================
@@ -610,8 +610,10 @@ static bool ParseDirective(StringRef S, ExpectedData *ED,
SourceManager &SM,
}
FileID FID = SM.translateFile(*File);
+ // FIXME: Figure out character-encoding converter treatment.
if (FID.isInvalid())
- FID = SM.createFileID(*File, Pos, SrcMgr::C_User);
+ FID = SM.createFileID(*File, Pos, SrcMgr::C_User,
+ /*InputEncodingName=*/{});
----------------
cor3ntin wrote:
That should use the input encoding.
However, you should consider that all clang own test files are utf8-ended and
should not be converted.
So you might want to modify lit to force utf-8 for test files (of anything
compiled with clang),
or document that if there is no tag set for the entiere llvm source tree,
nothing will be expected to work.
https://github.com/llvm/llvm-project/pull/206519
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
