-
-
Notifications
You must be signed in to change notification settings - Fork 9.3k
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
Fix regression in API caused by commit 85400d8d6751071ef78f042d1efa72… #4052
Conversation
- The added optional parameter changes API and should default to None This utility call is used by for example requestbuilder package directly which breaks because it passes only one argument to the function as it used to be.
I don't believe this is a public API, and therefore we don't gaurentee API compatibility. But, I could be wrong — either way, it seems like a nice change to make if it doesn't hurt anything. @nateprewitt any objections? |
I think this is a good idea, but we should work out why the builders are failing first. ;) |
Yeah, I don't see any issue with this, even if we're not supporting this API publicly. As for the build, something in our dependency pipeline must have changed yesterday or earlier. Python <= 3.3 builds started failing on |
I think this has to be a pytest-httpbin thing. The error is ultimately because we now believe that the cert that pytest-httpbin is presenting on the secure endpoint now has a subjectAlternativeName field. I'm trying to locate it and work out why that warning isn't firing. |
Hrm, no, that's not true. This test passes if run by itself, so this is a new state based test failure for some reason. |
Ok I have a theory, testing now. |
The new release of Pytest from yesterday has some interesting notes in the changelog.
|
Yeah, this is the warning issue: see #4056. |
Ok cool, can we rebase this on the new master? |
@Lukasa no conflicts! |
@kennethreitz, I kicked the Travis builds and they seem to be working with the changes to master. If you can kick appveyor, we should be able to merge. |
Hi everyone, sorry for barging in, but was this regression related to pytest's 3.1.0 warnings capture? |
Hey @nicoddemus, no need to apologize 😊 The regression is unrelated to the pytest 3.1.0 changes but those changes were effecting our build when this PR was opened. We've since implemented the suggested workaround in #4056. |
Great, thanks for the answer! 👍 |
So I'm not in favor of merging this. As was said earlier, this is not a public API. |
@sigmavirus24, if we're not going to maintain this as a public API, it should probably be moved to Edit: or modify the docstring. There's a fair amount of stuff in |
Codecov Report
@@ Coverage Diff @@
## master #4052 +/- ##
=========================================
+ Coverage 88.15% 89.65% +1.5%
=========================================
Files 15 16 +1
Lines 1941 1962 +21
=========================================
+ Hits 1711 1759 +48
+ Misses 230 203 -27
Continue to review full report at Codecov.
|
…bdcf76cc0e
The added optional parameter changes API and should default to None
This utility call is used by for example requestbuilder package directly
which breaks because it passes only one argument to the function as it
used to be.