Reference data

Return exchange reference entries stored in MongoDB.

Fetch constituents for stored exchanges and merge them into ticker_reference.

Fetch all world exchange indices from OnVista and persist/overwrite in MongoDB.

Safe to call repeatedly — subsequent calls update existing documents.

Retrieve a single exchange by entityValue, name, url_name, or search_term.

Return stocks for an exchange/index from ticker_reference. If the cache is empty, the endpoint auto-syncs that exchange once.

Fetch constituents for one exchange/index and merge them into ticker_reference.

Return cached branch/industry list; auto-fetches on first call.

Return cached country list; auto-fetches on first call.

Return cached currency list (official=true); auto-fetches on first call.

Return cached instrument-type list; auto-fetches on first call.

Return cached sector list; auto-fetches on first call.

Force-refresh all five reference categories from onvista.

Force-refresh a single reference category from onvista.

Return ticker reference entries stored in MongoDB.

Results are sorted by `ranking` ascending (lower rank first; tickers without a ranking last), then alphabetically by name/ticker. The search covers ticker, name and ISIN.

Backfills CIKs from the SEC ticker-to-CIK map (company_tickers.json) into ticker_reference — the precondition for /financials/sync/all to cover all US securities. Collision-safe (only where the CIK is empty and the ISIN is US or empty).

All stocks whose keywords_multilingual contain the given keyword_id — language-independent (the keyword_id is the same across all languages).

• group=company (default): different listings of the same company

(ordinary share, ADR, CDR, …) are joined into ONE company via
company_id — never into a different company. Every result carries
company_id, company_name, iso_country, the primary main listing,
listings[] (all listings including ISINs and tickers, even those
without a keyword of their own) as well as tickers/isins — so a
newsroom can serve the articles for ALL listings. count = number of
companies.

• group=listing: a flat list per individual listing (count = listings).

Sorted by keyword confidence descending. Example: /tickers/by-keyword/enterprise_software.

Background job: enrich every ticker in ticker_reference with ISIN/WKN/logo from onvista.de.

• only_missing: when true (default), skip tickers that already have an isin
• rate_limit: max requests per minute to onvista (default 30)
• concurrency: parallel requests (default 5)

Returns immediately; check progress via GET /tickers/enrich/isin/status (job is tracked in-memory while server is running).

Dry-run: shows what plausibility-based search finds for this ticker, without writing to MongoDB.

Both ticker-symbol and name-keyword searches are always performed. All structurally valid candidates are returned with their plausibility scores so you can see exactly why one was chosen over another.

Background job: enrich ticker_reference rows directly from onvista stock detail payloads.

Strategy:

• use onvista_id when present
• otherwise resolve via exact ISIN match
• on success patch the row and move on to the next one

Incremental onvista refresh. Re-fetches the onvista stock detail (which also rebuilds calendar_events — results, dividends and annual general meetings) for only those tickers whose calendar events are stale.

A ticker counts as stale when it has no calendar_events at all, or its newest calendar_events.synced_at is older than max_age_days. Returns a job_id — poll GET /tickers/jobs/{job_id}.

Lists every keyword_id that occurs, with the number of listings, multilingual labels (en/de/fr/es/it/pt) and average confidence. Sorted by frequency descending. Example: /tickers/keywords?search=cloud.

Background job: compute and persist data_confidence for every enriched ticker. Returns immediately with a count of documents to process.

Aggregate data_confidence statistics across all scored tickers. Returns counts per level and the worst matches (low confidence).

The central identifier resolver: ISIN / ticker / CIK / WKN / onvista entity_value / id_path / symbol → a canonical bundle including entity_type.

The basis for every capability that needs a uniform identifier (dealings, earnings, scoring, taxonomy). Known securities come from ticker_reference (no live call). 404 when nothing matches.

Trigger a full ticker sync from the given origin.

Resumable earnings-calendar sync over all equity tickers (default types STOCK / CS / ADRC). For each ticker it runs the onvista stock-detail enrichment (which rebuilds calendar_events with upcoming results, dividends and annual general meetings) and records the response time in the sync_earnings ledger.

Tickers synced within the last min_days_between days (default 7) are skipped, so a re-run after a crash or after a couple of days processes only what is actually due — never the whole universe again.

Returns a job_id — poll GET /tickers/jobs/{job_id}.

Retrieve a single ticker from the reference collection.

Unlike the list response, this includes keywords and keywords_multilingual.

List recent wiki-sync jobs.

Get the current status of one wiki-sync job.

Request cancellation of a running wiki-sync job.

Create a background job that scans sec_filings.ticker_reference for rows with origin = "onvista.stock_csv", resolves Wikipedia/Wikidata content using name, isin, wkn, and home_symbol, and persists the result into the sec_filings.wiki collection.

This endpoint does not process the CSV file itself. It only works on rows that were already imported into ticker_reference by the Onvista CSV import flow.

The request returns immediately with a job document. Use GET /wiki/jobs/{job_id} to poll progress and POST /wiki/jobs/{job_id}/cancel to stop a running job.