Enhanced Reading Log Exports

[cclauss]

Closes #6519
Blocked by #6734 -- Rebase this PR after 6734 lands.
Related to #6645 and #6653 which also use csv_string().

Export your Reading Log with attention to adding more fields and without building large lists in memory which makes this process unwieldy for users with large reading lists.

• Create iterate_users_logged_books() to get around the bug discussed at the bottom of get_users_logged_books()
  • Repeatedly get blocks of books and yield one book at a time. (lower memory footprint)
• Create new generate_reading_log() to build the CSV by yielding one line at a time. (lower memory footprint)
  • Need to use database (NOT requests!) to access works, editions, authors, redirects
  • Need to know how to deal with long entries caused by long lists (ex. Tom Sawyer's Subjects, People, Places)
  • Need to know how to access other data fields such as my_rating, date_read, read_count

Current output: OpenLibrary_ReadingLog(65).csv or see Screenshots below.

What code returns the results for:

• http://localhost:8080/works/OL54120W.json
• http://localhost:8080/books/OL2058361M.json

The /api/books endpoint does not return enough information...

• http://localhost:8080/api/books?bibkeys=OLID:OL54120W&format=json
• http://localhost:8080/api/books?bibkeys=OLID:OL2058361M&format=json

from openlibrary.plugins.books.dynlinks import ol_get_many_as_dict
ol_get_many_as_dict(['/books/OL2058361M', '/works/OL54120W'])

ol_get_many_as_dict() results are missing keys covers and description but we don't need them for CSV.

Being able to instantiate a openlibrary.code.models.Work would substantially reduce code complexity but this class has no .__init__() method. If I know the key (e.g. /works/OL8193488W), how can I get an openlibrary.code.models.Work instance for that key?

Answer: web.ctx.site.get(key)

As in work: Work = web.ctx.site.get("/works/OL8193488W") returns an openlibrary.plugins.upstream.models.Work

Technical

Testing

In one terminal tab: docker compose up
In another terminal tab: open http://localhost:8080
Search on Twain and set one book each to: Want to Read, Currently Reading, Already Read and add star ratings.
On http://localhost:8080/account/import click Export your Reading Log --> Download (.csv format)

Screenshot

Stakeholders @mheiman, @Fl0rent

[mekarpeles]

LGTM, only concern is Reading Logs with e.g. 10k+ items (if we're building the csv in memory) OR if we have multiple patrons doing exports at the same time, a more tenable solution could be to use a temp file (instead of memory).