Extract ToC from Deutsche Nationalbibliothek

[zorae]

Proposal

Deutsche Nationalbibliothek has high-quality scans of the table of contents for a large part of their holdings. These can be freely accessed as a PDF. For each OL edition that has a DNB identifier attached, OL could attempt to download the corresponding PDF and extract the ToC text. Note that some of the PDFs have a wildly inaccurate text layer, so it makes sense to run our own OCR.

Example edition page at DNB:
https://d-nb.info/973546166

Example TOC:
https://d-nb.info/973546166/04

See also #8756

Justification

Problem: OL currently only has a table of contents for a small fraction of editions. This impacts patrons’ ability to learn what a book is about.

Impact: Increase the number of ToCs, especially for German-language books.

Research: I’ve been manually OCRing and/or transcribing a number of TOCs from DNB for use on OL and can attest that the scans are of consistently high quality.

Breakdown

Requirements Checklist

• [ ]

Related files

Stakeholders

---

Instructions for Contributors

Please run these commands to ensure your repository is up to date before creating a new branch to work on this issue and each time after pushing code to Github, because the pre-commit bot may add commits to your PRs upstream.