Show HN: Librario, a book metadata API that aggregates G Books, ISBNDB, and more

TLDR: Librario is a book metadata API that aggregates data from Google Books, ISBNDB, and Hardcover into a single response, solving the problem of no single source having complete book information. It's currently pre-alpha, AGPL-licensed, and available to try now[0].

My wife and I have a personal library with around 1,800 books. I started working on a library management tool for us, but I quickly realized I needed a source of data for book information, and none of the solutions available provided all the data I needed. One might provide the series, the other might provide genres, and another might provide a good cover, but none provided everything.

So I started working on Librario, a book metadata aggregation API written in Go. It fetches information about books from multiple sources (Google Books, ISBNDB, Hardcover. Working on Goodreads and Anna's Archive next.), merges everything, and saves it all to a PostgreSQL database for future lookups. The idea is that the database gets stronger over time as more books are queried.

You can see an example response here[1], or try it yourself:

  curl -s -H 'Authorization: Bearer librario_ARbmrp1fjBpDywzhvrQcByA4sZ9pn7D5HEk0kmS34eqRcaujyt0enCZ' \
  'https://api.librario.dev/v1/book/9781328879943' | jq .
This is pre-alpha and runs on a small VPS, so keep that in mind. I never hit the limits in the third-party services, so depending on how this post goes, I’ll or will not find out if the code handles that well.

The merger is the heart of the service, and figuring out how to combine conflicting data from different sources was the hardest part. In the end I decided to use field-specific strategies which are quite naive, but work for now.

Each extractor has a priority, and results are sorted by that priority before merging. But priority alone isn't enough, so different fields need different treatment.

For example:

- Titles use a scoring system. I penalize titles containing parentheses or brackets because sources sometimes shove subtitles into the main title field. Overly long titles (80+ chars) also get penalized since they often contain edition information or other metadata that belongs elsewhere.

- Covers collect all candidate URLs, then a separate fetcher downloads and scores them by dimensions and quality. The best one gets stored locally and served from the server.

For most other fields (publisher, language, page count), I just take the first non-empty value by priority. Simple, but it works.

Recently added a caching layer[2] which sped things up nicely. I considered migrating from net/http to fiber at some point[3], but decided against it. Going outside the standard library felt wrong, and the migration didn't provide much in the end.

The database layer is being rewritten before v1.0[4]. I'll be honest: the original schema was written by AI, and while I tried to guide it in the right direction with SQLC[5] and good documentation, database design isn't my strong suit and I couldn't confidently vouch for the code. Rather than ship something I don't fully understand, I hired the developers from SourceHut[6] to rewrite it properly.

I've got a 5-month-old and we're still adjusting to their schedule, so development is slow. I've mentioned this project in a few HN threads before[7], so I’m pretty happy to finally have something people can try.

Code is AGPL and on SourceHut[8].

Feedback and patches[9] are very welcome :)

[0]: https://sr.ht/~pagina394/librario/

[1]: https://paste.sr.ht/~jamesponddotco/a6c3b1130133f384cffd25b3...

[2]: https://todo.sr.ht/~pagina394/librario/16

[3]: https://todo.sr.ht/~pagina394/librario/13

[4]: https://todo.sr.ht/~pagina394/librario/14

[5]: https://sqlc.dev

[6]: https://sourcehut.org/consultancy/

[7]: https://news.ycombinator.com/item?id=45419234

[8]: https://sr.ht/~pagina394/librario/

[9]: https://git.sr.ht/~pagina394/librario/tree/trunk/item/CONTRI...

106 points | by jamesponddotco 11 hours ago

13 comments

  • culi 8 hours ago
    I find WikiData to be perfect for aggregating identifiers. I mostly work with species names and it's perfect for getting the iNaturalist, GBIF, Open Tree of Life, Catalogue of Life, etc identities all in one query

    I haven't tried it for books. I imagine it's not sufficiently complete to serve as a backbone but a quick look at an example book gives me the ids for OpenLibrary, Librarything, Goodreads, Bing, and even niche stuff like the National Library of Poland MMS ID.

    https://www.wikidata.org/wiki/Q108922801

    [-]
    • bbor 2 hours ago
      Cracks me up that OP is trying Anna's Archive before Wikidata, NGL! Both great sources, though.

      I recently (a year ago... wow) dipped my toe into the world of library science through Wikidata, and was shocked at just how complex it is. OP's work looks really solid, but I hope they're aware of how mature the field is!

      For illustration, here are just the book-relevant ID sources I focused on from Wikidata:

        ARCHIVERS: 
  Library of Congress Control Number    `P1144` (173M)
  Open Library                          `P648`  (39M)
  Online Computer Library Center        `P10832` (10M)
  German National Library               `P227`  (44M)
  Smithsonian Institute                 `P7851` (155M)
  Smitsonian Digital Ark                `P9473` (3M)
  U.S. Office of Sci. & Tech. Info.     `P3894`

  PUBLISHERS:
  Google Books                          `P675`  (1M)
  Project Gutenberg                     `P2034` (70K)
  Amazon                                `P5749`

  CATALOGUERS:
  International Standard Book Number    `P212`
  Wikidata                              `P8379` (115B)
  EU Knowledge Graph                    `P11012`
  Factgrid Database                     `P10787` (0.4M)
  Google Knowledge Graph                `P2671` (500B)
      [-]
      • jamesponddotco 1 hour ago
        Not gonna lie, I didn't even know Wikidata existed until now. I'll look into it today and create a ticket for a new extractor.

        Thanks for letting me know!

  • karlding 8 hours ago
    Do you handle books with no ISBN?

    I’ve recently acquired some photo books that don’t appear to have any ISBN but are listed on WorldCat and have OCLC Numbers and are catalogued in the Japanese National Diet Library. Not sure if they actually don't have ISBNs or if I just haven't been able to find them, but from what I got from some research it's quite common for self-published books.

    [-]
    • jamesponddotco 1 hour ago
      No, right now you need an ISBN to search for a book. At a later date I'll implement search by title or author, which should help with this use case.
  • cmaury 10 hours ago
    Are you able to pull upcoming titles? All I want is a weekly/monthly list of books by authors I've ready which are coming out, and I've not been able to find it or to build it.
    [-]
    • jamesponddotco 10 hours ago
      If the data is present in one of the extractors, yes, but I think only Amazon and similar stores keep this kind of data right now. We don't have a extractor for Amazon yet.

      After v1.0.0 is out I plan to add the ability to add books manually to the database, at which point we'll be able to start improving the database without relying on third-party services.

    • culi 4 hours ago
      A couple years I looked for a similar service and failed to find it. I did however find this incredible podcast network called New Books In where they interview authors about their new books. It's a massive network that's broken down by categories that can get pretty niche. Everything from "Digital Humanities" to "Diplomatic History" to "Critical Theory". Episodes appear in multiple categories so broad categories like "Science" also exist

      https://newbooksnetwork.com/subscribe

      It's definitely biased towards academia which I personally see as a pro not a con

  • wredcoll 11 hours ago
    I applaud the effort, but last time I tried this the major issue was the sheer amount of book data only available from amazon.com and scraping that is tedious to put it mildly.
    [-]
    • jamesponddotco 11 hours ago
      Hardcover and ISBNDB have a good amount of data, with Hardcover being excellent for getting good covers and genres.

      I'm hoping Goodreads and Anna's Archive will help fill in the gaps, especially since Anna's Archive have gigantic database dumps available[1].

      [1]: https://todo.sr.ht/~pagina394/librario/12

      [-]
      • culi 4 hours ago
        You should also consider OpenLibrary and LibraryThing. Both of which have good coverage on WikiData which also aggregates identifiers.

        In fact, now that I think about it, you could also contribute your work to WikiData. I don't see ISBNdb ids on WikiData so you could write a script to make those contributions. Then anyone else using WikiData for this sort of thing can benefit from your work

  • apublicfrog 9 hours ago
    Wow. I don't have any use for this personally, but your post is really well presented, detailed and sourced. I hope it goes well!
    [-]
    • jamesponddotco 9 hours ago
      Thanks! It's a compilation of several random comments I made for a few months, haha.
  • krick 9 hours ago
    Does it handle languages other than English? I remember trying out some APIs like that for some tasks, and while I managed to find titles in English somewhat successfully, any other languages (be it the original title, or a translation of some fairly well-known book) were basically inaccessible.
    [-]
    • jamesponddotco 9 hours ago
      I only tested English and Brazilian Portuguese so far, and Brazilian Portuguese worked, with translator information included.
      [-]
      • fc417fc802 8 hours ago
        You're most likely to run into issues with non-latin languages. Particularly picograms and the associated schemes for how to interpret them in a context sensitive manner. Substring search for example is likely to be broken in my experience.
        [-]
  • eichin 6 hours ago
    Tried throwing a batch of known-to-be-in-Amazon ISBN's through (from a recent "export my data", so even if they're old amazon fundamentally knows them.) Got 500's for a handful of the first hundred, then a bunch of 502/503s (so, single threaded, but part of the HN hug to death, sorry!)

    (Only the first 4 or so were json errors, the rest were html-from-nginx, if that matters.)

    [-]
    • jamesponddotco 1 hour ago
      No hug of death, the server is sitting at 3% CPU usage under current load; it seems someone found a bug that triggered a panic, and systemd failed to restart the service because the PID file wasn't removed. Fixed now, should be back online :)
  • ksr 11 hours ago
    Nice, I might try your API for my ISBN extractor / formatter at https://github.com/infojunkie/isbn-info.js

    Right now, I use node-isbn https://www.npmjs.com/package/node-isbn which mostly works well but is getting old in the tooth.

    [-]
  • moritzruth 10 hours ago
    What do you think about BookBrainz?

    https://bookbrainz.org/

    [-]
    • jamesponddotco 10 hours ago
      First time I'm seeing it, to be honest, but it looks interesting. I do plan on having an UI for Librario (built a few mockups yesterday[1][2][3]), and I think the idea is similar, but BookBrainz looks bigger in scope.

      I could add them as an extractor, I suppose :thinking:

      [1]: https://i.cpimg.sh/pexvlwybvbkzuuk8.png

      [2]: https://i.cpimg.sh/eypej9bshk2udtqd.png

      [3]: https://i.cpimg.sh/6iw3z0jtrhfytn2u.png

      [-]
      • nmstoker 10 hours ago
        This is great - the service and that you're extending it and considering a UI.

        Personally I would go with option 2 as the colour from the covers beats the anaemic feel of 1 and it seems more original than the search with grid below of 3.

        [-]
        • jamesponddotco 10 hours ago
          Glad you liked the idea!

          Number two is what my wife and I prefer too, and likely what's going to be chosen in the end.

    • WillAdams 8 hours ago
      Doesn't seem to have a very compleat dataset --- the first book I thought to lok for, Hal Clement's _Space Lash_ (originally published as _Small Changes_) is absent, and I didn't see the later collection _Music of Many Sphere_ either:

      https://www.goodreads.com/book/show/939760.Music_of_Many_Sph...

  • wizzwizz4 10 hours ago
    Please ensure that your database keeps track of whence data was obtained, and when. It's exceptionally frustrating when automated data ingesting systems overwrite manually-corrected data with automatically-generated wrong data: keeping track of provenance is a vital step towards keeping track of authoritativeness.
    [-]
    • jamesponddotco 9 hours ago
      We don't support POST, PATCH, and whatnot yet so I didn't take that into account yet, but it's in the plans.

      Still need to figure out how this will work, though.

      [-]
      • fc417fc802 8 hours ago
        Since you support merging fields you likely would want to track provenance (including timestamp) on a per-field basis. Perhaps via an ID for the originating request.

        Although I would suggest that rather than merge (and discard) on initial lookup it might be better to remember each individual request. That way when you inevitably decide to fix or improve things later you could also regenerate all the existing records. If the excess data becomes an issue you can always throw it out later.

        I say all this because I've been frustrated by the quantity of subtle inaccuracies encountered when looking things up with these services in the past. Depending on the work sometimes the entries feel less like authoritative records and more like best effort educated guesses.

  • omederos 6 hours ago
    502 Bad Gateway :|
    [-]
    • jamesponddotco 1 hour ago
      It seems someone found a bug that triggered a panic, and systemd failed to restart the service because the PID file wasn't removed. Fixed now, should be back online :)
  • ocdtrekkie 8 hours ago
    Library of Congress data seems like a huge omission especially for something named after a librarian. ;) It is a very easy API to consume too.
    [-]
    • jamesponddotco 1 hour ago
      I didn't look into it yet because I assumed the current extractors had the information from them, but it's in my list of future extractors!
  • sijirama 7 hours ago
    hella hella cool

    goodluck