I apologize if I am not supposed to mail through this list if I am a newcomer, I’ve never used mailing lists and I can’t find any documentation about how to use it properly.
Either way, I’ve started using clangd9 lately after enjoying cquery for a while, I really like it but the feature I miss the most is the macro referencing. I tend to work with the linux kernel a lot so the migration to clangd improved the indexing time incredibly but I still need to find references to macros many times per work-day.
I glanced a over the source code and I think I can write the feature and submit it for review, but I am unsure about the overall procedure. Where do I submit it to? Who looks over it and who accepts it?
I hope I am not making a fool of myself.
Thank you, Avishay.
Thanks for your interesting in contributing to clangd! Yes, this mailing list is a fine place to ask questions of this sort.
the feature I miss the most is the macro referencing
I would find this useful as well.
I glanced a over the source code and I think I can write the feature and
submit it for review, but I am unsure about the overall procedure.
Where do I submit it to? Who looks over it and who accepts it?
The procedures followed for clangd are largely the same as those for the larger LLVM community, which are documented here . One thing that's a bit different is clangd has its own issue tracker on Github .
Patches are submitted to LLVM's Phabricator instance . In terms of who to ask for review, you can look at who has reviewed other recent changes to the files you're touching. However, you can also submit a patch without explicitly setting reviewers. I think the clangd team monitors and will notice all review requests with "[clangd]" in the commit message.
Hope that helps. Feel free to ask more questions if you have them!
Thank you for the guidance Nate!
I’ll do my best to contribute ASAP
Welcome and thanks for your interest in clangd!
In addition to Nathan’s comments, I would also suggest discussing the approach you want to take before submitting a patch for review.
Especially if the approach can be condensed to a few sentences.
Things usually go faster if reviewers and the author are aligned on the direction of the patch beforehand.