ioeric added inline comments.
================
Comment at: clangd/index/SymbolCollector.cpp:631
+
+bool SymbolCollector::isSelfContainedHeader(FileID FID) {
+ // The real computation (which will be memoized).
----------------
sammccall wrote:
> ioeric wrote:
> > sammccall wrote:
> > > this has been moved to a member so it can use the cache and the
> > > (non-threadsafe) regex, and its logic has obviously been changed.
> > "self-contained header" doesn't sound accurate anymore. Maybe something
> > like `isIncludable`?
> Let me explain how I think about it, and then I'll change the name if you
> still don't think it fits.
>
> What we're detecting here (usually) is that this header can't be included
> unless a certain symbol/state is defined first from the outside.
> Whether that's for some direct technical reason (e.g. configuration is
> needed) or to enforce a coding style, the dependency on external preprocessor
> state means the header isn't self contained. (And the non-self-containedness
> is the reason we can't include it in arbitrary contexts)
That sounds good. Thanks for explaining!
Repository:
rCTE Clang Tools Extra
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D60815/new/
https://reviews.llvm.org/D60815
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
