Switch books page lists to solr from db

[benbdeitch]

Closes #11020

This PR enables Edition pages to fetch the lists that a given edition is from the Solr database in a single query, rather than having to perform multiple queries on the infogami database. This should improve the loading times. It also allows us to only display non-trivial lists on these pages (i.e: lists that contain more than a single entry.)

Technical

Testing

Step 1: Create two new lists on your local environment. One with two books, and one with only one book.
Step 2: Navigate to the page of the book they share in common, and examine the Lists section.
Step 3: If only one list is shown on that page, then the code is behaving properly.

Screenshot

Stakeholders

@cdrini

[cdrini]

Nice work @benbdeitch ! A few comments, but one core blocker. One of the goals of the issue is to fix "We have to query twice against the DB and show 2 rows base on lists including the Work v. Edition". Looking at the code, I think we might need to rename/tweak _get_lists_solr_uncached and friends, since this requires take advantage of a new type of solr query to search for both work and edition keys. Eg something like:

@public
def get_book_lists(work_key: str, edition_key: str) -> list[List]:
    q = f'seed:("{work_key}' OR "{edition_key}"')
    ... the query you have now in `_get_lists_solr_uncached`

That would let us achieve the stated goal, and change the partials.py code to use this method, and only call render_template("lists/widget", ...) once with the results!

[cdrini]

Ahh, I think this is the issue! We don't need the split the keys, the entire key is stored in solr, eg /works/OL3261646W

[cdrini]

Oh and we'll want to forward along both, eg get_book_lists(work and work.key, edition and edition.key), and update the method to handle None. That'll let it do the magic of combining both!

[cdrini]

Now that the solr reindex finished, this is no longer necessary!

Suggested change

	filter_query = "seed_count:[2 TO *] OR (NOT seed_count:*)"
	filter_query = "seed_count:[2 TO *]"

[cdrini]

Lgtm! Did some refactoring of the lists widget html since it was kind of tricky to work through. The lists look much better now with the seed_count filter, too!