Glossary
Elasticsearch-Style Search API
Elasticsearch-style search API is a documented REST query endpoint backed by a search index. The catalog's example is EU CosIng, which exposes its cosmetics ingredient records — roughly 24,526 hits per itemType on broad single-letter queries — with INCI, CAS/EC, annex, function and SCCS fields.
What is Elasticsearch-style search API?
It is a query interface built on a search index rather than a relational endpoint: free-text and field-scoped queries return ranked hits as JSON. The cataloged instance is EU CosIng, 'searchable through a documented Elasticsearch-style REST API with INCI, CAS/EC, annex, function and SCCS fields', and tagged 'search api'.
Its scale is measurable from the API itself: 'Search API reports ~24,526 hits for a broad single-letter query on both itemType ingredient and substance; earlier plain-term queries reported 36,195 records.' Those two counts differ because query style changes hit counts — a property worth knowing before you quote a database size.
Why does Elasticsearch-style search API matter when choosing a dataset?
Search APIs trade completeness for convenience. You get precise, field-scoped lookups — an INCI name here, an annex restriction there — but no guarantee you have seen every record, which is why the coexisting nightly bulk dump matters for full copies.
Governance is the sharper issue. CosIng's terms note that 'the embedded search integration key is published openly in the app config but no formal API terms are posted; heavy programmatic use should be coordinated with DG GROW.' A technically accessible endpoint with no posted terms is not the same as a licensed feed, and hit counts that shift with query style (~24,526 versus 36,195) mean size claims need their method attached.
How do you evaluate Elasticsearch-style search API in a data source?
- Check the documented query fields. INCI, CAS/EC, annex, function and SCCS are the fields CosIng exposes — a wrapper should pass them through.
- Attach a method to every size claim. ~24,526 hits per itemType on broad single-letter queries versus 36,195 records from earlier plain-term queries are different measurements.
- Read the terms posture. No formal API terms are posted; coordinate heavy programmatic use with DG GROW rather than relying on the openly published config key.
- Line up the bulk alternative. For complete copies, the nightly-bulk-dump beats paginated search.
Category context: personal-care-products data.
Related terms
- nightly-bulk-dump — the completeness counterpart to query-based retrieval.
- Category context: personal-care-products data.
Frequently asked questions
How many records does the CosIng search API expose?
About 24,526 hits per itemType for a broad single-letter query on both ingredient and substance types; earlier plain-term queries reported 36,195 records.
Is the CosIng search API formally licensed?
No formal API terms are posted. The embedded key appears openly in the app config, and heavy programmatic use should be coordinated with DG GROW.
Should I use the search API or a bulk dump?
Search for targeted lookups; take the nightly bulk dump when you need the complete record set.
Datasets containing this field
Datasets containing Elasticsearch-Style Search API
6 datasets carry elasticsearch-style search api in the catalog. Open one, count the fields, judge for yourself.
COSDNA — Cosmetics Ingredient Analysis Database
EU CosIng — European Commission Cosmetic Ingredients Database
SCCS opinions · OJ publication flags …+9 more
INCIDecoder / INKEEDecoder — Ingredient Encyclopedia and Product INCI Database
ingredient_name · also_called · what_it_does …+5 more
Open Beauty Facts — Global Cosmetics Product Database (Exports + Live API)
Sephora — Structured Product Catalog (Scrapeable)
Every listing shows the field dictionary, sample rows, and coverage before you commit. API, files, or your warehouse. Daily, weekly, or hourly.
Get sample rows