Update ninja to include fortran support

[jcfr]

Checklist

• [x] Used a fork of the feedstock to propose changes
• [x] Bumped the build number (if the version is unchanged)
• [x] Reset the build number to 0 (if the version changed)
• [ ] Re-rendered with the latest conda-smithy (Use the phrase code>@<space/conda-forge-admin, please rerender in a comment in this PR for automated rerendering)
• [x] Ensured the license file is being packaged.

List of changes since v1.9.0:

$ git shortlog v1.9.0..28a7d14 --no-merges
Brad King (20):
      ManifestParser: Fix typo {expectd => expected}
      Factor out a base class of ManifestParser
      Assert precondition in BuildStatus::BuildEdgeStarted
      Allow EdgeFinished and NodeFinished to fail with errors
      Teach FakeCommandRunner to support multiple active commands
      Factor out edge marking logic from Plan::AddSubTarget
      Factor out output edge ready check from Plan::NodeFinished
      Make a Builder optionally available to Plan
      Explicitly avoid repeat deps loading
      Add a parser for a new "dyndep" file format
      Add a "dyndep" reserved binding to the manifest format
      Teach DependencyScan to load a dyndep file
      Teach RecomputeDirty to load dyndep files that are ready
      Teach builder to load dyndep files when they are ready
      clean: remove unnecessary Cleaner constructor variant
      clean: remove outputs specified by dyndep files
      graph: load dyndep files
      query: load dyndep files for queried edges
      Document `dyndep` binding behavior and the dyndep file format
      Document example dyndep use cases

Graham Christensen (1):
      Docs: Make builds reproducible by generating the same IDs from the same inputs

Jan Niklas Hasse (5):
      Fix Fuchsia typo
      Take CPU set limitations into account when calculating processor count
      Remove trailing whitespace from all files
      Ignore .ccls-cache directory
      Fix UB "member call on null pointer of type 'DepsLog'", see #1248

Nicolas Despres (1):
      Resurrect the 'rules' tool.

ddrone (1):
      Update link to premake

goshhhy (1):
      make inline.sh more portable

jhuels (1):
      Feature/add term env dumb to win32 (#1550)

[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.

[jcfr]

This PR simply updates the version of ninja by downloading the latest version, it does NOT downnload the source from a different repository like what was done in https://github.com/conda-forge/ninja-feedstock/pull/12

[frol]

It is not 1.9.0 version. It would be much easier for everyone if the upstream just release a new version. If they cannot (e.g. they feel that the current changes are not polished enough, yet), it does not make sense for us to package this early version as a stable release replacement. The only thing we can do about it is to release it as a dev version, but I don't remember all the incantations.

[jcfr]

The only thing we can do about it is to release it as a dev version

Makes sense. Is it a matter on appending the .dev suffix ?

[frol]

It needs to go to a separate branch of the feedstock and it may also require some configuration tuning. I will only have bandwidth to look into it later next week.

[jcfr]

Sounds good. If you could point me toward some documentation, I could help as well. In the mean time, thanks again for your input and help.

[frol]

That is the whole point that I need to find the documentation or an example to follow :)

[jcfr]

Following a suggestion of @msarahan , looking at the pre-release package for python 3.8 seems like a good approach. See https://github.com/conda-forge/python-feedstock/blob/pre-3.8/recipe/meta.yaml

We now need to choose a release number for this pre-release, for example pre-1.10.0 (current version if 1.9.0), I will send an email to the ninja mailing list to know what version number they anticipate for their next release.

[wolfv]

this is released now, I guess?