References to method in object types

[Julow]

References to methods in object types are not working.

type t = < m : int >
class cl : object method m : int end

In this example, cl.m resolves but t.m doesn't.

[lpw25]

I think that one is intended behaviour. I don't think we allow any references to parts of type expressions.

[Drup]

Don't we allow references to fields of record types ?

[Julow]

Indeed, we only allow references to nominal things (constructors, record fields). We also disallow references to polymorphic variants constructors.
However, should we consider adding this feature ? For example, we have very big polymorphic variants that we create type aliases for, as if they were nominal: https://github.com/ocaml/odoc/blob/master/src/model/paths_types.ml

[lpw25]

Those are not type expressions, they are datatype definitions. That is an important distinction in OCaml, and currently it is where odoc draws the line about what may be referenced.

We could change where we draw the line about what is linkable, but that is a broader question than just whether to link to methods in objects. At one extreme you could allow linking to the foo method in:

type t = < foo : int > -> < bar : string >

as t.left.foo. At the other extreme you could ban references to class methods since there is no case where you ever refer to a class method directly in OCaml itself.

[yawaramin]

Today in function docstrings we have the @param tag to describe parameters, since we similarly can't refer to individual parameters. Perhaps we should have a @method tag for object methods? E.g.,

type t = < foo : int >
(** Blah.

    @method foo get the foo. *)

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. The main purpose of this is to keep the issue tracker focused to what is actively being worked on, so that the amount and variety of open yet inactive issues does not overwhelm contributors.

An issue closed as stale is not rejected — further discussion is welcome in its closed state, and it can be resurrected at any time. odoc maintainers regularly check issues that were closed as stale in the past, to see if the time is right to reopen and work on them again. PRs addressing issues closed as stale are as welcome as PRs for open issues. They will be given the same review attention, and any other help.

[Julow]

Let's leave this closed.