Ambiguous fixity annotation when only a single operator out of a grouped type signature inside a class definition has a fixity annotation

[ghc-mirror]

Original reporter: haddock@

To reproduce:

class Bug a where
  (<*<), (>*>) :: a -> a -> a
  infix 9 <*<

The haddock output just shows | infix 9 | here since there is only one fixity annotation for that line, except here it is not clear which operator the annotation corresponds to.

I think the proper way to address this would be to split apart lines of a class in general, instead of adding complex special case logic to detect the scenario of attaching a single fixity annotation to a method group (which does not fit into the current design whatsoever).

I'll look into implementing this feature whenever I get some free time and energy unless somebody else wants to.

[harpocrates]

This appears to have been fixed somewhere between 2.17.3 and 2.18.2 since on the latter I get:

[alexbiehl]

Alec, could you try Haddock 2.19? Especially Operators.hs test should be a good test case cfir various things related to fixities.

[harpocrates]

@alexbiehl Just did, and things are looking spiffy!

[alexbiehl]

Alright, good. I fiddled around there a few days ago. What is strange thoughy, I think **< and friends are missing documentation! I think they should have the same docs as **>! Alec Theriault <notifications@github.com> schrieb am Fr., 9. Feb. 2018, 18:07:

…

@alexbiehl <https://github.com/alexbiehl> Just did, and things are looking spiffy! [image: screen shot 2018-02-09 at 9 06 43 am] <https://user-images.githubusercontent.com/10766081/36039870-a0b4857e-0d78-11e8-8e36-7feea7bc58f1.png> — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#287 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AByiiecoF_c6Bca7mL5RF1UFzqMIOQvUks5tTHtggaJpZM4B5Q_6> .

[alexbiehl]

Fixed, see 4b4733a