Matt Westcott

Results 440 comments of Matt Westcott

As they say, the two hard problems in computer science are cache invalidation and naming things... There is already a `get_object_list` method, which returns the list of objects to present...

The only real reason is that I was replicating the functionality of the built-in choosers, where the search box always uses Wagtail search backends. Supporting SQL-based search as a fallback...

Design references from elsewhere in Wagtail: ModelAdmin flat page listing ![Screenshot 2023-01-18 at 10 29 17](https://user-images.githubusercontent.com/85097/213148970-ed8dec23-1f66-449f-8eb8-c9d8f2a3d02a.png) ModelAdmin "choose parent page" step ![Screenshot 2023-01-18 at 10 30 43](https://user-images.githubusercontent.com/85097/213149035-5f54322b-11e9-4cf2-98c8-ac82f7fc0979.png) Report with filters...

After iterating on this with the design team, the focus of this feature has now shifted considerably away from a ModelAdmin-like view existing as its own menu item, and towards...

@mohammadareeb95 As mentioned in my comment quoted above - I'm pretty sure that putting it in requirements.txt is not enough for it to be handled automatically as a dependency when...

For background, the use-case that originally prompted the desire to validate against circular redirects was a site that added various `/foo` -> `/foo/` redirects (redundantly, since CommonMiddleware already handled this)....

> Should we also confirm whether the old_path matches an existing view? I take it the idea here is to provide some sort of feedback to the user to indicate...

[Converting URL redirects to page redirects] > Should this be automatic, or a warning during creation or a report? Adding links in RichText automatically detects internal links, but prompts the...

Picking this up again for 6.2 planning... I think we're happy to go ahead with validating against circular redirects, as per [#11659](https://github.com/wagtail/wagtail/issues/11659). Re converting URL redirects to page redirects -...

> Could we perhaps rename the proposed ImageBlock to ImageAltChooserBlock? The rationale for calling it `ImageBlock` is that we want to steer developers towards it as the default option whether...