Scholar Citation Meta Scanner

For pages that list many works: finds every DOI on the page and offers the full citations to Zotero as a pick-list (COinS). Off by default, enabled per page.

What it does

Scholar Citation Meta Scanner is for a page that lists many works. Switch it on for a post or page, and it finds every DOI on it, looks each one up, and embeds the results as hidden COinS tags. The Zotero browser button then shows a folder icon; clicking it opens a pick-list of every work on the page, with full citations.

For a page about a single work, where you want Google Scholar to index the page and link your PDF, use the companion plugin Scholar Citation Meta. The Scanner works on its own or together with it.

Examples

• One paper (Meta): The urban-tech feedback loop, a journal article in Digital Geography & Society. Google Scholar has what it needs to index the page as that article, and Zotero saves it as a journal article.
• One book chapter (Meta): Land for AI, a chapter in The Inner World of Artificial Intelligence (CRC Press). Zotero saves it as a book section with the book title and publisher.
• A list of works (Scanner): Papers + Writings, a full publications list. Zotero offers every listed work with a DOI as a pick-list.
• Both together: Doing Critical GIS, a special-issue introduction with the issue's other articles listed below it. Meta describes the introduction itself for Google Scholar. The Scanner gives Zotero a pick-list with the introduction first, followed by the other articles.

Using both

The two plugins are built to run side by side. On a page with both:

• Google Scholar still reads Meta's tags, so the page is still described as its own work.
• Zotero prefers the Scanner's list, and the page's own work is put first in it (you can turn this off per page).
• If a DOI on a list page belongs to one of your own Meta posts, the Scanner uses that post's details, including any BibTeX fixes such as missing page numbers.

Install

1. Download scholar-citation-meta-scanner-<version>.zip from the latest release. Download the release asset, not GitHub's "Source code" archive.
2. In WordPress, go to Plugins → Add New → Upload Plugin, choose the zip, and activate. If an older copy is installed, choose Replace current with uploaded.
3. In the editor sidebar, open the Zotero DOI list box, tick Scan this page for DOIs, and click Update.
4. Reload the editor to see what was found, including any DOIs with no metadata. The page is scanned again every time you update it.

If Scholar Citation Meta is active, a second option puts the page's own article first in the list (on by default).

What it finds

• Links to doi.org / dx.doi.org
• Publisher links that contain a DOI (e.g. tandfonline.com/doi/full/10.1080/...)
• doi:10.xxxx/... and bare 10.xxxx/... text
• Duplicates are removed, so the same DOI linked twice is listed once.
• Title slugs after a DOI in publisher links are dropped.
• If a book DOI and one of its chapter DOIs both appear (10.1201/9781003480167 and 10.1201/9781003480167-9), only the chapter is kept.

Where the metadata comes from

1. Your own posts: if a post has this DOI in Scholar Citation Meta, its metadata is used, including any BibTeX fixes. It is re-read every time the page loads, so later fixes show up without rescanning.
2. Crossref.
3. DataCite's API, if Crossref has no record (Zenodo, ICPSR, figshare, and many OJS journals such as ACME).
4. doi.org, for any other registration agency (mEDRA, JaLC, ...).

Page fields that aren't page numbers (e.g. "Vol. 21 No. 4 (2022): ...") are dropped, and the volume and issue are taken from that text.

Lookups are cached for 30 days and shared across posts. Failed lookups are retried after 24 hours. Each save spends up to 15 seconds on new lookups; anything left over finishes in the background via WP-Cron, about 30 seconds later.

How Zotero behaves

• Zotero prefers COinS over the single-article citation_* tags. On pages where the scanner is on, the browser button shows the list (with your article first, if that option is ticked).
• On pages where the scanner is off, Zotero still saves the single article.
• Google Scholar ignores COinS, so the citation_* tags still control Scholar.
• Zotero's COinS reader attaches a snapshot of the page to each saved item. Users can turn this off with the automatic-attachments option in Zotero's General settings.

Settings (optional, wp-config.php)

define( 'SCMS_POST_TYPES', 'post,page' );
define( 'SCMS_MAX_DOIS', 200 );

Updates

Version 1.4.0 and later update from this repository's Releases. When a new release is published, WordPress shows it on the Plugins and Updates screens, and you update with one click (or let auto-updates handle it). WordPress checks GitHub at most every 6 hours.

Versions before 1.4.0 don't have the updater, so upload 1.4.0 once by hand.

Releasing a new version (maintainers)

1. Bump the version in the plugin header (Version:) and in readme.txt (Stable tag:), and add a changelog entry to readme.txt.
2. Commit and push, then create a release whose tag is v<version>, e.g. gh release create v1.2.3 --generate-notes.
3. The Release workflow checks that the tag matches both version numbers, builds scholar-citation-meta-scanner-<version>.zip, and attaches it to the release. Sites running the plugin will then see the update.

Changelog

• 1.4.1: Update checks fall back to github.com when GitHub's API is rate-limited (common on shared hosting).
• 1.4.0: Updates from GitHub Releases: new versions appear on the Plugins screen like any other plugin.
• 1.3.0: Handles publisher links with a title slug after the DOI; treats tags and line breaks as spaces so text is not glued onto a DOI; keeps a chapter DOI over its book DOI.
• 1.2.0: Checks the DataCite API directly when Crossref has no record, before falling back to doi.org.
• 1.1.0: Uses metadata from your own Scholar Citation Meta posts first; page-field cleanup.
• 1.0.0: First release.

License

GPL-2.0-or-later. See LICENSE.