[bugfix] Environment shell script path is now formatted based on the subsystem

[jellespijker]

The shell script for the environments should be formatted based on the subsystem now.

Fixes #11980

Changelog: (Bugfix): Environment shell script path is now formatted based on the subsystem Docs: https://github.com/conan-io/docs/pull/XXXX

• [X] Refer to the issue that supports this Pull Request.
• [ ] If the issue has missing info, explain the purpose/use case/pain/need that covers this Pull Request.
• [X] I've read the Contributing guide.
• [X] I've followed the PEP8 style guides for Python code.
• [ ] I've opened another PR in the Conan docs repo to the develop branch, documenting this one.

^{Note: By default this PR will skip the slower tests and will use a limited set of python versions. Check here how to increase the testing level by writing some tags in the current PR body text.}

[jellespijker]

Yes I agree with the risk there, I will investigate this in more detail. Also try out how it behaves in more user cases.

The new win-bash approach looks promising btw

[jellespijker]

@memsharded the failing test occurred in #11987 which turned out to be a non issue. I do think this PR is save to merge and actually fixes the bug in described in #11980

[jellespijker]

@memsharded / @lasote has this fix been considered to make it to the next release. The error you were concerned about happened in another PR (which is now closed since it turned out to be a non-issue) This fix is relatively safe and contained imo.

And I think I will run into issues with some open PR's:

• https://github.com/conan-io/conan-center-index/pull/12892
• https://github.com/conan-io/conan-center-index/pull/12898#issuecomment-1246196467

[jellespijker]

As my colleague @Joeydelarago pointed out, the hint is in the comment:

This would need a test that would be failing without this fix.

I will write a testcase for this PR

[memsharded]

Hi @jellespijker

It has been some time, maybe this was no longer a priority? Failed to write the testcase? It is true that it is weird that subsystem is not passed there, but on the other hand, there have been no reports about this, so it seems subsystems are fine with the native path? It would be good to know a bit better, have failing tests for this.

[jellespijker]

I'm fine with closing this PR. Our initial work, which required this fix, didn't end up in our production env. So this isn't PR isn't need by us anymore.

[CLAassistant]

All committers have signed the CLA.