Extract a work's structured metadata fields

The shared source of truth for the work landing page's metadata panel (`generate_work_metadata_html()`), its YAML frontmatter (`generate_work_metadata_yaml()`) and its JSON representation (`render_register_json()`), the same three-way split used for venues (see [get_venue_metadata_fields()]).

Usage

get_work_metadata_fields(doi, register_table)

Arguments

doi

The work's DOI (`table_details[["name"]]` on a work page).

register_table

The work's filtered register rows (one per certificate that checked this DOI), still carrying `Repository`, `Paper Title`, `OpenAlex`, `Venue`, `Check date`.

Value

A list with `title`, `doi`, `openalex` (`NA_character_` if none), `publication_date` (`NA_character_` if none), `venues` (unique, comma-joined), `check_count`, `first_check_date`, `last_check_date`, and `authors` (a list of `name`/`orcid` lists, `orcid` `NULL` when not known - possibly an empty list).

Details

Authors are fetched once, from the first certificate's `codecheck.yml` (via the same cache every other lookup in the render uses) - a work's author list does not vary by which certificate checked it, so refetching per certificate would only mean more requests for the same answer.