Continuing Feature/record download info #507
Conversation
…cord it in info.ini
| @@ -12,5 +12,6 @@ notifications: | |||
| branches: | |||
| only: | |||
| - develop | |||
| - feature/record-download-info | |||
There was a problem hiding this comment.
this addition should not be part of the pull request
|
Don't worry about Python 2.4 testing, we've bumped the minimum supported version to 2.5 for the next pip release. |
|
I'm seeing the two bundle failures as well (and not seeing them in develop branch), I'll try to take a look soon and see if I can figure out what's going on with those. |
|
Hmm I'm pretty sure I've seen those failures with the tests cache sporadically when running locally - it's possible we have a non-hermetic test which doesn't clean up after itself and this CL changes the ordering causing the failure. |
|
@dstufft , this is fairly old. still important? if it is, can you explain what scenario is for this? I guess just to fill in the history of how that package got there? something more to give people in |
|
I'm going to close this, the use case I had in mind is one I no longer thing is actually a good idea. |
lock
bot
added
the
auto-locked
I don't know if there's a better way to do this, but I'm continuing where @dstufft left off over in #453.
I've tried to address the concerns raised in the original (hardcoded paths and other test issues).
I'm a little puzzled as to why the tests failed so spectacularly in python 3, but I started with replacing the iteritems() and the errors all seemed to go away. Here's a Travis CI build: http://travis-ci.org/#!/tgecho/pip/builds/1061109
The two test failures in 2.7/3.1/3.2 are for bundling, but I can't quite wrap my head around them yet. If someone with a better clue about pip internals could take a stab we'd be grateful.
There's another error in test_editable_git_upgrade in 2.5/2.6 that I'm not sure is related, but I'm looking at that. I'll also try to get a test on 2.4.