BUG: read_excel leads to segfault

[tharwan]

• [x ] I have checked that this issue has not already been reported.

• [ x] I have confirmed this bug exists on the latest version of pandas.

• (optional) I have confirmed this bug exists on the master branch of pandas.

---

Note: Please read this guide detailing how to provide the necessary information for us to reproduce your bug.

Code Sample, a copy-pastable example

from pathlib import Path
import pandas as pd

filename = 'file.xlsm'
template_df = pd.read_excel(filename, sheet_name="Muster", usecols="A:F", skiprows=1)
template_df.index = template_df["Beschreibung"]

path = 'file2.xlsx'
df = pd.read_excel(path, sheet_name="Januar", usecols="A:F", skiprows=1)
df.index = df["Beschreibung"]

template_df.loc['Fee EPEX ID Kundenhandel', "Short/Kauf/NEG"] = 0
template_df.update(df)
template_df.loc["Direktvermarktung Vertragspreis", "Long/Verkauf/POS"]

Problem description

First of all, I know that the example is not a proper minimal snippet. I also can not provide the excel files in question. However I have no idea how to debug this issue any further.

If I run the script above with python -X faulthandler I get:

Current thread 0x00007f76b30a3280 (most recent call first):
  File ".../.venv/lib/python3.8/site-packages/pandas/core/frame.py", line 3133 in _get_value
  File ".../.venv/lib/python3.8/site-packages/pandas/core/indexing.py", line 888 in __getitem__
  File ".../debug.py", line 22 in <module>
fish: “python -X faulthandler .../de…” terminated by signal SIGSEGV (Address boundary error)

From what I can deduce there is something wrong with the template_df DataFrame. I was trying to generate an equivivalent copy of it without copying the memory so I ran it through

tempalte_df = pd.Dataframe.from_dict(template_df.to_dict()). 

This does indeed seem to help. However by accident I discovered just doing

d = template_df.to_dict() 

also fixes it. As if to_dict() would somehow alter the DataFrame?

So I might be completely on the wrong track. Any help on how to drill down on this would be appreciated.

Expected Output

No segfault :-)

Output of pd.show_versions()

INSTALLED VERSIONS

commit : f2c8480
python : 3.8.2.final.0
python-bits : 64
OS : Linux
OS-release : 5.4.0-65-generic
Version : #73-Ubuntu SMP Mon Jan 18 17:25:17 UTC 2021
machine : x86_64
processor : x86_64
byteorder : little
LC_ALL : None
LANG : en_US.UTF-8
LOCALE : en_US.UTF-8

pandas : 1.2.3
numpy : 1.20.1
pytz : 2021.1
dateutil : 2.8.1
pip : 21.0.1
setuptools : 52.0.0
Cython : None
pytest : None
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : 1.3.7
lxml.etree : 4.6.2
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : 2.11.3
IPython : 7.21.0
pandas_datareader: None
bs4 : None
bottleneck : None
fsspec : None
fastparquet : None
gcsfs : None
matplotlib : 3.3.4
numexpr : None
odfpy : None
openpyxl : 3.0.6
pandas_gbq : None
pyarrow : None
pyxlsb : None
s3fs : None
scipy : None
sqlalchemy : 1.3.23
tables : None
tabulate : None
xarray : None
xlrd : 1.2.0
xlwt : None
numba : None

[mzeitlin11]

Thanks for the report @tharwan. My best guess upon a quick glance is that the issue here is related to the line
df.index = df["Beschreibung"] (see #34364 for something similar). The same workaround should hopefully work here (using set_index instead of directly assigning). Since the index is immutable, direct assignment with a mutable object should generally be avoided (but clearly we should handle it more gracefully than a segfault).

[mzeitlin11]

@tharwan going to close in favor of #34364, but please reopen if you think it is a different issue.