aaron.ballman added a comment. In D145228#4174829 <https://reviews.llvm.org/D145228#4174829>, @sammccall wrote:
> @aaron.ballman > > If someone wants to work on producing a more modular, component-based design, > that's probably a good idea (but tricky to get right). Probably a good place > to start is working out what pieces can be moved into separate libraries with > e.g. a coherent layering story, thinking about how they might be reused in > other tools etc. > > Exposing all clangd's headers as-is is not that. I don't think the idea that > llvm-project is modular is at odds with the idea that leaf subprojects exist, > though clangd is big enough that it may make sense to split up. And private > headers (those that live next to source files in the tree) are not installed > (with the recent exception of `clang-tidy`, which IMO should have been > restructured to publish in this way). Ah, okay, thank you for the clarification! I agree we have to go about splitting the project up into libraries very carefully and this approach isn't appropriate. But I was getting the impression there was push-back against the idea of making clangd into a series of libraries in general and that worried me. > FWIW, I do think clang is a fair example of a project where the default of > "everything is a public library" has made the library design somewhat > incoherent and hard to evolve. OTOH it's enabled tons of useful stuff, so... IMO, we've not really had too many issues evolving Clang's libraries, but I do think there's room for improvement (it's hard to figure out how to split up Sema, for example). As you say, it's not an easy problem to solve but it enables a lot of really useful stuff we wouldn't otherwise be able to achieve as easily. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D145228/new/ https://reviews.llvm.org/D145228 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
