-
-
Notifications
You must be signed in to change notification settings - Fork 528
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Unbreak the build when pyenv is present #29285
Comments
comment:1
Who's doing something with pyenv? I don't think we need to provide work-arounds for every conceivable user environment misconfiguration. |
comment:2
Replying to @embray:
https://dblp.org/pers/k/Kruppa:Alexander.html
|
comment:3
Pyenv creates major issues with packages sometimes, I can imagine what Sagemath might be going about. My easiest fix would be symlinking the two paths together which allows communication, between your bin and other paths, it did work for me when I was trying simple builds, of course now I use Sagemath as an application separately. Pyenv is popular with MacOs users, so this will definitely need a fix |
comment:4
In fact, the breaking of the path isn't something restricted to Sage, it existed with pip too, where pip was not able to define the paths correctly for the packages to go in |
This comment has been minimized.
This comment has been minimized.
comment:6
Replying to @embray:
I do. Because I tried installing a standalone cygdb on Arch Linux for 3 days and only succeeded when I used pyenv, because I wanted two different python versions on my machine: A normal one and one compiled with --with-pydebug. Achieving this without pyenv is very hard (on Arch Linux).
If it is too hard or not possible to make sagemath work with pyenv that is ok. But there should be a good error message, because no one looks at the message:
And thinks: "Ok, it looks like I need to uninstall pyenv." An error message similar to this: In General, I think it is really important that misconfigurations result in nice error messages. Note: My issue can be found here: #29461 |
comment:9
I'm checking this with the up to date pyenv, and it seems that the issue with the PATH is no longer. Currently building latest beta with python 3.9-dev from pyenv on Debian stable, will report the results. |
comment:10
I was able to successfully install Sage using python 3.9 from pyenv (installed following Perhaps one should check this on Homebrew, but it seems it's OK. |
comment:11
Can be fixed as worksforme, that is. |
comment:12
Let us close this if nobody objects. |
Reviewer: Samuel Lelièvre |
comment:14
New issues with pyenv reported in #32749. |
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
comment:19
Test on macOS with
( pip's self-installation gets tricked into installing in the wrong place:
|
comment:20
|
This comment has been minimized.
This comment has been minimized.
Commit: |
Author: Matthias Koeppe |
New commits:
|
Changed reviewer from Samuel Lelièvre to none |
comment:25
Let's get this into 9.5 please |
Changed reviewer from https://github.com/mkoeppe/sage/runs/4682289891?check_suite_focus=true to John Palmieri |
comment:28
Works for me on OS X with homebrew's pyenv: without this branch, sage-bootstrap runs the "wrong" python3, while with this branch, it runs the right python3. |
comment:29
Thank you! |
Changed branch from u/mkoeppe/unbreak_the_build_when_pyenv_is_present to |
The build fails when pyenv is present, as reported in report and also in #32749/#33103.
We fix it by rejecting the pyenv shims for
sage-bootstrap-python
.Tickets:
SAGE_ROOT/tox.ini
: Add variants that provision a system python usingpyenv
CC: @embray @mkoeppe @slel @tobiasdiez @jhpalmieri
Component: build: configure
Author: Matthias Koeppe
Branch/Commit:
376932f
Reviewer: John Palmieri
Issue created by migration from https://trac.sagemath.org/ticket/29285
The text was updated successfully, but these errors were encountered: