You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have the following dep listed in my py2 project on a LINUX machine:
pandas==0.17.1
Since Pypi has no linux wheels until pandas v0.18.1 ( https://pypi.org/simple/pandas/ ),
pip install -r
Will download and attempt to build pandas from source.
I have numpy 1.16.4 installed, but get this error:
run_setup
...
File "/tmp/easy_install-cXiZ1Y/numpy-1.17.0rc1/setup.py", line 31, in
from setuptools import setup, Command
RuntimeError: Python version >= 3.5 required.
I believe that the pandas build is pulling in numpy itself in order to build numpy/pandas binaries from source (correct?), rather than use the installed numpy version.
HOWEVER, it pulls in the latest numpy, which ASOF is numpy-1.17.0r, where 1.17 is the first numpy to drop support for py27.
If my assumptions are correct, is it possible to set a numpy "max_version" to prevent the py27 incompatible numpys from being pulled in for older pandas being built on py27?
The text was updated successfully, but these errors were encountered:
Problem description
I have the following dep listed in my py2 project on a LINUX machine:
Since Pypi has no linux wheels until pandas v0.18.1 ( https://pypi.org/simple/pandas/ ),
pip install -r
Will download and attempt to build pandas from source.
I have numpy 1.16.4 installed, but get this error:
I believe that the pandas build is pulling in numpy itself in order to build numpy/pandas binaries from source (correct?), rather than use the installed numpy version.
HOWEVER, it pulls in the latest numpy, which ASOF is numpy-1.17.0r, where 1.17 is the first numpy to drop support for py27.
If my assumptions are correct, is it possible to set a numpy "max_version" to prevent the py27 incompatible numpys from being pulled in for older pandas being built on py27?
The text was updated successfully, but these errors were encountered: