Ivan Perez-4 wrote:
Matthew Bromberg wrote:
I can't help you about the haddock thing but, in my experience, if you forgive that kind of mistakes, many users will write documentation with lots of warnings and say "ok, these docs ain't right but, it works for me". I guess this is mostly what has happened to HTML, you can find lots and lots of documents that "work" (well, depending on the browser you are using), but they are obviously wrong.
Maybe the solution to this kind of problems is haddock to give more information about the parse error, like: "Found token A when expecting one of B or D", so at least you know what you should have written instead. _______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
I understand your point, but this is documentation, not mission critical code. I am fortunate that I even have time to try to fix this right now. Had I seen this last week I would have just given up in frustration. Here is another arcane example -- | compute cos (theta / 2) assuming the branch [-pi, pi] halfcos :: Double -> Double halfcos cs = sqrt $ (cs + 1)/2 This produces a parse error parse error in doc string: [haddock.exe: reading EOF! on the h of the halfcos :: Double -> Double line Don't get me wrong, I am appreciative of the fact that this tool exists, I'm just having trouble using it properly. Also I'm the kind of person that views warnings when compiling my source code as errors. Thus I wouldn't accept unproperly formatted html. For code documentation, however, I can't get that worried about it. -- View this message in context: http://www.nabble.com/Haddock-Parse-Errors-tf2396951.html#a6684878 Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.