Since 2.8 example output looks like this:
======= test session starts ========
platform linux -- Python 3.4.3, pytest-2.8.7, py-1.4.31, pluggy-0.3.1
rootdir: $REGENDOC_TMPDIR, inifile:
collected 0 items
======= no tests ran in 0.12 seconds ========
before it looked like this:
=========================== test session starts ============================
platform linux -- Python 3.4.1 -- py-1.4.27 -- pytest-2.7.1
rootdir: /tmp/doc-exec-162, inifile:
collected 0 items
This is basically a cosmetic thing, but I think it would be better to have something looking like a real path there again to decrease the confusion potential - especially for newcomers.
Might also be an easy first issue for someone wanting to make themselves familiar with some of pytests build tools?
GitMate.io thinks a possibly related issue is https://github.com/pytest-dev/pytest/issues/62 (better html output for examples (sphinx highlighter)).
Hi @obestwalter,
I agree that using a real path would be better.
I thought this would be configured here:
But it doesn't seem to be the case. I'm sure @RonnyPfannschmidt knows where this is configured though.
@obestwalter at each execution its a different folder, thats simpy not directly usable
im happy to accept a pr to regendoc to fix this to something more visually pleasing
Yeah, I think the idea is to hard-code a folder: /tmp/regendoc for example
@nicoddemus its already on github https://github.com/pytest-dev/regendoc/
Oh right, even better then 馃榿
If all occurrences can be replaced with something generic I would go for something like /path/to/project to make clear that this is a generic example folder in a documentation snippet (and that the root is usually the root of your project).
@obestwalter i disagree, since its not the project dir in that case
edit: s/sine/since/
@RonnyPfannschmidt sorry, what do you mean by sine?
rootdir: $REGENDOC_TMPDIR, inifile:
I agree with @obestwalter that rootdir: /path/to/project seems appropriate... do you have another thing in mind?
/path/to/example
Also good :)
Thanks for the fix @aswinm - closing this then :)