Suppressing .env loading message when PIPENV_QUIET is set #3457
Conversation
| with open(dotenv_path, 'w') as f: | ||
| pass | ||
|
|
||
| with mock.patch('pipenv.environments.PIPENV_DOTENV_LOCATION', dotenv_path): |
There was a problem hiding this comment.
It's better to set PIPENV_QUIET environment variable, rather than patching the function here.
There was a problem hiding this comment.
Well I wanted to to this, but the PIPENV_QUIET variable is removed at import-time :
Line 271 in 1f36ca1
Do you think patching PIPENV_VERBOSITY variable is acceptable instead ?
There was a problem hiding this comment.
Sorry I meant this line :
Line 270 in 1f36ca1
There was a problem hiding this comment.
Okay, setting PIPENV_VERBOSITY to -1 looks fine to me.
There was a problem hiding this comment.
I can squash in one commit if you prefer
There was a problem hiding this comment.
I don't have preference on this. Either is ok for me
There was a problem hiding this comment.
Ok, I just pushed the squashed commit
thunderk
force-pushed
the
quiet_env_loading
branch
from
6faf9e2
to
a6beb61
Compare
|
Why the CI in windows for python27 can't pass... |
|
I don't know why CI fails on windows, #3426 has the same error. It fails while creating the virtualenv, with the error: I'm not sure how this could be related with my changes, they do not include non-ascii characters. |
|
Your PR looks good to me. I don't think the failing CI is related with your PR since it is a so minor change. But it still worth investigating why the CI fails... |
|
One more thing to say is that we might still need to make the test pass first, otherwise merging this may make our CI failing all the time. @thunderk Do you mind add a news first? |
thunderk
force-pushed
the
quiet_env_loading
branch
from
d5dfd57
to
cca4861
Compare
|
@jxltom Sure, I added the news, I hope it's ok |
|
Offhand, are you using an editor that respects |
|
(We aren’t going to merge this until it’s sorted out either way, whether it’s a CI issue or a code issue, so let’s get to the bottom of it) |
|
@techalchemy Yes, I use an editor aware of .editorconfig. I checked again, and the files I edited seems to be saved as utf-8. I will try to add the coding:utf-8 header in the test file (it was missing, but I can't see non utf-8 chars in my diff). |
thunderk
force-pushed
the
quiet_env_loading
branch
2 times, most recently
from
054ddd2
to
c3545a3
Compare
|
OK, I'm stuck, I tried to go back from master and rewrite my changes with another editor (just to be sure). Also checked the modified files with some utf-8 validator, and still no luck... |
thunderk
force-pushed
the
quiet_env_loading
branch
from
c3545a3
to
1c6520c
Compare
thunderk
force-pushed
the
quiet_env_loading
branch
from
1c6520c
to
9f04c93
Compare
|
well, note that all tests were failing until recently, so that may be contributing... the changes seem ok to me so i'll just update your pr with the current changes and see if that works |
|
(I think the transient test failure issue is resolved and this is probably something we can merge now!) |
techalchemy
added
Category: CLI
|
Any news on this item. I am also interested to get ride this notification. |
This is a proposed fix for issue #2358 that have been closed without solution.
The fix suppresses the message "Loading .env environment variables" that is sent to stderr, when PIPENV_QUIET environment variable is set.
This is very useful for people that use pipenv run in cron commands, because writing to stderr may systematically report the cron as error.