Docs

README.md

@@ -4,7 +4,7 @@
 
 ## What is sedr?
 
-An experimental validator for OGC EDR APIs using schemathesis. Main focus will be on the Rodeo Profile, which is a subset of the OGC EDR API.
+An experimental validator for OGC EDR APIs. Main focus will be on the Rodeo Profile, which is a subset of the OGC EDR API.
 
 ## Who is responsible?
 
@@ -40,6 +40,10 @@ Run manually as noted in [Test it out](#test-it-out), or add it to your CI using
 
 ## Overview of architecture
 
+- __init__ includes tests from ogcapi, edrreq and rodeoprofile at startup. Tests are categorized as landing, conformance and collection.
+- Landing and conformance tests are run first, in the preflight phase.
+- Then schemathesis will validate the OpenAPI spec and run lots of automatic tests, including fuzzing of query parameters. Collection tests are run during this phase.
+
 ## Documentation
 
 ### Limitations
@@ -48,9 +52,13 @@ Run manually as noted in [Test it out](#test-it-out), or add it to your CI using
 - Assuming OGC EDR API version 1.2 (draft)
 - Few, basic tests for now
 - Will focus more on profiles (limitations within the EDR spec) like <https://github.com/EURODEO/rodeo-edr-profile> than the full EDR spec.
-- Use --rodeo-profile to force a test against the profile (will happen automatically if conformance includes the profile)
+- Use --rodeo-profile to force a test against the profile
+
+### Testing the sedr code to look for regressions
+
+For development, source a venv and run `tox p` to run all tests.
 
-### Understanding errors
+### Understanding errors from schemathesis
 
 For each "FAILED" line, you can scroll back to see the full error and, if relevant, with a curl-example to reproduce it.
 
@@ -96,7 +104,7 @@ ERROR sedr/schemat.py - schemathesis.exceptions.SchemaError: Failed to load sche
 
 #### Wrong API version / missing conformance link
 
-Sedr supports EDR 1.1, but the API is EDR 1.0.
+Sedr wants EDR 1.1, but the API is EDR 1.0.
 
 ```python
         if not requirementA2_2_A5:

sedr/schemat.py

@@ -115,10 +115,6 @@ def test_edr_collections(case):
             "test_collections found collection id %s", collection_json["id"]
         )
 
-        # Validation done in requrementA5_2
-        extent = util.parse_spatial_bbox(collection_json)
-        extents[collection_url] = tuple(extent[0])
-
         # Run edr, ogc, profile tests
         for f in util.test_functions["collection"]:
             status, msg = f(collection_json)
@@ -127,6 +123,10 @@ def test_edr_collections(case):
                 raise AssertionError(f"Test {f.__name__} failed with message: {msg}")
             util.logger.info("Test %s passed. (%s)", f.__name__, msg)
 
+        # Validation of spatial_bbox done above
+        extent = util.parse_spatial_bbox(collection_json)
+        extents[collection_url] = tuple(extent[0])
+
 
 for p in schema.raw_schema["paths"].keys():
     # Optionally include endpoints if they exist, otherwise schemathesis will refuse to run