos = len(before) + len(bundlesheets) + len(country) + len(all_links)
ext_sst = self.__ext_sst_rec(sst_stream_pos)
return before + bundlesheets + after + ext_sst + eof + sheets
def save(self, filename):
from . import CompoundDoc
doc = CompoundDoc.XlsDoc()
doc.save(filename, self.get_biff_data())
File diff suppressed because it is too large Load Diff
+9
View File
@@ -0,0 +1,9 @@
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 *
File diff suppressed because it is too large Load Diff