Add ign-rendering to feedstock

[JShep1]

Signed-off-by: John Shepherd [email protected]

Checklist

• Title of this PR is meaningful: e.g. "Adding my_nifty_package", not "updated meta.yaml"
• License file is packaged (see here for an example)
• Source is from official source
• Package does not vendor other packages. (If a package uses the source of another package, they should be separate packages or the licenses of all packages need to be packaged)
• If static libraries are linked in, the license of the static library is packaged.
• Build number is 0
• A tarball (url) rather than a repo (e.g. git_url) is used in your recipe (see here for more details)
• GitHub users listed in the maintainer section have posted a comment confirming they are willing to be listed there
• When in trouble, please check our knowledge base documentation before pinging a team.

I'm on the Ignition Gazebo team and attempting to get the remaining Ignition packages into conda-forge for increased Windows support. This is my first time contributing to conda-forge so I'm open to feedback for steps I may have missed or other improvements that I need to make. I followed the contributing instructions as closely as I could but understand I may have made mistakes along the way.

There are a few more Ignition packages that we would like to add to conda-forge (see here) so once I get feedback on this one, I'll open PRs for the remaining Ignition packages.

[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 (recipes/libignition-rendering-recipe) and found it was in an excellent condition.

[traversaro]

Hi @JShep1 , I think that at the moment the recipe is failing as the build.sh and bld.bat scripts are missing. Fortunately, those scripts look similar for most CMake-based projects, and so this applies also for ignition libraries. For an example of this build scripts, see:

• https://github.com/conda-forge/libignition-math4-feedstock/blob/master/recipe/bld.bat
• https://github.com/conda-forge/libignition-math4-feedstock/blob/master/recipe/build.sh

[JShep1]

Hi @JShep1 , I think that at the moment the recipe is failing as the build.sh and bld.bat scripts are missing. Fortunately, those scripts look similar for most CMake-based projects, and so this applies also for ignition libraries. For an example of this build scripts, see:

• https://github.com/conda-forge/libignition-math4-feedstock/blob/master/recipe/bld.bat
• https://github.com/conda-forge/libignition-math4-feedstock/blob/master/recipe/build.sh

Thanks for the advice! I'll get the ign-rendering equivalent scripts in soon.

[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 (recipes/libignition-rendering) and found it was in an excellent condition.

[Tobias-Fischer]

To get OpenGL, try adding this to the meta.yaml:

In the build section:
{{ cdt('mesa-libgl-devel') }}  # [unix]
{{ cdt('mesa-dri-drivers') }}  # [unix]
{{ cdt('libselinux') }}  # [linux]
{{ cdt('libxdamage') }}  # [linux]
{{ cdt('libxxf86vm') }}  # [linux]
{{ cdt('libxfixes') }}  # [linux]
{{ cdt('libxext') }}  # [linux]
{{ cdt('libxau') }}  # [linux]

In the host & run sections:
xorg-libx11  # [unix]
xorg-libxext  # [unix]

I don't think you'll need freeglut, but let's see (remark: freeglut is not available on osx).

[JShep1]

To get OpenGL, try adding this to the meta.yaml:

Thanks! Greatly appreciated

[Tobias-Fischer]

No worries - seems like it did the job :)

[wolfv]

hmm, the internet tells me that defining GLX_GLXEXT_LEGACY=1 might help the issue with the GLintptr?

Really happy to see more robotics here! Welcome!

[traversaro]

On Windows, a specific test is failing:

39/56 Test #39: UNIT_RenderingIface_TEST ..............***Failed    0.42 sec

Probably it may be easier to debug and check if this is happening also for a build of the library done "manually". If you want to temporary exclude that test by adding it to the exclusion option -E (that is already used for INTEGRATION|PERFORMANCE|REGRESSION .

[traversaro]

hmm, the internet tells me that defining GLX_GLXEXT_LEGACY=1 might help the issue with the GLintptr?

Note that this seems to induced by the inclusion of the GL/glx.h header in https://github.com/ignitionrobotics/ign-rendering/blob/f8ec77282d02a0b92de1ee5b24c5a3bd4948775e/ogre/src/OgreRenderEngine.cc#L22 , a reasonable first test may be to just check if the included header is actually necessary by removing it.

Note that (as it typically happens with ignition libs) a similar inclusion is also happening in Classic Gazebo ( https://github.com/osrf/gazebo/blob/10863ffa8ec28e8e0e3d3a3a588d21cdc78055ab/gazebo/rendering/RenderEngine.cc#L27 ), so you can also try to checkout the gazebo feedstock ( https://github.com/conda-forge/gazebo-feedstock/tree/master/recipe ), because that line seems to be working fine there

[traversaro]

For the macOS failure, it seems that the tests require an SDK targeting 10.13 . I remember we had a similar problem for another ignition library, but I can't remember which one it was, perhaps @Tobias-Fischer remembers. However, given that in this case the problem only affects the tests, it could make sense to disable the test compilation.
It was probably something related to this file: https://github.com/conda-forge/gazebo-feedstock/blob/master/recipe/conda_build_config.yaml .

[Tobias-Fischer]

Yes that's correct re OSX, See also the doc: https://conda-forge.org/docs/maintainer/knowledge_base.html#requiring-newer-macos-sdks

On gazebo we should add the run requirements as described in the docs .. we're naughty not doing that at the moment.

[Tobias-Fischer]

Hi @JShep1, could you please also add sth like in this commit for OSX? conda-forge/gazebo-feedstock@06af3f5

[JShep1]

Hi @JShep1, could you please also add sth like in this commit for OSX? conda-forge/gazebo-feedstock@06af3f5

Aha, thanks, I originally had this line but second guessed myself 😅

[Tobias-Fischer]

osx fails because we didn't re-render. Not sure how that works in staged recipes?!?

[traversaro]

osx fails because we didn't re-render. Not sure how that works in staged recipes?!?

I am not sure, but a possible strategy is just skip osx for now, merge the PR and add osx back in the feedstock once it is created.

[Tobias-Fischer]

Sounds good to me. Or even merge as is, as the problem should resolve by itself?

[traversaro]

Sounds good to me. Or even merge as is, as the problem should resolve by itself?

I am not sure if it is possible to merge PRs on staged-recipes with failing build status, I guess @wolfv knows better.

[wolfv]

Awesome!

I left some small comments.

[wolfv]

I am going to ask on the conda-forge gitter if we can merge with the failing build or if we need to skip OS X first.

[traversaro]

Something to consider for future ignition-* recipes is to understand if there is a way to generate the feedstock with a name different from the recipe, to avoid the usual confusion of a feedstock with a name with the number that generates recipes also subsequent major releases.