nmos-cpp: new release and conan v2 migration

[garethsb]

Specify library name and version: nmos-cpp/cci.20221124

---

• I've read the guidelines for contributing.
• I've followed the PEP8 style guides for Python code in the recipes.
• I've used the latest Conan client version.
• I've tried at least one configuration locally with the conan-center hook activated.

[garethsb]

Waiting for #13959 and #13962.

[garethsb]

This is still waiting for #13959 and #13962 in order to have any chance of building on CCI, but locally, I'm just struggling with the paths to the packaged executables, which means that the test package doesn't run successfully.

@samuel-emrys This is the PATH issue that we discussed on Slack.

I attempted a few different ways of adding the right directory in package_info():

The recipe's existing approach:

        self.env_info.PATH.append(bin_path)

seems to have no effect on the generated conanrunenv-release-x86_64.bat. Understandable I think as env_info is the "old" way.

However, my first attempt:

        self.cpp_info.bindirs = [bindir]
        self.cpp_info.libdirs = [libdir]

also seem to have no effect on that file. The default bin directory is prepended to the PATH, not my specified bindir...

My next attempt (updated after realising I needed append_path rather than plain append):

        self.runenv_info.prepend_path("PATH", bin_path)

works OK.

However, the test package can't run successfully, because the test package uses subprocess.Popen as well as self.run. I've updated the self.run call to use env="conanrun", but I think I need to get the variables from the VirtualRunEnv for the call to subprocess.Popen(..., env=???) and I can't work out the syntax for that.

[garethsb]

Now broken by issue discussed in #14223.

[garethsb]

Sigh... 🪲 Macos, x86_64, apple-clang, 12.0, libc++, Release....

ERROR: nmos-cpp/cci.20220120 (test package): Error in test() method, line 44
	self.run(bin_path + " node-config.json", env="conanrun", output=node_out)
	ConanException: Error 138 while executing ./test_package node-config.json

[prince-chrismc]

3 tiny comments

• source for the patches
• nicer comment
• more details for the patches (our goal is to have smaller patches so the extra comments need a bit more explaining)

[uilianries]

LGTM

[prince-chrismc]

Sorry for dismissing the review 🙊 I just wanted to make sure the linters worked correctly

[garethsb]

Sorry for dismissing the review 🙊 I just wanted to make sure the linters worked correctly

I copied that list format from another CCI recipe too 🙄

If https://github.com/conan-io/conan-center-index/blob/97b198659f0b6ac39b259a4d99232c1ac725d629/docs/adding_packages/conandata_yml_format.md#patch_source constitutes the spec, I'm not convinced the multiline format is any more valid than the list... 😖

[conan-center-bot]

Conan v1 pipeline

All green in build 28 (4e0dcd822dadc1ed3de69d5984b93052857efcf9):

• nmos-cpp/cci.20221124@:
  All packages built successfully! (All logs)

• nmos-cpp/cci.20220208@:
  All packages built successfully! (All logs)

• nmos-cpp/cci.20220620@:
  All packages built successfully! (All logs)

• nmos-cpp/cci.20220428@:
  All packages built successfully! (All logs)

• nmos-cpp/cci.20220120@:
  All packages built successfully! (All logs)

[garethsb]

Withdrawing this PR. Will clean it up and submit again. Thanks for all reviews and assistance so far @uilianries, @prince-chrismc.

[prince-chrismc]

If https://github.com/conan-io/conan-center-index/blob/97b198659f0b6ac39b259a4d99232c1ac725d629/docs/adding_packages/conandata_yml_format.md#patch_source constitutes the spec, I'm not convinced the multiline format is any more valid than the list... 😖

conan-center-index/linter/conandata_yaml_linter.py

Line 35 in 64577d2

Optional("patch_source"): Str(),

:)

This is only for humans to read and there's not really a way to make it "machine readable" that I could image so I did not make it more complicated 🤷 I understand it's not intuitive but I am pretty sure I need to redo that linter in the new year