Windows build for mpc

[isuruf]

No description provided.

[conda-forge-linter]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe) and found it was in an excellent condition.

[jakirkham]

Could we ping him about getting some tags?

What is special/what do we know about this commit?

[isuruf]

Nothing special, this is a commit including the MSVC build files that is in between 1.0.3 and the development version.

[isuruf]

I think the developer is using MPC original git repo and updates his repo from time to time with the MSVC build files rebased

[jakirkham]

Could you please confirm how you generated appveyor.yaml?

[isuruf]

conda-smithy regenerate, but it seems I generated it when python was used, not vc 14. Now when I regenerate appveyor.yml is disabled.

[jakirkham]

Yeah, we haven't added logic for vc yet. Sounds fine. Don't worry about disabled AppVeyor part.

[isuruf]

Anything to be done here?

[jakirkham]

Some sort of testing for Windows would be nice. Perhaps a simple program that exercises some functionality from mpc.

[isuruf]

Tests pass now

[jakirkham]

@msarahan, could you please take a look at this and see if this looks ok to you?

[isuruf]

@msarahan, ping

[isuruf]

pinging again

[jakirkham]

He's been pretty busy trying to fix conda and conda-build bugs. Let's give him a little more time.

[jakirkham]

In the mean time, @patricksnape could you please help review this when you have a chance?

[msarahan]

All looks sane to me. Why is there a custom appveyor.yaml, though?

[jakirkham]

What do you mean by "custom"? I believe this was generated with conda smithy regenerate.

[msarahan]

I see, sorry. I didn't understand why the file was part of this PR at all.

[jakirkham]

I tried to regenerate locally and basically get the same result. Think that @isuruf may have a slightly older version of conda smithy though.

However, there definitely seems to be something funny happening in conda-smithy w.r.t. the skip portion. Seems it decides to skip all Windows builds instead of all Windows builds that are not Python 3.5. 😕 @pelson, do you have any insight on why this is happening?

[jakirkham]

NVM we don't list python and conda-smithy doesn't know about vc yet. Sorry for the noise.

[jakirkham]

I've added a PR ( isuruf#1 ) against your PR, @isuruf, so that we can fix the re-rendering issue.

[jakirkham]

I see, sorry. I didn't understand why the file was part of this PR at all.

Yeah, so normally this is less pronounced as it is results in just moving disabled_appveyor.yml to appveyor.yml. Potentially some differences appear. However, we didn't have disabled_appveyor.yml either so it came across as a totally new file. I'm not totally sure why it didn't already exist though.

[jakirkham]

Does this still look good to you, @msarahan? If so, I'm happy to see this merged once it passes.

[msarahan]

LGTM, OK to merge when green.

[jakirkham]

The queues are really backed up so it might not finish until late. If it passes, please feel free to self-merge so that you can get started on other things, @isuruf.

[isuruf]

Thanks for the reviews