sammccall added a comment. In D90588#2370604 <https://reviews.llvm.org/D90588#2370604>, @kbobyrev wrote:
> In D90588#2368021 <https://reviews.llvm.org/D90588#2368021>, @sammccall wrote: > >> This doesn't feel quite right to me - we're going to need to get PP >> conditional regions, include blocks etc from the ParsedAST (they're not in >> ASTContext). >> My sense is that we'll need a fairly random subset of ParsedAST, and so >> ParsedAST is a reasonable abstraction unless it's hard to produce for tests. >> But it isn't! >> >> What's the motivation for this change? > > Aww that comment is regarding `syntax::buildSyntaxTree` weren't you? I > thought this was about `getFoldingRanges` API... Sorry. Oops, sorry - yes :-( My the point there was that the traversal roots for building the syntax tree are the ASTContext's traversal scope, which defaults to TUDecl, so passing in TUDecl is misleading there. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D90588/new/ https://reviews.llvm.org/D90588 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
