Message353912
This appears to be somehow caused by running Python 3.7.4 from an active virtualenv (which I'd forgotten I had active when I did the test).
Regardless, I'd have expected that the command would either correctly create a venv containing pip, or fail to create the venv at all. Creating an incomplete venv is extremely confusing for the user, if nothing else.
It looks like this is *not* a regression - it's been like this since 3.6.8 at least.
Actually, the created venv looks pretty broken to me - it has the *parent* environment in `sys.path`, but does not have its *own* environment there:
>.\ve-368a\Scripts\python.exe
Python 3.6.8 (tags/v3.6.8:3c6b436a57, Dec 24 2018, 00:16:47) [MSC v.1916 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import sys
>>> sys.path
['', 'C:\\Work\\Scratch\\vv\\ve-368a\\Scripts\\python36.zip', 'C:\\Users\\UK03306\\.virtualenvs\\f1980c81dac32d4\\DLLs', 'C:\\Users\\UK03306\\.virtualenvs\\f1980c81dac32d4\\lib', 'C:\\Users\\UK03306\\.virtualenvs\\f1980c81dac32d4\\Scripts', 'C:\\Utils\\PythonVersions\\Python-3.6.8\\Lib', 'C:\\Utils\\PythonVersions\\Python-3.6.8\\DLLs', 'C:\\Users\\UK03306\\.virtualenvs\\f1980c81dac32d4', 'C:\\Users\\UK03306\\.virtualenvs\\f1980c81dac32d4\\lib\\site-packages']
Note the lack of ve-368a directories, and the presence of a lot of directories from .virtualenvs (the parent environment). |
|
Date |
User |
Action |
Args |
2019-10-04 09:37:28 | paul.moore | set | recipients:
+ paul.moore, steve.dower |
2019-10-04 09:37:27 | paul.moore | set | messageid: <1570181847.97.0.421572324708.issue38369@roundup.psfhosted.org> |
2019-10-04 09:37:27 | paul.moore | link | issue38369 messages |
2019-10-04 09:37:27 | paul.moore | create | |
|