barrucadu / bookdb Goto Github PK

I don't really have any intention on doing any further work on it, so I should migrate to a more updated framework. Something with digestive-functors support would be nice, as then I can simplify all the form processing.

There should be an optional sorting key, to allow authors with different display names and sorting. The motivation for this is Lord Dunsany:

The display name should be "Lord Dunsany", but for sorting purposes, the name is just "Dunsany".

I just read a blog post about history tables, and it reminded me that I've wanted that sort of feature for a long time in bookdb. If there was a log of changes to the reading flag and location, the system would be much more flexible.

I've never got around to implementing it because I won't have any history before that point, which feels like a waste. But putting it off is just increasing the waste!

If a book already has a cover image and the description is updated without a new cover image provided, the old one will be lost (the file is still there, but the database entry doesn't reference it).

Solution 1: Don't overwrite the path to cover image if it's already set, and a new one isn't provided.
Solution 2: Don't store the path to cover image in the database, just use the ISBN (requires normalising file extension, which is currently the only reason it's in the db)

Solution 2 is probably better, although it would require hitting the filesystem to check if books have cover images. Maybe a combined approach is optimal, where the db has a "has cover" flag (which can be set once, and is then never unset), and normalised filenames.

Volume numbers are currently sorted lexicographically rather than numerically.

To work around non-numeric volume numbers (eg, "4a") sorting can attempt to split into a numeric prefix and a textual suffix.

When looking at some search results, or a filtered book list, the suggestion in the top-right corner should be drawn from books in the list.

I have timestamps going back to 2012 or so, it shouldn't be too hard to produce a few graphs and tables. Things which might be interesting:

• Number of books read per month, overlaid with last year and average over all years
• Books read this year (with books read this month or week highlighted)
• Least recently read books

Because of one really long book tittle, there is a lot of wasted space in the Title column, so I end up having to scroll left and right a lot for no real reason

1. Drop the Stylesheet :: Sitemap and Javascript :: Sitemap routes.
2. Add a Static :: String -> Sitemap route, which renders as "/static/[string]"
3. Use wai-middleware-static or something to serve static files, or just put it into the nginx config.
4. Stop relying on externally-hosted jquery and chart.js

I've run this in my local docker using the python:3.8 image.

When I run the app, i get this page: The search server is unavailable. Try again in a minute or two.
I did pip install -y requirements-freeze.txt
basically everything that was in the Dockerfile.

This is after editing serve.py because the redirect breaks the site completely.
from : BASE_URI = os.getenv("BASE_URI", "http://bookdb.nyarlathotep")
to: BASE_URI = os.getenv("BASE_URI", "http://192.168.1.76:8888")

After that I gave up. I dont think this is a repeatable install inside or outside docker.
You also have a port variabl 9200 that is on localhost but not open or called to be open. so i opened that on my container.

I like the project idea... hopefully you can make it so that it works correctly.
I cant spend more time trying to get it to work :(

See translator for "The Foundation Pit": https://www.barrucadu.co.uk/bookdb/list

Authors get broken up into a list and individually hyperlinked, other fields don't.

Would be nice to have a view containing a list of books chosen to be read, with the ability to cross off books and see completion progress. Maybe even have deadlines.