[GHC] #8226: Remove the old style -- # Haddock comments.

[GHC]

#8226: Remove the old style -- # Haddock comments.
-------------------------------------+------------------------------------
        Reporter:  Fuuzetsu          |            Owner:  adinapoli
            Type:  task              |           Status:  patch
        Priority:  normal            |        Milestone:  7.10.1
       Component:  Compiler          |          Version:  7.7
      Resolution:                    |         Keywords:
Operating System:  Unknown/Multiple  |     Architecture:  Unknown/Multiple
 Type of failure:  None/Unknown      |       Difficulty:  Unknown
       Test Case:                    |       Blocked By:
        Blocking:                    |  Related Tickets:
-------------------------------------+------------------------------------

Comment (by adinapoli):

 Hi guys,

 I haven't forgot, is that I'm making basically no progress on it.
 After Fuuzetsu's suggestion of removing the old data structure which held
 the 0.x style comments, I stumbled upon a series of difficulties, which
 are, unordered, the result of different edits on the Lexer;

 * Removing the data structure caused immediately GHC to complain on old
 style comments which are actually leftovers, an example is inside
 libraries/ghc-prim/GHC/Classes.hs. I couldn't understand from the sole
 documentation what to do, but it seems to me that, as Fuuzetsu said, they
 can be removed because they are equivalent to a pragma (in the example --
 #hide == {-# OPTIONS_HADDOCK hide #-}. Unfortunately the lexer chokes in
 trying to parse "-- #hide" once the data structure has been removed, even
 though, theoretically, should be possible to modify the Lexer to just
 treat them as a normal comment ; after all the only thing which prevents
 this to happen, I believe, is the "isNormalComment" function.
 Unfortunately I can't modify it as per my first patch (aka removing the
 "#") because this will cause GHC to interpret RULES as normal comments (I
 discovered this thanks to an insight from SPJ).

 Sorry for the noise, I thought it was worth documenting my efforts my
 myself in the future :)
 If someone has some insights, please shout :)
 Alfredo

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8226#comment:7>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler