kadircet added a comment.
In D84839#2187222 <https://reviews.llvm.org/D84839#2187222>, @ilya-golovenko
wrote:
> @kadircet I decided to add `OnlyChildren` to `enum VisitKind` instead of
> making a `VisitKindSet` - this helped to minimize the changes in
> `shouldVisit()`. What do you think?
SGTM.
================
Comment at: clang-tools-extra/clangd/FindSymbols.cpp:203
return;
- llvm::Optional<DocumentSymbol> Sym = declToSym(AST.getASTContext(), *ND);
- if (!Sym)
- return;
- if (Visit == VisitKind::DeclAndChildren)
- traverseChildren(D, Sym->children);
- Results.push_back(std::move(*Sym));
+ if (Visit == VisitKind::OnlyChildren) {
+ traverseChildren(D, Results);
----------------
ah this is complicated now, but at least for the sake of nestedness:
```
if (Visit == OnlyChildren)
return traverseChildren;
// We must have the Decl in the result set.
auto *ND = llvm::cast<NamedDecl>(D);
auto Sym = declToSym(AST.getASTContext(), *ND);
if (!Sym) return;
Results.push_back(std::move(*Sym));
if(Visit == OnlyDecl) return;
assert(Visit == DeclAndChildren && "Unexpected VisitKind");
traverseChildren(Result.back()->children);
```
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D84839/new/
https://reviews.llvm.org/D84839
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
