Merge branch 'hotfix/dict' into develop

- Added testing for data set width/height

.gitignore

@@ -4,8 +4,8 @@ dist/*
 MANIFEST
 
 # python skin
-.pyc
+*.pyc
-.pyo
+*.pyo
 
 # osx noise
 .DS_Store
@@ -15,3 +15,5 @@ profile
 .idea
 .idea/*
 
+# vi noise
+*.swp

HISTORY.rst

@@ -1,6 +1,27 @@
 History
 =======
 
+0.7.0 (2010-09-20)
+------------------
+
+* Renamed DataBook Databook for consistiency.
+* Export properties changed to methods (XLS filename / StringIO bug).
+* Optional Dataset.xls(path='filename') support (for writing on windows).
+* Added utf-8 on the worksheet level.
+
+
+0.6.4 (2010-09-19)
+------------------
+
+* Updated unicode export for XLS.
+* More exhaustive unit tests.
+
+
+0.6.3 (2010-09-14)
+------------------
+* Added Dataset.append() support for columns.
+
+
 0.6.2 (2010-09-13)
 ------------------
 * Fixed Dataset.append() error on empty dataset.

README.rst

@@ -31,11 +31,11 @@ Usage
     
 Populate fresh data files: ::
     
-    headers = ('first_name', 'last_name', 'gpa')
+    headers = ('first_name', 'last_name')
 
     data = [
-        ('John', 'Adams', 90),
+        ('John', 'Adams'),
-        ('George', 'Washington', 67)
+        ('George', 'Washington')
     ]
     
     data = tablib.Dataset(*data, headers=headers)
@@ -43,7 +43,11 @@ Populate fresh data files: ::
 
 Intelligently add new rows: ::
 
-    >>> data.append(('Henry', 'Ford', 83))
+    >>> data.append(('Henry', 'Ford'))
+
+Intelligently add new columns: ::
+
+    >>> data.append(col=('age', 90, 67, 83))
     
 Slice rows:  ::
 
@@ -66,7 +70,7 @@ JSON!
 +++++
 ::
 
-	>>> print data.json
+	>>> print data.json()
 	[
 	  {
 	    "last_name": "Adams",
@@ -85,7 +89,7 @@ YAML!
 +++++
 ::
 
-	>>> print data.yaml
+	>>> print data.yaml()
 	- {age: 90, first_name: John, last_name: Adams}
 	- {age: 83, first_name: Henry, last_name: Ford}
 	
@@ -93,7 +97,7 @@ CSV...
 ++++++
 ::
 
-	>>> print data.csv
+	>>> print data.csv()
 	first_name,last_name,age 
 	John,Adams,90 
 	Henry,Ford,83 
@@ -102,7 +106,7 @@ EXCEL!
 ++++++
 ::
 
-	>>> open('people.xls').write(data.xls)
+	>>> data.xls('people.xls')
 
 It's that easy.
 	
@@ -122,7 +126,7 @@ Or, if you absolutely must: ::
 Contribute
 ----------
 
-If you'd like to contribute, simply fork `the repository`_, commit your changes, and send a pull request. Make sure you add yourself to AUTHORS_.
+If you'd like to contribute, simply fork `the repository`_, commit your changes to the **develop** branch (or branch off of it), and send a pull request. Make sure you add yourself to AUTHORS_.
 
 
 Roadmap
@@ -133,6 +137,7 @@ Roadmap
 - Auto-detect import format
 - Add possible other exports (SQL?)
 - Possibly plugin-ify format architecture
+- Ability to assign types to rows (set, regex=, &c.)
 - Plugin support
 
 .. _`the repository`: http://github.com/kennethreitz/tablib

setup.py

@@ -18,7 +18,7 @@ if sys.argv[-1] == "publish":
 
 setup(
 	name='tablib',
-	version='0.6.2',
+	version='0.7.0',
 	description='Format agnostic tabular data library (XLS, JSON, YAML, CSV)',
 	long_description=open('README.rst').read() + '\n\n' +
 	                 open('HISTORY.rst').read(),

tablib/core.py

@@ -8,7 +8,7 @@
 
 
 import csv
-import cStringIO
+import StringIO
 import random
 
 import simplejson as json
@@ -21,8 +21,8 @@ from helpers import *
 # __all__ = ['Dataset', 'DataBook']
 
 __name__ = 'tablib'
-__version__ = '0.6.1'
+__version__ = '0.7.0'
-__build__ = 0x000601
+__build__ = 0x000700
 __author__ = 'Kenneth Reitz'
 __license__ = 'MIT'
 __copyright__ = 'Copyright 2010 Kenneth Reitz'
@@ -54,7 +54,7 @@ class Dataset(object):
 
 
 	def __getitem__(self, key):
-		if is_string(key):
+		if isinstance(key, basestring):
 			if key in self.headers:
 				pos = self.headers.index(key) # get 'key' index from each data
 				return [row[pos] for row in self._data]
@@ -80,10 +80,15 @@ class Dataset(object):
 			return '<dataset object>'
 
 
-	def _validate(self, row=None, safety=False):
+	def _validate(self, row=None, col=None, safety=False):
 		"""Assures size of every row in dataset is of proper proportions."""
 		if row:
 			is_valid = (len(row) == self.width) if self.width else True
+		elif col:
+			if self.headers:
+				is_valid = (len(col) - 1) == self.height
+			else:
+				is_valid = (len(col) == self.height) if self.height else True
 		else:
 			is_valid = all((len(x)== self.width for x in self._data))
 
@@ -130,33 +135,39 @@ class Dataset(object):
 		"""Headers property."""
 		return self.__headers
 
+
 	@headers.setter
 	def headers(self, collection):
 		"""Validating headers setter."""
 		self._validate(collection)
-		self.__headers = collection
+		if collection:
+			try:
+				self.__headers = list(collection)
+			except TypeError, why:
+				raise TypeError
+		else:
+			self.__headers = None
+
 
 	@property
 	def dict(self):
 		"""Returns python dict of Dataset."""
 		return self._package()
 
-	@property
+
 	def json(self):
 		"""Returns JSON representation of Dataset."""
 		return json.dumps(self.dict)
 
 
-	@property
 	def yaml(self):
 		"""Returns YAML representation of Dataset."""
 		return yaml.dump(self.dict)
 
 
-	@property
 	def csv(self):
 		"""Returns CSV representation of Dataset."""
-		stream = cStringIO.StringIO()
+		stream = StringIO.StringIO()
 		_csv = csv.writer(stream)
 
 		for row in self._package(dicts=False):
@@ -165,34 +176,58 @@ class Dataset(object):
 		return stream.getvalue()
 
 
-	@property
+	def xls(self, path=None):
-	def xls(self):
 		"""Returns XLS representation of Dataset."""
-		stream = cStringIO.StringIO()
 
-		wb = xlwt.Workbook()
+		wb = xlwt.Workbook(encoding='utf8')
 		ws = wb.add_sheet(self.title if self.title else 'Tabbed Dataset')
+
 		for i, row in enumerate(self._package(dicts=False)):
 			for j, col in enumerate(row):
-				ws.write(i, j, str(col))
+				ws.write(i, j, col)
 
-		wb.save(stream)
+		if path:
-		return stream.getvalue()
+			wb.save(path)
+			return True
+		else:
+			stream = StringIO.StringIO()
+			wb.save(stream)
+			return stream.getvalue()
 
 
-	def append(self, row):
+	def append(self, row=None, col=None):
 		"""Adds a row to the end of Dataset"""
-		self._validate(row)
+		if row:
-		self._data.append(tuple(row))
+			self._validate(row)
+			self._data.append(tuple(row))
+		elif col:
+			self._validate(col=col)
+
+			if self.headers:
+				# pop the first item off, add to headers
+				self.headers.append(col[0])
+				col = col[1:]
+		
+			if self.height and self.width:
+
+				for i, row in enumerate(self._data):
+					_row = list(row)
+					_row.append(col[i])
+					self._data[i] = tuple(_row)
+			else:
+				self._data = [tuple([row]) for row in col]
 
 
-	def index(self, i, row):
+	def insert(self, i, row=None, col=None):
 		"""Inserts a row at given position in Dataset"""
-		self._validate(row)
+		if row:
-		self._data.insert(i, tuple(row))
+			self._validate(row)
+			self._data.insert(i, tuple(row))
+		elif col:
+			pass
 
 
-class DataBook(object):
+class Databook(object):
 	"""A book of Dataset objects.
 	   Currently, this exists only for XLS workbook support.
 	"""
@@ -200,12 +235,14 @@ class DataBook(object):
 	def __init__(self, sets=[]):
 		self._datasets = sets
 
+
 	def __repr__(self):
 		try:
 			return '<%s databook>' % (self.title.lower())
 		except AttributeError:
 			return '<databook object>'
 
+
 	def add_sheet(self, dataset):
 		"""Add given dataset ."""
 		if type(dataset) is Dataset:
@@ -213,6 +250,7 @@ class DataBook(object):
 		else:
 			raise InvalidDatasetType
 
+
 	def _package(self):
 		collector = []
 		for dset in self._datasets:
@@ -222,48 +260,58 @@ class DataBook(object):
 			))
 		return collector
 
+
 	@property
 	def size(self):
 		"""The number of the Datasets within DataBook."""
 		return len(self._datasets)
 
 
-	@property
+	def xls(self, path=None):
-	def xls(self):
 		"""Returns XLS representation of DataBook."""
 
-		stream = cStringIO.StringIO()
-		wb = xlwt.Workbook()
 
-		for dset in self._datasets:
+		wb = xlwt.Workbook(encoding='utf8')
-			ws = wb.add_sheet(dset.title if dset.title else 'Tabbed Dataset %s' % (int(random.random() * 100000000)))
+
+		for i, dset in enumerate(self._datasets):
+			ws = wb.add_sheet(dset.title if dset.title else 'Sheet%s' % (i))
 
 			#for row in self._package(dicts=False):
 			for i, row in enumerate(dset._package(dicts=False)):
 				for j, col in enumerate(row):
-					ws.write(i, j, str(col))
+					ws.write(i, j, col)
+
+
+
+		if path:
+			wb.save(path)
+			return True
+		else:
+			stream = cStringIO.StringIO()
+			wb.save(stream)
+			return stream.getvalue()
 
-		wb.save(stream)
-		return stream.getvalue()
 
-	@property
 	def json(self):
 		"""Returns JSON representation of Databook."""
 
 		return json.dumps(self._package())
 
-	@property
+
 	def yaml(self):
 		"""Returns YAML representation of Databook."""
 
 		return yaml.dump(self._package())
 
 
+
 class InvalidDatasetType(Exception):
 	"Only Datasets can be added to a DataBook"
 
+
 class InvalidDimensions(Exception):
 	"Invalid size"
 
+	
 class UnsupportedFormat(NotImplementedError):
 	"Format is not supported"

tablib/helpers.py

@@ -10,16 +10,12 @@ class Struct(object):
 		self.__dict__.update(entries)
 		
 	def __getitem__(self, key):
-		return getattr(self, key)
+		return getattr(self, key, None)
 
 
 def piped():
-	"""Returns piped input via stdin, else False"""
+	"""Returns piped input via stdin, else False."""
 	with sys.stdin as stdin:
+		# TTY is only way to detect if stdin contains data
 		return stdin.read() if not stdin.isatty() else None
 
-		
-def is_string(obj):
-	"""Tests if an object is a string"""
-
-	return True if type(obj).__name__ == 'str' else False

test_tablib.py

@@ -1,39 +1,185 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
+"""Tests for tablib."""
+
 import unittest
 
 import tablib
 
+
 class TablibTestCase(unittest.TestCase):
+	"""Tablib test cases."""
+
 	def setUp(self):
-		pass
+		"""Create simple data set with headers."""
-
+		global data
-	def tearDown(self):
-		pass
-
-	def test_empty_append(self):
-
 		data = tablib.Dataset()
 
-		new_row = (1,2,3)
+		self.headers = ('first_name', 'last_name', 'gpa')
+		self.john = ('John', 'Adams', 90)
+		self.george = ('George', 'Washington', 67)
+		self.tom = ('Thomas', 'Jefferson', 50)
+
+		self.founders = tablib.Dataset(headers=self.headers)
+		self.founders.append(self.john)
+		self.founders.append(self.george)
+		self.founders.append(self.tom)
+
+
+	def tearDown(self):
+		"""Teardown."""
+		pass
+
+
+	def test_empty_append(self):
+		"""Verify append() correctly adds tuple with no headers."""
+		new_row = (1, 2, 3)
 		data.append(new_row)
 
+		# Verify width/data
 		self.assertTrue(data.width == len(new_row))
+		self.assertTrue(data[0] == new_row)
 
 
 	def test_empty_append_with_headers(self):
-
+		"""Verify append() correctly detects mismatch of number of
-		data = tablib.Dataset()
+		headers and data.
-
+		"""
 		data.headers = ['first', 'second']
-		new_row = (1,2,3,4)
+		new_row = (1, 2, 3, 4)
 
 		self.assertRaises(tablib.InvalidDimensions, data.append, new_row)
 
-	# def test_adding_header with (self):
+
+	def test_add_column(self):
+		"""Verify adding column works with/without headers."""
+
+		data.append(['kenneth'])
+		data.append(['bessie'])
+
+		new_col = ['reitz', 'monke']
+
+		data.append(col=new_col)
+
+		self.assertEquals(data[0], ('kenneth', 'reitz'))
+		self.assertEquals(data.width, 2)
+
+		# With Headers
+		data.headers = ('fname', 'lname')
+		new_col = ['age', 21, 22]
+		data.append(col=new_col)
+
+		self.assertEquals(data[new_col[0]], new_col[1:])
 
 
+	def test_add_column_no_data_no_headers(self):
+		"""Verify adding new column with no headers."""
+
+		new_col = ('reitz', 'monke')
+
+		data.append(col=new_col)
+
+		self.assertEquals(data[0], tuple([new_col[0]]))
+		self.assertEquals(data.width, 1)
+		self.assertEquals(data.height, len(new_col))
+
+
+	def test_add_column_no_data_with_headers(self):
+		"""Verify adding new column with headers."""
+
+		data.headers = ('first', 'last')
+
+		new_col = ('age',)
+		data.append(col=new_col)
+
+		self.assertEquals(len(data.headers), 3)
+		self.assertEquals(data.width, 3)
+
+		new_col = ('foo', 'bar')
+
+		self.assertRaises(tablib.InvalidDimensions, data.append, col=new_col)
+
+
+	def test_header_slicing(self):
+		"""Verify slicing by headers."""
+
+		self.assertEqual(self.founders['first_name'],
+						[self.john[0], self.george[0], self.tom[0]])
+		self.assertEqual(self.founders['last_name'],
+						[self.john[1], self.george[1], self.tom[1]])
+		self.assertEqual(self.founders['gpa'],
+						[self.john[2], self.george[2], self.tom[2]])
+
+
+	def test_data_slicing(self):
+		"""Verify slicing by data."""
+
+		# Slice individual rows
+		self.assertEqual(self.founders[0], self.john)
+		self.assertEqual(self.founders[:1], [self.john])
+		self.assertEqual(self.founders[1:2], [self.george])
+		self.assertEqual(self.founders[-1], self.tom)
+		self.assertEqual(self.founders[3:], [])
+
+		# Slice multiple rows
+		self.assertEqual(self.founders[:], [self.john, self.george, self.tom])
+		self.assertEqual(self.founders[0:2], [self.john, self.george])
+		self.assertEqual(self.founders[1:3], [self.george, self.tom])
+		self.assertEqual(self.founders[2:], [self.tom])
+
+
+	def test_delete(self):
+		"""Verify deleting from dataset works."""
+
+		# Delete from front of object
+		del self.founders[0]
+		self.assertEqual(self.founders[:], [self.george, self.tom])
+
+		# Verify dimensions, width should NOT change
+		self.assertEqual(self.founders.height, 2)
+		self.assertEqual(self.founders.width, 3)
+
+		# Delete from back of object
+		del self.founders[1]
+		self.assertEqual(self.founders[:], [self.george])
+
+		# Verify dimensions, width should NOT change
+		self.assertEqual(self.founders.height, 1)
+		self.assertEqual(self.founders.width, 3)
+
+		# Delete from invalid index
+		self.assertRaises(IndexError, self.founders.__delitem__, 3)
+
+
+	def test_csv_export(self):
+		"""Verify exporting dataset object as CSV."""
+
+		# Build up the csv string with headers first, followed by each row
+		csv = ''
+		for col in self.headers:
+			csv += col + ','
+
+		csv = csv.strip(',') + '\r\n'
+
+		for founder in self.founders:
+			for col in founder:
+				csv += str(col) + ','
+			csv = csv.strip(',') + '\r\n'
+
+		self.assertEqual(csv, self.founders.csv())
+
+
+	def test_unicode_append(self):
+		"""Passes in a single unicode charecter and exports."""
+
+		new_row = ('å', 'é')
+		data.append(new_row)
+
+		data.json()
+		data.yaml()
+		data.csv()
+		data.xls()
 
 	
 if __name__ == '__main__':