[PATCH] D129973: [clang] Pass FoundDecl to DeclRefExpr creator for operator overloads

[Shafik Yaghmour via Phabricator via cfe-commits]

shafik accepted this revision.
shafik added a comment.
This revision is now accepted and ready to land.

LGTM



================
Comment at: clang/test/Index/annotate-operator-call-expr.cpp:21
+// CHECK1: Punctuation: "(" [7:6 - 7:7] CallExpr=operator():3:7
+// CHECK1: Punctuation: ")" [7:7 - 7:8] CallExpr=operator():3:7
 // CHECK1: Punctuation: ";" [7:8 - 7:9] CompoundStmt=
----------------
SimplyDanny wrote:
> I updated the test expectation without actually knowing whether the test references are correct(er) now. There seems to be some loss of information and some `CallExpr`s are now seen as `DeclRefExpr`s. Please tell me if these changes are okay. I rather made them to have a basis for discussions.
`DeclRerExpr`  to `CallExpr` looks right.


================
Comment at: clang/test/Index/cursor-ref-names.cpp:36
 // CHECK: cursor-ref-names.cpp:18:5: DeclRefExpr=inst:17:9 Extent=[18:5 - 18:9]
-// CHECK: cursor-ref-names.cpp:19:5: CallExpr=operator[]:4:9 SingleRefName=[19:9 - 19:12] RefName=[19:9 - 19:10] RefName=[19:11 - 19:12] Extent=[19:5 - 19:12]
+// CHECK: cursor-ref-names.cpp:19:5: CallExpr=operator[]:4:9 Extent=[19:5 - 19:12] 
 // CHECK: cursor-ref-names.cpp:19:5: DeclRefExpr=inst:17:9 Extent=[19:5 - 19:9]
----------------
It looks like we lost some information here but I am not sure if that is expected or not. 

AFAIK CIndex is supported on a best effort basis, I think this is ok.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D129973/new/

https://reviews.llvm.org/D129973