Add show --json functionality

[MarckK]

Fixes #5261

[brcrista]

@MarckK FYI, I just deleted the "Pip CI (Docs)" pipeline, so that's why I cancelled your build there

[MarckK]

Hi @brcrista, Pip CI(macOS) is failing an integration test with python 2.7. Not sure how to resolve. Advice appreciated :)

[pfmoore]

@MarckK The MacOS error is here: https://dev.azure.com/pypa/pip/_build/results?buildId=2521&view=logs (I think...)

@brcrista It's quite hard to get to the actual error details from the "Details" link in the Github CI checks. Is this something that can be improved? Either by documentation or (better!) by configuring things so that the test output is easier to get to?

[MarckK]

Hi @pfmoore, thanks, yeah, I found it. Just wasn't sure how to fix the failing test, and the following didn't clarify for me: https://dev.azure.com/pypa/pip/_build/results?buildId=2521&view=logs

[pfmoore]

Ah, OK. Yeah, that test failure looks like pip --force-reinstall isn't updating any files. But on MacOS only, and only for Python 2.7. That's weird.

[dstufft]

It might be intermittent. Some of our tests are really flakey

…

Sent from my iPhone

On Nov 22, 2018, at 4:23 AM, Paul Moore ***@***.***> wrote: Ah, OK. Yeah, that test failure looks like pip --force-reinstall isn't updating any files. But on MacOS only, and only for Python 2.7. That's weird. — You are receiving this because you are on a team that was mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[pfmoore]

I hit re-run on that test. Let's see what happens...

[pfmoore]

First impression - Github hasn't noticed that I did so...

@brcrista can you help? Why isn't hitting the "re-run" button showing up here?

[MarckK]

Hi @dstufft, @pfmoore, @brcrista, I'd love to think it was a flaky test, but I've rebased to update today's changes, and (force) pushed, triggering the tests again, and the same test has failed again.
Pip CI(macOS) is failing an integration test with python 2.7: pip --force-reinstall isn't updating any files.

[brcrista]

Hey everyone, I was gone on the Thanksgiving holiday in the US.

@pfmoore there's a known bug where the "re-run" button on the GitHub PR won't trigger builds from forks. You can still re-run on the Azure Pipelines side (you need to have permission to run a build):

[brcrista]

@pfmoore about what you said about the GitHub checks page -- could you explain what info you're looking for here? I see the log lines for the failing test under "Annotations." When I click "Raw output" it shows the output from the failing test.

[BrownTruck]

Hello!

I am an automated bot and I have noticed that this pull request is not currently able to be merged. If you are able to either merge the master branch into this pull request or rebase this pull request against master then it will be eligible for code review and hopefully merging!

[cjerdonek]

I think this PR is getting too long and too hard to review because it's not just adding the json format. There is a lot going on. For example, one important thing is making sure the default format is being preserved correctly in the refactor, and I'm not sure there is adequate test coverage of the default format for us to be falling back on to rely on. Basically, a bunch of refactoring is happening at the same time as an addition of new behavior, and it's hard to separate them. (And there are 27 commits.)

My suggestion would be to start by biting off a smaller chunk. Namely, submit a PR that only refactors the existing implementation into (1) getting the info, and (2) formatting it with the existing format. That PR can also start off by making sure there is adequate test coverage of the default case to make sure the refactor is preserving the output. Then, after that is committed, the second PR would only have to format the info as JSON and add tests only for that.

(Also, if written correctly, I'm not sure you'd even need more than a trivial JSON-specific test, since all the JSON formatter does is call json.dumps(). Most of the testing effort, then, should be going into testing the function that generates the data structure you pass into the JSON formatter or default formatter. Those tests would be useful for both the JSON and default formatter, but I'm not seeing tests of that form here.)

Even as a maintainer I strive for PR's with limited scope and frequently break things up into smaller PR's. It makes writing and reviewing PR's easier for everyone, and makes it easier to make forward progress.

[BrownTruck]

Hello!

I am an automated bot and I have noticed that this pull request is not currently able to be merged. If you are able to either merge the master branch into this pull request or rebase this pull request against master then it will be eligible for code review and hopefully merging!

[chrahunt]

Hello! Thanks so much for your contribution. It has been some time since this has seen activity, so I will close this PR. This is not saying these changes are not useful, just that it is not something we're considering actively worked on right now.

@MarckK or anyone else interested in picking this up, please don't hesitate to open a new PR including these changes. I would pay special attention to @cjerdonek's comment above - if there's some part of the code which needs to be cleaned up, doing that in a separate PR before the functional change can really help move things along smoothly.

Thanks again!