Don't remove the doc of inline includes #595
Conversation
Julow
force-pushed
the
inline-include-doc
branch
from
94c7be8 to
e80b926
Compare
|
Are you intending to implement the rest of the suggested fix? |
|
Not in this PR. |
|
I'm not entirely clear on the status after this PR, but I do think that we should have the rules for where to put comments settled before 2.0.0. We're about to ask people to update their docs to match the new rules to get good docs.ocaml.org output, and I think we should try to avoid having to ask them twice. I think the important thing for this bit is that you should be able to add a preamble comment to a module like: https://github.com/janestreet/base/blob/master/src/hash.mli by adding a first comment to the associated module type here: https://github.com/janestreet/base/blob/master/src/hash_intf.ml#L144 |
Julow
force-pushed
the
inline-include-doc
branch
from
e80b926 to
2abaed2
Compare
This makes adding a new test slightly less painful by: - Don't require 'expected' files to be present. Promote will create them. - Add a 'make promote-html' rule. This rule still promote one file at a time but it abstracts the _scratch/actual thing and works with the first goal.
Julow
force-pushed
the
inline-include-doc
branch
from
2abaed2 to
95237ca
Compare
|
It's likely we'll revisit this code with the other module doc changes, but we'll merge this now for the tests. |
This should work fine since #627 |
Fixes the initial issue of #478 but doesn't implement @lpw25's suggestion.
The first doc-comment of the expansion of an include was always removed. This PR avoids this for
@inlineincludes.The first two commits are unrelated.