Hahnfeld added inline comments.
================
Comment at: clang/lib/Parse/ParseTentative.cpp:94
+ // involve namespaces and friend declarations.
+ if (NextToken().is(tok::identifier))
+ return true;
----------------
v.g.vassilev wrote:
> I am wondering what is the false positive rate of this change? That is, if we
> enable incremental parsing by default (in a local build) and then run all
> tests which do not produce diagnostics.
With this patch and enabling `IncrementalExtensions` by default, it's at least
not worse than current `main`: For all of `check-clang`, I see a number of
failing unit tests that start hoarding some memory and are killed with
`std::bad_alloc` or are stuck using 100% CPU. Those we probably want to debug
independently of this change...
================
Comment at: clang/test/Interpreter/disambiguate-decl-stmt.cpp:74
+// expected-error@-1 2 {{'Inner' is a private member of 'PR13642'}}
+
// Deduction guide
----------------
v.g.vassilev wrote:
> Could we move the diagnostic-producing cases in a separate file?
We could, but I'm actually not a big fan because it moves the expected failing
test away from the related disambiguation tests above...
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D157838/new/
https://reviews.llvm.org/D157838
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
