[PATCH] D143319: [clangd] Support iwyu pragma: no_include

Younan Zhang via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Thu Feb 9 19:49:54 PST 2023


zyounan added a comment.

In D143319#4115186 <https://reviews.llvm.org/D143319#4115186>, @kadircet wrote:

> thanks a lot for the patch!
>
> we're migrating IWYU related functionality in clangd to make use of include-cleaner library nowadays. so can you actually move the IWYU no_include pragma handling logic into https://github.com/llvm/llvm-project/blob/main/clang-tools-extra/include-cleaner/lib/Record.cpp?
> This won't have affect on clangd regarding the particular problem you're facing immediately, but we're going to merge the logic inside include-cleaner and clangd soon and such discrepancies should disappear then.
>
> in the meanwhile regarding the particular problem you're facing for experimental symbols. sorry for shutting down D142836 <https://reviews.llvm.org/D142836>, but that actually made us re-evaluate the approach of hand-curated lists and we've decided to introduce such mappings. so if you can revamp D142836 <https://reviews.llvm.org/D142836> and put `TsStdSymbolMap.inc` in https://github.com/llvm/llvm-project/tree/main/clang/lib/Tooling/Inclusions/Stdlib/StdTsSymbolMap.inc we can land it and make sure it work as intended inside clangd, even before we merge the IWYU pragma handling logic. Happy to do that myself if you don't feel like doing it, sorry for creating some confusion here :(

Although I'm convinced that they're not stable enough to migrate into Stdlib directory, I'm still glad to hear that we're going to support TS symbols.
And certainly, I'd love to revamp D142836 <https://reviews.llvm.org/D142836>, while I believe that this `IWYU no-include` is a more general approach than that ad-hoc way.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D143319/new/

https://reviews.llvm.org/D143319



More information about the cfe-commits mailing list