Python2's stdlib csv module is nice, but it doesn't support unicode. This module is a drop-in replacement which *does*.
Project Links
Meta
Author: Jeremy Dunck
Classifiers
Development Status
- 5 - Production/Stable
Intended Audience
- Developers
License
- OSI Approved :: BSD License
Natural Language
- English
Programming Language
- Python :: 2.6
- Python :: 2.7
- Python :: 3.3
- Python :: 3.4
- Python :: 3.5
- Python :: Implementation :: PyPy
- Python :: Implementation :: CPython
The unicodecsv is a drop-in replacement for Python 2.7’s csv module which supports unicode strings without a hassle. Supported versions are python 2.7, 3.3, 3.4, 3.5, and pypy 2.4.0.
More fully
Python 2’s csv module doesn’t easily deal with unicode strings, leading to the dreaded “‘ascii’ codec can’t encode characters in position …” exception.
You can work around it by encoding everything just before calling write (or just after read), but why not add support to the serializer?
>>> import unicodecsv as csv
>>> from io import BytesIO
>>> f = BytesIO()
>>> w = csv.writer(f, encoding='utf-8')
>>> _ = w.writerow((u'é', u'ñ'))
>>> _ = f.seek(0)
>>> r = csv.reader(f, encoding='utf-8')
>>> next(r) == [u'é', u'ñ']
TrueNote that unicodecsv expects a bytestream, not unicode – so there’s no need to use codecs.open or similar wrappers. Plain open(…, ‘rb’) will do.
(Version 0.14.0 dropped support for python 2.6, but 0.14.1 added it back. See c0b7655248c4249 for the mistaken, breaking change.)
0.14.1
Sep 22, 2015
0.14.0
Sep 21, 2015
0.13.0
May 27, 2015
0.12.0
Apr 15, 2015
0.11.2
Apr 13, 2015
0.11.1
Apr 12, 2015
0.11.0
Mar 11, 2015
0.9.4
Jan 29, 2013
0.9.3
Jan 25, 2013
0.9.0
Dec 02, 2011
0.8.0
Nov 27, 2011
Files in release
No dependencies