Previously, when viewing a /videos listing, there was a link
called (Chan) to bring you to the channel page, but since videos
only carried author_id and not author_name it was always kind of
ugly. This will inject that attribute on the way out.
I know, this would be more properly written as an SQL join in the
first place, but my row-dict conversion isn't set up for that and
I'm planning on converting this all to object-based returns instead
of dicts soon.
This makes the page much easier to use on mobile. On desktop I think
the new size may be a little overwhelming but I'll try getting used
to it, and anyway I think it simply makes more sense than the
arbitrary size from earlier.
Example: If you are using the /pending filter, and use the search
box to find some videos and ignore them, then those videos will
be removed from the DOM. That way they aren't still there when you
clear the text box to see the other pending videos.
- Download / Ignore on multiple videos is batched into a single
request, like it should have been all along.
- Text box at the top of the page lets you search for terms live
instead of loading various ?q urls.
- Doing a force refresh will actually update the metadata of old
videos instead of skipping them.
- Add thumbnails to the card (dynamically loaded)
- Add button for embedding the video
- Store video duration in db (use update script)
- Separate links for filter views with and without query.
voussoir