BUG: Segmentation Fault in MacOS installed by Pip

[Fazatholomew]

Pandas version checks

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

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

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

from pandas import DataFrame

Issue Description

I tried to import pandas but it's straight to segmentation fault. Numpy can be imported separately. Installed by pip in venv and no other pandas installation exists.

Below is the backtrace from lldb.

(lldb) target create "python"
Current executable set to 'python' (x86_64).
(lldb) settings set -- target.run-args  "-c" "import pandas as pd"
(lldb) r
Process 57543 launched: '/.venv/bin/python' (x86_64)
Process 57543 stopped
* thread #2, stop reason = exec
    frame #0: 0x0000000100011000 dyld`_dyld_start
dyld`_dyld_start:
->  0x100011000 <+0>: popq   %rdi
    0x100011001 <+1>: pushq  $0x0
    0x100011003 <+3>: movq   %rsp, %rbp
    0x100011006 <+6>: andq   $-0x10, %rsp
Target 0: (Python) stopped.
(lldb) bt
* thread #2, stop reason = exec
  * frame #0: 0x0000000100011000 dyld`_dyld_start
(lldb)

System Version: macOS 11.7.6 (20G1231)
Kernel Version: Darwin 20.6.0
Model Name: MacBook Pro
Model Identifier: MacBookPro11,1
Python: 3.12.2

Expected Behavior

No segmentation fault

Installed Versions

I can't even import it but pip said it's 2.2.2

[jbrockmendel]

There isn't enough here to reproduce. Can you confirm that importing numpy, pytz, dateutil.tz, dateutil.parser work OK?

[Fazatholomew]

I don't think I am using lldb correctly. I tried using GDB, it kept giving me .venv/bin/python": not in executable format: file format not recognized. To answer your question, all of those libraries are imported fine. Until pandas:

python                                                            [3:05:10]
Python 3.12.2 (v3.12.2:6abddd9f6a, Feb  6 2024, 17:02:06) [Clang 13.0.0 (clang-1300.0.29.30)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import numpy
>>> import pytz
>>> import dateutil.tz
>>> import dateutil.parser
>>> import pandas
[1]    55969 segmentation fault  python

[rhshadrach]

When you do pip install pandas, you should see precisely which wheel being used to install. Can you post that information? E.g.

pandas-2.2.2-cp312-cp312-macosx_10_9_x86_64.whl

[Fazatholomew]

Actually, I just created a new venv and did the installation again and it works fine! However, it turns out, when I install pyarrow that it has that error. Should I take this to pyarrow repo? Here's the --verbose output

DEBUG Using Python 3.12.2 environment at .venv/bin/python3
DEBUG Acquired lock for `.venv`
DEBUG At least one requirement is not satisfied: pyarrow
DEBUG Using request timeout of 30s
DEBUG Solving with installed Python version: 3.12.2
DEBUG Adding direct dependency: pandas*
DEBUG Adding direct dependency: pyarrow*
DEBUG Found fresh response for: https://pypi.org/simple/pyarrow/
DEBUG Found fresh response for: https://files.pythonhosted.org/packages/d4/62/ce6ac1275a432b4a27c55fe96c58147f111d8ba1ad800a112d31859fae2f/pyarrow-17.0.0-cp312-cp312-macosx_10_15_x86_64.whl.metadata
DEBUG Found stale response for: https://pypi.org/simple/pandas/
DEBUG Sending revalidation request for: https://pypi.org/simple/pandas/
DEBUG Found not-modified response for: https://pypi.org/simple/pandas/
DEBUG Searching for a compatible version of pandas (*)
DEBUG Selecting: pandas==2.2.2 [compatible] (pandas-2.2.2-cp312-cp312-macosx_10_9_x86_64.whl)
DEBUG Found fresh response for: https://files.pythonhosted.org/packages/dd/49/de869130028fb8d90e25da3b7d8fb13e40f5afa4c4af1781583eb1ff3839/pandas-2.2.2-cp312-cp312-macosx_10_9_x86_64.whl.metadata
DEBUG Adding transitive dependency for pandas==2.2.2: numpy{python_version >= '3.12'}>=1.26.0
DEBUG Adding transitive dependency for pandas==2.2.2: python-dateutil>=2.8.2
DEBUG Adding transitive dependency for pandas==2.2.2: pytz>=2020.1
DEBUG Adding transitive dependency for pandas==2.2.2: tzdata>=2022.7
DEBUG Searching for a compatible version of pyarrow (*)
DEBUG Selecting: pyarrow==17.0.0 [compatible] (pyarrow-17.0.0-cp312-cp312-macosx_10_15_x86_64.whl)
DEBUG Adding transitive dependency for pyarrow==17.0.0: numpy>=1.16.6
DEBUG Found stale response for: https://pypi.org/simple/python-dateutil/
DEBUG Sending revalidation request for: https://pypi.org/simple/python-dateutil/
DEBUG Found stale response for: https://pypi.org/simple/pytz/
DEBUG Sending revalidation request for: https://pypi.org/simple/pytz/
DEBUG Found stale response for: https://pypi.org/simple/tzdata/
DEBUG Sending revalidation request for: https://pypi.org/simple/tzdata/
DEBUG Found stale response for: https://pypi.org/simple/numpy/
DEBUG Sending revalidation request for: https://pypi.org/simple/numpy/
DEBUG Found not-modified response for: https://pypi.org/simple/python-dateutil/
DEBUG Found fresh response for: https://files.pythonhosted.org/packages/ec/57/56b9bcc3c9c6a792fcbaf139543cee77261f3651ca9da0c93f5c1221264b/python_dateutil-2.9.0.post0-py2.py3-none-any.whl.metadata
DEBUG Found not-modified response for: https://pypi.org/simple/pytz/
DEBUG Found fresh response for: https://files.pythonhosted.org/packages/9c/3d/a121f284241f08268b21359bd425f7d4825cffc5ac5cd0e1b3d82ffd2b10/pytz-2024.1-py2.py3-none-any.whl.metadata
DEBUG Found not-modified response for: https://pypi.org/simple/numpy/
DEBUG Found not-modified response for: https://pypi.org/simple/tzdata/
DEBUG Found fresh response for: https://files.pythonhosted.org/packages/65/58/f9c9e6be752e9fcb8b6a0ee9fb87e6e7a1f6bcab2cdc73f02bb7ba91ada0/tzdata-2024.1-py2.py3-none-any.whl.metadata
DEBUG Searching for a compatible version of numpy{python_version >= '3.12'} (>=1.26.0)
DEBUG Selecting: numpy==2.0.1 [compatible] (numpy-2.0.1-cp312-cp312-macosx_10_9_x86_64.whl)
DEBUG Adding transitive dependency for numpy==2.0.1: numpy==2.0.1
DEBUG Adding transitive dependency for numpy==2.0.1: numpy{python_version >= '3.12'}==2.0.1
DEBUG Searching for a compatible version of numpy (==2.0.1)
DEBUG Selecting: numpy==2.0.1 [compatible] (numpy-2.0.1-cp312-cp312-macosx_10_9_x86_64.whl)
DEBUG Found fresh response for: https://files.pythonhosted.org/packages/64/1c/401489a7e92c30db413362756c313b9353fb47565015986c55582593e2ae/numpy-2.0.1-cp312-cp312-macosx_10_9_x86_64.whl.metadata
DEBUG Searching for a compatible version of numpy{python_version >= '3.12'} (==2.0.1)
DEBUG Selecting: numpy==2.0.1 [compatible] (numpy-2.0.1-cp312-cp312-macosx_10_9_x86_64.whl)
DEBUG Searching for a compatible version of python-dateutil (>=2.8.2)
DEBUG Selecting: python-dateutil==2.9.0.post0 [compatible] (python_dateutil-2.9.0.post0-py2.py3-none-any.whl)
DEBUG Adding transitive dependency for python-dateutil==2.9.0.post0: six>=1.5
DEBUG Searching for a compatible version of pytz (>=2020.1)
DEBUG Selecting: pytz==2024.1 [compatible] (pytz-2024.1-py2.py3-none-any.whl)
DEBUG Searching for a compatible version of tzdata (>=2022.7)
DEBUG Selecting: tzdata==2024.1 [compatible] (tzdata-2024.1-py2.py3-none-any.whl)
DEBUG Found stale response for: https://pypi.org/simple/six/
DEBUG Sending revalidation request for: https://pypi.org/simple/six/
DEBUG Found not-modified response for: https://pypi.org/simple/six/
DEBUG Searching for a compatible version of six (>=1.5)
DEBUG Selecting: six==1.16.0 [compatible] (six-1.16.0-py2.py3-none-any.whl)
DEBUG Found fresh response for: https://files.pythonhosted.org/packages/d9/5a/e7c31adbe875f2abbb91bd84cf2dc52d792b5a01506781dbcf25c91daf11/six-1.16.0-py2.py3-none-any.whl.metadata
DEBUG Tried 7 versions: numpy 1, pandas 1, pyarrow 1, python-dateutil 1, pytz 1, six 1, tzdata 1
DEBUG Split specific environment resolution took 0.306s
Resolved 7 packages in 311ms
DEBUG Requirement already cached: numpy==2.0.1
DEBUG Requirement already cached: pandas==2.2.2
DEBUG Requirement already cached: pyarrow==17.0.0
DEBUG Requirement already cached: python-dateutil==2.9.0.post0
DEBUG Requirement already cached: pytz==2024.1
DEBUG Requirement already cached: six==1.16.0
DEBUG Requirement already cached: tzdata==2024.1
DEBUG Preserving seed package: pip==24.0

[Fazatholomew]

This is a known pyarrow issue for MacOS 11. Pyarrow >16 doesn't support MacOS 11 anymore. So, in order to use pyarrow in MacOS 11 is to use pyarrow==16.0.0.