#10543: MacOS: validate fails on \u -------------------------------------+------------------------------------- Reporter: trommler | Owner: Type: bug | Status: closed Priority: normal | Milestone: Component: Build System | Version: 7.10.1 Resolution: invalid | Keywords: cpp Operating System: MacOS X | Architecture: Type of failure: Building GHC | Unknown/Multiple failed | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Revisions: Phab:D1004 -------------------------------------+------------------------------------- Comment (by rodlogic): Replying to [comment:5 geoff]:
So, I was the one to add this line — overriding `-Wno-invalid-pp-token -Wno-unicode -Wno-trigraphs` — because these flags are not universally supported by clang. If anyone could help me understand this better, it would be appreciated. https://github.com/Homebrew/homebrew/issues/41777#issuecomment-122826887
I am not that familiar with GHC's build system, so I can't answer that. Hopefully someone else can? Here is the question, inlined: {{{ @rodlogic Is there a reason why the stage 0 bootstrap has to be built with -Werror? These warnings are pretty benign (indeed, more benign than the flags to suppress them). And what exactly is the significance of the \u in the source file, in this particular case? Support for unicode literals are not mandatory in POSIX in many places, and Mac OS X being a certified UNIX system, this is unlikely to change. The only way to reliably encode an arbitrary Unicode character in ASCII is is to use the octal sequences that comprise the glyph. Or, you could simply use the actual glyph, treating the source file as UTF-8, of course. But this seems particularly odd to me here, since the character that follows is t! So why is this needed here at all? What am I missing? }}} -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/10543#comment:6 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler