homepage

Getting some extremely strange behavior when attempting to parse a fairly standard CSV in Python 2.6.6.
I've tried a whole different mixture of dialects, quoting options, line terminators etc, and none seem to get a happy ending.
Spent about 2 hours banging my head against a brick wall on this, and struggling to see how the CSV libs could be so fundamentally broken, given that I couldn't find any other related bugs.
I have attempted to parse the following CSV data:
"First","Middle","Last","Nickname","Email","Category"
"Moe","","Howard","Moe","moe@3stooges.com","actor"
"Jerome","Lester","Howard","Curly","curly@3stooges.com","actor"
"Larry","","Fine","Larry","larry@3stooges.com","musician"
"Jerome","","Besser","Joe","joe@3stooges.com","actor"
"Joe","","DeRita","CurlyJoe","curlyjoe@3stooges.com","actor"
"Shemp","","Howard","Shemp","shemp@3stooges.com","actor"
The code used to parse was this:
datx = open("data.txt", "rb").read()
rows = csv.reader( datx , dialect="wat")
for row in rows:
 print x
The output given is this:
['First']
['', '']
['Middle']
['', '']
['Last']
['', '']
['Nickname']
['', '']
['Email']
['', '']
['Category']
[]
['Moe']
['', '']
['']
['', '']
['Howard']
['', '']
['Moe']
['', '']
['moe@3stooges.com']
['', '']
['actor']
[]
['Jerome']
['', '']
['Lester']
['', '']
['Howard']
['', '']
['Curly']
['', '']
['curly@3stooges.com']
['', '']
['actor']
[]
['Larry']
['', '']
['']
['', '']
['Fine']
['', '']
['Larry']
['', '']
['larry@3stooges.com']
['', '']
['musician']
[]
['Jerome']
['', '']
['']
['', '']
['Besser']
['', '']
['Joe']
['', '']
['joe@3stooges.com']
['', '']
['actor']
[]
['Joe']
['', '']
['']
['', '']
['DeRita']
['', '']
['CurlyJoe']
['', '']
['curlyjoe@3stooges.com']
['', '']
['actor']
[]
['Shemp']
['', '']
['']
['', '']
['Howard']
['', '']
['Shemp']
['', '']
['shemp@3stooges.com']
['', '']
['actor']
[]

Sorry, accidently pasted the wrong code snippet previously. The correct code snippet is:
datx = open("data.txt", "rb").read()
rows = csv.reader( datx )
for row in rows:
 print x

This bug also seems to be showing in 2.7.3

Okay, just found the reason for this.. It's because I was putting a .read() on the file descriptor..
I really think that the CSVReader should raise an assertion in the event that it is passed an object which has no iterator, or if it is given a string, as this is a fairly easy mistake to make.

It already produces a TypeError with a specific message if the input is
not iterable. You seem to be using a homegrown dialect; with the
conventional list (csv.reader("the quick brown fox")) you very quickly
see that you're iterating over a string.

We don't generally do that kind of type checking.

@david Gotcha - I had a feeling that would be the case. Thank you for the quick replies anyway guys! Hopefully this will help others in the future :)