CSV dialect is not auto-detected
|Reported by:||anonymous||Owned by:||athomas|
Contrary to the comment in #324, my environment did not autodetect the format of the CSV data. I'm going to assume this is a problem in my version of the csv module (i.e., out of scope of this macro), but it might affect others.
- trac 0.9.6
- Debian unstable
- python2.3 (2.3.5-15)
- csv.__version__ = 1.0
My fix was to capture txt.encode('ascii', 'replace') into txt and then pass sniffer.sniff(txt) into reader as was removed in changeset .