在pandas read_csv中转义引号

Question

I am unable to create a dataframe which has escaped quotes when using read_csv .
(Note: R's read.csv works as expected.)

My code:

import pandas as pd
pd.read_csv('data.csv')
#error!
CParserError: Error tokenizing data. C error: Expected 2 fields in line 4, saw 3

data.csv

SEARCH_TERM,ACTUAL_URL
"bra tv bord","http://www.ikea.com/se/sv/catalog/categories/departments/living_room/10475/?se%7cps%7cnonbranded%7cvardagsrum%7cgoogle%7ctv_bord"
"tv på hjul","http://www.ikea.com/se/sv/catalog/categories/departments/living_room/10475/?se%7cps%7cnonbranded%7cvardagsrum%7cgoogle%7ctv_bord"
"SLAGBORD, \"Bergslagen\", IKEA:s 1700-tals serie","http://www.ikea.com/se/sv/catalog/categories/departments/living_room/10475/?se%7cps%7cnonbranded%7cvardagsrum%7cgoogle%7ctv_bord"

How can I read this csv and avoid this error?

My guess is that pandas is using some regular expressions which cannot handle the ambiguity and trips on the third row, or more specifically: \\"Bergslagen\\" .

Answer 1

It does work, but you have to indicate the escape character for the embedded quotes:

In [1]: data = '''SEARCH_TERM,ACTUAL_URL
"bra tv bord","http://www.ikea.com/se/sv/catalog/categories/departments/living_room/10475/?se%7cps%7cnonbranded%7cvardagsrum%7cgoogle%7ctv_bord"
"tv p\xc3\xa5 hjul","http://www.ikea.com/se/sv/catalog/categories/departments/living_room/10475/?se%7cps%7cnonbranded%7cvardagsrum%7cgoogle%7ctv_bord"
"SLAGBORD, \\"Bergslagen\\", IKEA:s 1700-tals serie","http://www.ikea.com/se/sv/catalog/categories/departments/living_room/10475/?se%7cps%7cnonbranded%7cvardagsrum%7cgoogle%7ctv_bord"'''

In [2]: df = read_csv(StringIO(data), escapechar='\\', encoding='utf-8')

In [3]: df
Out[3]: 
                                      SEARCH_TERM                                         ACTUAL_URL
0                                     bra tv bord  http://www.ikea.com/se/sv/catalog/categories/d...
1                                      tv på hjul  http://www.ikea.com/se/sv/catalog/categories/d...
2  SLAGBORD, "Bergslagen", IKEA:s 1700-tals serie  http://www.ikea.com/se/sv/catalog/categories/d...

see this gist .