same .tsv file, get different data-frame structure using engine 'python' and 'c'

[Jane-Eyre]

In my Mac, I have a Tab-Separated values file with encoding UTF-8, and the version of pandas is
0.24.2.
74872_zh_CN_UI.txt

when I use read_csv function with engine 'python' like this:
b = pd.read_csv("/Users/GHIBLI/Documents/vmware-L10n/bert/zh_CN/74872_zh_CN_UI.tsv", engine="python", delimiter="\t")
print(b.shape)
I got (8, 1)

if with default engine:
b = pd.read_csv("/Users/GHIBLI/Documents/vmware-L10n/bert/zh_CN/74872_zh_CN_UI.tsv", delimiter="\t")
print(b.shape)
I got (8,22)

In contrast to 'C' engine, this 'python' engine seems that is not as simple as just 'feature-complete' I think.

[Liam3851]

I cannot reproduce using pandas 0.24.2 on either Ubuntu or Windows -- both show the shape as (8, 22) using both engines. It could be a Mac-specific issue, though it would be odd that the Python-engine implementation would behave differently based on the OS.

[WillAyd]

I can reproduce this on master. If you would like to take a deeper look and see what's going on would certainly appreciate it!

[Liam3851]

@WillAyd Are you using a Mac? I can't reproduce on master using Windows/python 3.7.3 (I suppose perhaps this could also be a python version difference, rather than an OS difference)

In [1]: import pandas as pd

In [2]: pd.__version__
Out[2]: '0.25.0.dev0+615.g998a0deea'

In [3]: pd.read_csv('74872_zh_CN_UI.txt', delimiter='\t', engine='python').shape
Out[3]: (8, 22)

[WillAyd]

Yes macOS on master

…

Sent from my iPhone

On May 28, 2019, at 8:37 AM, David Krych ***@***.***> wrote: @WillAyd Are you using a Mac? I can't reproduce on master using Windows/python 3.7.3 (I suppose perhaps this could also be a python version difference, rather than an OS difference) In [1]: import pandas as pd In [2]: pd.__version__ Out[2]: '0.25.0.dev0+615.g998a0deea' In [3]: pd.read_csv('74872_zh_CN_UI.txt', delimiter='\t', engine='python').shape Out[3]: (8, 22) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[luckydenis]

I was able to play on linux 19.04 Python 3.7 Pandas 0.24.2

[luckydenis]

Good afternoon, I found where in what a mistake. But have not yet figured out how best to fix it. Do you mind if I take this task?