在熊猫read_csv中使用utf-8记录分隔符

Question

I have a delimited file where the separation character is the NOT character (¬) and I am unable to parse it using pandas - see below, the columns are not properly split.

test = pd.read_csv("file.csv", sep="¬", encoding="latin-1")
test.head(1)
0       1231�XXX7791�BBB9991�22999KKKK... 
test.shape
Out[128]: (7001001, 1)

I am using ipython 3.2.0, pandas 0.16.2, 2.7.10.final.0 on OS X Yosemite.

Answer 1

import pandas as pd

df = pd.read_csv('data.csv', sep='\u00AC', encoding ='utf-8', header=None, engine='python')

print(df)

The previous code will give me this, which is what you wanted. You just needed to call the correct UTF-8 encoding as a sep

      0        1        2          3
0  1231  XXX7791  BBB9991  22999KKKK

You need engine=python because by default pandas uses engine=c which does not support regex seperators.

From IPython