Skip to main content

Extracting Metadata via Crossref API

In addition to OpenAlex, EcoLit synchronizes with the CrossRef API — the world's largest DOI (Digital Object Identifier) registration agency for academic publications. This module enables you to look up precise metadata for any paper via its DOI.

1. Crossref Endpoint Mapping​

The platform provides local endpoints designed to match the native structure of Crossref:

Upstream Crossref EndpointApp API EndpointLookup Domain
GET /worksGET /api/crossref/worksSearch and filter works across the Crossref database.
GET /works/{doi}GET /api/crossref/works/{doi}Retrieve detailed metadata for a paper via its DOI.
GET /journalsGET /api/crossref/journalsList academic journals registered with Crossref.
GET /journals/{issn}/worksGET /api/crossref/journals/{issn}/worksRetrieve works published in a specific journal via ISSN.
GET /members/{id}/worksGET /api/crossref/members/{id}/worksRetrieve works belonging to a specific publisher member.

2. Crossref Query Parameter Syntax​

Unlike OpenAlex, Crossref uses its own naming conventions for query parameters and filters, which are fully supported by EcoLit:

  • query= (Keyword search): Free text keyword search across all publication metadata. E.g., query=inflation targeting
  • filter= (Crossref filters): Filter by metadata criteria.
    • Date Range: filter=from-pub-date:2020-01-01
    • Author ORCID: filter=has-orcid:true (Only return papers where at least one author has an ORCID).
    • Research Funder: filter=funder:10.13039/100000001
  • rows= (Page size): Crossref allows up to 1,000 records per request. E.g., rows=500
  • cursor= (Deep Pagination): When downloading thousands of records, use the cursor (cursor=*) to paginate instead of page numbers.
  • mailto= (Polite Pool): The system automatically appends the email address configured in settings to place requests into Crossref's Polite Pool, increasing response speeds and preventing rate-limiting.

3. Fallback Proxy Workflow​

Similar to OpenAlex, when you request a paper by DOI (e.g., /api/crossref/works/10.1016/j.jfineco.2020.04.019):

  1. The system searches for this DOI in the papers_core table in the local PostgreSQL database.
  2. If the record is found, it is returned immediately in Crossref JSON format.
  3. If not found, EcoLit acts as a Proxy: sending the request to the upstream Crossref server (https://api.crossref.org/works/10.1016/j.jfineco.2020.04.019) with the configured mailto header.
  4. The retrieved metadata is stored in the local PostgreSQL database and the citation references are extracted (paper_references table) to build interactive citation graphs.