#12574: Consistent use of sigs vs signatures in warnings -------------------------------------+------------------------------------- Reporter: crockeea | Owner: Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 Resolution: | Keywords: newcomer Operating System: Unknown/Multiple | Architecture: Type of failure: Incorrect | Unknown/Multiple warning at compile-time | Test Case: Blocked By: | Blocking: Related Tickets: #11583, #10752 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by martinceresa): Hi! I've found the error. It seems that when looking for warning messages in DynFlags.hs/wWarningFlagsDeps for 'Opt_WarnMissingLocalSignatures' it finds first the depecrated version message. Swapping those two lines, getting something like this: {{{#!hs wWarningFlagsDeps :: [(Deprecation, FlagSpec WarningFlag)] ..... flagSpec "missing-local-signatures" Opt_WarnMissingLocalSignatures, depFlagSpec "missing-local-sigs" Opt_WarnMissingLocalSignatures "it is replaced by -Wmissing-local-signatures", ..... }}} solves this problem, but it is not the best solution. It is clear that there is an implicit order when looking for warning massages, and we need to check that function to see why it selects just the first it finds. I am new to GHC development so I am still fighting with the code, and it may be worth open a new ticket to modify such function, and solve possibly any other 'bad warning massage'. Cheers. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/12574#comment:3 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler