sammccall added a comment.
I think the design should be more thoroughly considered here.
- what are the latency consequences of the extra index lookup in different
scenarios?
- how does this compare to doing it at LSP resolve time instead?
- if we're going to do the extra lookup, can we make use of ranking signals
from the index too?
================
Comment at: clangd/CodeComplete.cpp:1369
+ // Keys are indices into Output vector.
+ llvm::DenseMap<size_t, SymbolID> OutputIndex;
----------------
I don't think we can inline this much logic into `runWithSema()` for each
feature we add - need to find a clearer way to structure the code.
================
Comment at: clangd/CodeComplete.cpp:1398
+ llvm::DenseMap<SymbolID, std::string> FetchedDocs;
+ Opts.Index->lookup(DocIndexRequest, [&](const Symbol &S) {
+ if (!S.Documentation.empty())
----------------
If we're going to query the index again here, it seems we should do it earlier
so we can use the results for ranking.
Repository:
rCTE Clang Tools Extra
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D56492/new/
https://reviews.llvm.org/D56492
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
