aaron.ballman added inline comments.
================
Comment at: clang/unittests/AST/AttrTest.cpp:19
+TEST(Attr, Doc) {
+ EXPECT_THAT(Attr::getDocumentation(attr::Used),
+ testing::HasSubstr("The compiler must emit the definition even "
----------------
It seems this is failing the premerge CI -- I think you need a call to `.str()`
in here to convert the `StringRef` to a `std::string`?
================
Comment at: clang/utils/TableGen/ClangAttrEmitter.cpp:4237
+ static const llvm::StringRef AttrDoc[] = {
+ #define ATTR(NAME) AttrDoc_##NAME,
+ #include "clang/Basic/AttrList.inc"
----------------
kadircet wrote:
> i am not well-versed in tablegen, so sorry if i am being dense here, butt
> what happens to the attributes we skipped above (e.g. the ones without a
> `ASTNode` flag)?
>
> Maybe we should be emitting a `... AttrDoc_X[] = "";` for all attributes, no
> matter what?
The ones without an AST node don't get an `attr::Kind` enumeration generated
for them, so I believe this interface is safe.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D107703/new/
https://reviews.llvm.org/D107703
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
