GNU bug report logs -
#46345
27.1; Issue with `run-python' when `exec-path' is set buffer-locally
Previous Next
To reply to this bug, email your comments to 46345 AT debbugs.gnu.org.
Toggle the display of automated, internal messages from the tracker.
Report forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#46345
; Package
emacs
.
(Sat, 06 Feb 2021 13:05:01 GMT)
Full text and
rfc822 format available.
Acknowledgement sent
to
Augusto Stoffel <arstoffel <at> gmail.com>
:
New bug report received and forwarded. Copy sent to
bug-gnu-emacs <at> gnu.org
.
(Sat, 06 Feb 2021 13:05:01 GMT)
Full text and
rfc822 format available.
Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):
It is reasonable to set `exec-path' locally in Python buffers to reflect
a virtualenv, so that the external tools invoked by Eglot, Flycheck,
etc., work correctly.
This leads to two issues with `run-python'. First, it doesn't inherit
the buffer-local `exec-path', which is probably the correct behavior.
Second, just having a buffer-local `exec-path' breaks the mechanism that
uses `python-shell-virtualenv-root' to find the Python interpreter.
`process-environment' is likely affected by similar issues.
On a related note: is there a good reason for
`python-shell-calculate-process-environment' not to update the PATH
entry? It's not inconceivable that a Python script needs to call a
program which is only available in the virtualenv. Moreover, from the
end-user perspective, it would be convenient if adding a call to
(setq-local process-environment (python-shell-calculate-process-environment))
in some hook would make tools like `M-x compile' follow the
`python-shell-virtualenv-root' settings.
This bug report was last modified 4 years and 128 days ago.
Previous Next
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.