-
Notifications
You must be signed in to change notification settings - Fork 352
feat: use output from new doc-ci with odoc 3 features in the package documentation area #3124
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
Co-authored-by: Jon Ludlam <[email protected]>
Co-authored-by: Jon Ludlam <[email protected]>
3c7ef34
to
62e8c97
Compare
Co-authored-by: Jon Ludlam <[email protected]>
It was added in ocaml/odoc#1314 and needs not be ignored Co-authored-by: Jon Ludlam <[email protected]>
Co-authored-by: Jon Ludlam <[email protected]>
Co-authored-by: Jon Ludlam <[email protected]>
9a5f418
to
62d32b9
Compare
I converted this to a draft, but feel free to ping me any time when it makes sense to test this. I will put this on https://staging.ocaml.org, and hook it up with the staging docs-ci. |
Co-authored-by: Jon Ludlam <[email protected]>
Co-authored-by: Jon Ludlam <[email protected]>
Co-authored-by: Jon Ludlam <[email protected]>
Co-authored-by: Jon Ludlam <[email protected]>
Odoc 3 allows to have breadcrumbs without href. Currently, the breadcrumbs are still computed by ocaml.org, but soon it will be taken from odoc3's json output. Co-authored-by: Jon Ludlam <[email protected]>
With odoc 3 and hierarchical documentation, breadcrumbs will include pages. They are rendered with `/` between them (while other components are rendered with `.` between them). So We can have `library1/Module.Submodule`. Pages cannot yet happen since we still compute breadcrumbs ourselves. Co-authored-by: Jon Ludlam <[email protected]>
Co-authored-by: Jon Ludlam <[email protected]>
Co-authored-by: Jon Ludlam <[email protected]>
Does this make #1963 obsolete? |
Yes. |
From https://alpinejs.dev/directives/cloak : > Sometimes, when you're using AlpineJS for a part of your template, there is a > "blip" where you might see your uninitialized template after the page loads, > but before Alpine loads. > > `x-cloak` addresses this scenario by hiding the element it's attached to until > Alpine is fully loaded on the page. > > For `x-cloak` to work however, you must add the following CSS to the page. > > ``` > [x-cloak] { display: none !important; } > ```
52920b2
to
3f494dd
Compare
Since odoc 3, and the replacement of voodoo by odoc_driver as the odoc driver, the layout for docs has changed. Since we do not want to break previously existing links, we redirect the old layout to the new one.
@sabine @jonludlam This is ready for testing on staging! (I don't know what makes the "deployability" CI fail but I don't think it's related) |
Deploying on staging.ocaml.org (https://deploy.ci.ocaml.org/job/2025-06-06/074257-ocluster-build-384752) |
Did you take the screenshot from staging.ocaml.org? Here is how the same page looks for me: I'll try to reproduce with different browser and settings, but if you have any hint, I'll take them. |
Documentation assets are not present as json files. Therefore, if we do not find a .json target, we try to load it as an asset. This is not ideal: assets always require two requests to the documentation server. 404 pages also trigger two requests...
Ok, testing on chromium yields the same CSS problem... |
Deploying updated PR to staging. |
6c9e7de
to
b30d4c1
Compare
Replaces #3123, but PR is on
main
.This PR uses the output from the new doc-ci @jonludlam has been working on. The docs are built with odoc 3, benefiting from several features, some of them that were previously "emulated" on this repo (such as a global sidebar).
This currently implements:
This PR is not yet complete, however all pushed commits can already be reviewed, so I think it was already worth opening.
I tried to make the commits as atomic as possible to ease the review, which can then be made "commit by commit". All commits compile, although they may fail at runtime (eg due to wrong format for deserializing).
I'll post updates and warn when I consider the PR complete!