Unconsistent (random) behaviour of pd.concat with different indexes · Issue #17344 · pandas-dev/pandas (original) (raw)

Code Sample, a copy-pastable example if possible

The following code

import pandas as pd

dfs_sq = []

dfs_sq.append(pd.DataFrame(index=[0, 'sess'], columns=range(1,3))) for i in range(5): dfs_sq.append(pd.DataFrame(index=[0, 1, 'sess'], columns=range(1,3)))

df_sq = pd.concat(dfs_sq, axis=1)

assert df_sq.index[1] == 'sess', df_sq.index

... saved as dependable.py, succeeds if called as PYTHONHASHSEED=40 python3 dependable.py and fails if called as PYTHONHASHSEED=41 python3 dependable.py:

Traceback (most recent call last): File "dependable.py", line 11, in assert df_sq.index[1] == 'sess', df_sq.index AssertionError: Index([0, 1, 'sess'], dtype='object')

Problem description

This is not very nice.

Expected Output

The same - which one is not particularly important.

Output of pd.show_versions()

INSTALLED VERSIONS

commit: None
python: 3.5.3.final.0
python-bits: 64
OS: Linux
OS-release: 4.9.0-3-amd64
machine: x86_64
processor:
byteorder: little
LC_ALL: None
LANG: it_IT.UTF-8
LOCALE: it_IT.UTF-8

pandas: 0.21.0.dev+389.g276f3089a
pytest: 3.0.6
pip: 9.0.1
setuptools: None
Cython: 0.25.2
numpy: 1.12.1
scipy: 0.19.0
pyarrow: None
xarray: None
IPython: 5.1.0.dev
sphinx: 1.5.6
patsy: 0.4.1
dateutil: 2.6.0
pytz: 2017.2
blosc: None
bottleneck: 1.2.1
tables: 3.3.0
numexpr: 2.6.1
feather: 0.3.1
matplotlib: 2.0.2
openpyxl: None
xlrd: 1.0.0
xlwt: 1.1.2
xlsxwriter: 0.9.6
lxml: None
bs4: 4.5.3
html5lib: 0.999999999
sqlalchemy: 1.0.15
pymysql: None
psycopg2: None
jinja2: 2.9.6
s3fs: None
fastparquet: None
pandas_gbq: None
pandas_datareader: 0.2.1