Official website of Department of Mathematics at the University of Virginia.
The documentation (including howtos on updating content) is a part of the website.
Phase 1 adds an unlinked /arxiv/ placeholder and infrastructure for a UVA Mathematics arXiv tracker. It includes shared arXiv DB checks, roster-history extraction, arXiv source-affiliation evidence, and Semantic Scholar/CrossRef smoke caches. It does not publish paper data yet.
Useful commands:
make uva-arxiv-checkverifies the configured shared SQLite DB, source corpus path, and API-key presence without printing secret values.make uva-arxiv-db-since-dryreports the plannedsinceupdate without fetching or writing; setSINCE=YYYY-MM-DDand pass extra script flags throughARGS=....make uva-arxiv-db-sinceruns the updater; useARGS="--limit N"for smoke updates.make uva-arxiv-roster-historyprints roster-history counts without writing outputs.make uva-arxiv-source-smoke ID=2501.01234dry-runs one arXiv source fetch.make uva-arxiv-api-smoke ID=2501.01234runs one Semantic Scholar smoke check.
Direct scripts expose the underlying smoke flags: update_arxiv_db.py since --db --since --overlap-days --limit --dry-run --source --endpoint, sources.py fetch --id --sources-dir --force --dry-run --endpoint, roster_history.py --dry-run --no-write --as-of, and the API clients' --cache-path, --refresh, and --no-cache.
The scripts use Python 3.11+, git, make, and SQLite. Shared defaults live in scripts/uva_arxiv/config.yml and can be overridden with ARXIV_DB and ARXIV_SOURCES_DIR; ignored .env files may also provide S2_API_KEY, SEMANTIC_SCHOLAR_API_KEY, CROSSREF_MAILTO, and CROSSREF_API_KEY.
External metadata and evidence caches are stored under scripts/uva_arxiv/cache/ and are ignored except for .gitkeep. Missing Semantic Scholar/CrossRef metadata is incomplete metadata, not publication evidence. Missing or absent UVA source-affiliation evidence is also an evidence state, not a rejection decision.