Simon Marlow wrote:
Generating a single unified index is entirely possible, given all the .haddock files for the relevant packages. I'll try to get this into the next version of Haddock.
Great! Thanks!
As for the rest, this is all stuff that could/should be done by the library infrastructure (which is currently in in the design stage).
Fine - just take it as a data point what a single user would like to see. Whatever that may be worth :-)
2. Remove the split between "type/class index" and "function/constructor index" in the Haddockish doc-index.html file.
Yes, this is also something I'd like to do.
3. Some links in the indexes are bold, others aren't.
The bold ones represent the *original* defining module for the entity, as compared to a module which just re-exports it.
Aaah... a mystery explained :-)
However, knowing this information isn't very useful,
Agreed.
and arguably it shouldn't be displayed to the user at all.
Indeed. I'd consider the original defining point of an entity to be an implementation detail: if it will work the same whether it's defined in the top-level module or in some helper module, the caller wouldn't (and shouldn't have to) care about the difference.
I'll change this for the next Haddock version.
Again, thank you! And your response was very much appreciated. Regards, Jo