Scrapers
A scraper is a provider that searches for or extracts metadata about an existing library item. It enriches or identifies a record; it is not primarily responsible for acquiring the source media.
Inputs and results
Section titled “Inputs and results”Scrapers can use record metadata, URLs, file names, codes, or user-supplied queries to find candidates. A result can contain descriptive fields, relationships, artwork, URLs, and provider-scoped remote IDs. Capabilities vary by scraper and media type.
Searching, selecting a match, and applying metadata are separate stages. A candidate can be reviewed without modifying the record. Applying it follows the workflow’s merge or replacement choices and can attach provenance to imported values.
Implementations
Section titled “Implementations”A compiled extension can implement scraper interfaces and services. A YAML scraper pack can contribute declarative scraping definitions without a compiled assembly. Both are installed extensions and must use supported host contracts.
Remote sites can change independently of Cove. Scraper failure can reflect upstream markup, authentication, rate limits, or network access rather than an invalid local record.