openEO Back-End Extractor
Pricing
from $0.35 / 1,000 record extracteds
openEO Back-End Extractor
Point at ANY openEO back-end (the standard API for Earth-observation cloud back-ends — VITO/Terrascope, Copernicus Data Space, EODC). Extract collections, the process registry, capabilities, file formats, service types or UDF runtimes as flat rows + raw JSON. No auth. Pay per record.
Pricing
from $0.35 / 1,000 record extracteds
Rating
0.0
(0)
Developer
Datamule
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
8 days ago
Last modified
Categories
Share
A generic "point at any openEO back-end" extractor. openEO is the open, uniform REST API for Earth-observation cloud back-ends — the standard that lets one client talk to VITO/Terrascope, the Copernicus Data Space Ecosystem, EODC and every other openEO-conformant processing back-end the same way, over satellite data cubes (Sentinel-1/2/3, Landsat, and hundreds of derived collections).
Because the grammar is uniform across operators, one actor spans them all.
Point it at a back-end URL, pick a read mode, and it flattens that back-end's
public discovery documents into clean tabular rows — with the same normalized
shape regardless of the operator — plus a lossless raw copy of every record.
No per-operator scraper, no authentication.
How it resolves a back-end
openEO back-ends advertise themselves at {host}/.well-known/openeo — a list of
API versions, each with a production flag and a versioned base URL. Give the actor
the host root (e.g. https://openeo.vito.be) and it resolves the well-known
document to the highest production version base for you, then reads the core
discovery endpoints against it. You can also pass an already-versioned base
(e.g. https://openeo.terrascope.be/openeo/1.2) and skip that step, or pin a
specific version with apiVersion.
Seven read modes
| Mode | What you get (one flat row per…) |
|---|---|
| collections (default) | …EO data collection (STAC-flavored) — id, title, description, license, keywords, extentSpatial (bbox), extentTemporal (interval), providers, raw |
| processes | …process in the back-end's registry — id, title (summary), categories, parameterCount, parameters, returns, deprecated, experimental, raw |
| capabilities | the landing document (id, title, api/backend/stac versions, production, advertised endpoints) plus one row per advertised API version |
| collection-detail | …requested collection id — the full STAC Collection incl. cubeDimensions, summaries, providers, extents |
| file-formats | …supported input and output file format — id, direction, title, gisDataTypes |
| service-types | …secondary web service type the back-end can publish (WMTS/XYZ/…). Optional in the spec — processing-only back-ends have none. |
| udf-runtimes | …user-defined-function runtime (Python/R/…) — id, runtime url (type), available versions in keywords |
Every row also carries provenance (_backend, _mode, _type) and the lossless
raw source object.
Input
{"backend": "https://openeo.dataspace.copernicus.eu","mode": "collections","collectionId": [],"maxItems": 10000}
- backend (required) — the URL of any openEO back-end. Host root preferred
(the actor resolves
/.well-known/openeo); a versioned base also works. - mode — one of the seven above. Default
collections. - collectionId — only used by
collection-detailmode; the collection id(s) to fetch full detail for. Runcollectionsfirst to discover the exact ids. - apiVersion — optionally pin the openEO API version (e.g.
1.2.0); defaults to the highest production version advertised. - maxItems — total row cap (each row is one billable event). Default 10000.
- bearerToken / extraHeaders — optional; not required — every read mode targets public, unauthenticated discovery endpoints. Never logged.
Output (collections mode, example row)
{"_backend": "https://openeo.dataspace.copernicus.eu/openeo/1.2","_mode": "collections","_type": "collection","id": "SENTINEL3_OLCI_L1B","title": "Sentinel 3 OLCI","license": "proprietary","keywords": null,"extentSpatial": [[-180.0, -85.0, 180.0, 85.0]],"extentTemporal": [["2016-04-17T11:33:13Z", null]],"stacVersion": "1.0.0","raw": { "…": "the full STAC Collection" }}
Behaviour & honesty
- Version resolution is automatic. The well-known document is parsed to the best production version base; an already-versioned base is used as-is.
- STAC collections are flattened (id/title/description/license/extents) with
the full STAC Collection preserved in
raw— nothing is lost. - Pagination handled.
collectionsandprocessesauto-follow the top-levelrel="next"link (relative or absolute) until exhausted or the cap is reached. - Optional endpoints degrade gracefully.
service_types/udf_runtimesare optional in the openEO spec; a back-end that doesn't offer one is reported, not crashed. - Zero records is never faked. If the chosen mode returns nothing (an optional endpoint the back-end doesn't serve, or an unknown collection id), the run exits non-zero instead of emitting fabricated rows.
Pricing
Pay-per-event: one record charge per row returned.
Verified
Tested live against three independent operators — VITO/Terrascope
(openeo.vito.be → 89 collections / 150 processes), Copernicus Data Space
Ecosystem (openeo.dataspace.copernicus.eu → 83 / 143) and EODC
(openeo.eodc.eu) — confirming identical normalized output across back-ends.
