Fix: search limit for items with same ids, different collections

CHANGELOG.md

@@ -55,8 +55,6 @@ and this project adheres to [Semantic Versioning](http://semver.org/).
   toolchain, for production deployments where the Rust build environment is not needed.
 - Dependabot coverage expanded to Docker base images and pip packages (two new
   ecosystems with grouped update policies).
-
-### Changed
 - `pypgstac migrate` now delegates runtime migration planning and apply logic to
   `pgstac-migrate`; `src/pypgstac/src/pypgstac/migrate.py` remains as a
   compatibility wrapper.
@@ -111,6 +109,8 @@ and this project adheres to [Semantic Versioning](http://semver.org/).
   `python-runtime` groups).
 - `docker-compose.yml` removes explicit `container_name` entries to avoid conflicts
   between concurrent local instances.
+- Appended `collection` as a secondary sorting tie-breaker before `id` inside `keyset_sortkeys`. This ensures deterministic pagination tokens globally and enables PostgreSQL partition pruning for optimized multi-collection searches.
+- Updated pgTAP test suite expectations and output snapshots to match the new `collection, id` sorting structure.
 
 ### Removed
 - PL/Rust support: `pgstacbase-plrust` and `pgstac-plrust` Docker targets removed; the
@@ -135,6 +135,8 @@ and this project adheres to [Semantic Versioning](http://semver.org/).
 - `scripts/container-scripts/test` now derives the active database from
   `PGDATABASE`/`POSTGRES_DB` when checking server extensions and refreshing
   collation versions, instead of assuming `postgis`.
+- Fixed an edge-case bug where keyset pagination returned 0 features on Page 2 if items across different collections shared identical IDs and timestamps (#392).
+- Resolved a Python type-checking diagnostic in `pypgstac.load` regarding dynamic file opening modes.
 
 
 ## [v0.9.11]

src/pgstac/pgstac.sql

@@ -2293,7 +2293,7 @@ BEGIN
     END IF;
 
     IF jsonb_typeof(j) = 'object' THEN
-        -- GeoJSON geometry args (Point/Polygon/.../GeometryCollection) are not cql1 expressions;
+        -- GeoJSON geometry args (Point/Polygon/.../GeometryCollection) are not cql2 expressions;
         -- pass them through unchanged so spatial ops keep their geometry intact.
         IF j ? 'type' AND (j ? 'coordinates' OR j ? 'geometries') THEN
             RETURN j;
@@ -2870,9 +2870,10 @@ RETURNS TABLE(ord int, field text, expr text, dir text, notnull boolean) AS $$
         SELECT coalesce(_search->'sortby','[{"field":"datetime","direction":"desc"}]'::jsonb) AS s
     ),
     app AS (
+        -- Append collection first (helps partition pruning), then id to guarantee global uniqueness
         SELECT s
-               || jsonb_build_object('field','id','direction', s->0->>'direction')
-               || jsonb_build_object('field','collection','direction', s->0->>'direction') AS s
+               || jsonb_build_object('field','collection','direction', s->0->>'direction')
+               || jsonb_build_object('field','id','direction', s->0->>'direction') AS s
         FROM base
     ),
     rows AS (
@@ -5829,7 +5830,6 @@ BEGIN
     END LOOP;
 END;
 $$ LANGUAGE PLPGSQL;
-
 -- END FRAGMENT: 997_maintenance.sql
 
 -- BEGIN FRAGMENT: 998_idempotent_post.sql

src/pgstac/sql/002d_keyset.sql

@@ -26,9 +26,10 @@ RETURNS TABLE(ord int, field text, expr text, dir text, notnull boolean) AS $$
         SELECT coalesce(_search->'sortby','[{"field":"datetime","direction":"desc"}]'::jsonb) AS s
     ),
     app AS (
+        -- Append collection first (helps partition pruning), then id to guarantee global uniqueness
         SELECT s
-               || jsonb_build_object('field','id','direction', s->0->>'direction')
-               || jsonb_build_object('field','collection','direction', s->0->>'direction') AS s
+               || jsonb_build_object('field','collection','direction', s->0->>'direction')
+               || jsonb_build_object('field','id','direction', s->0->>'direction') AS s
         FROM base
     ),
     rows AS (

src/pgstac/tests/basic/collection_searches.sql.out

@@ -75,10 +75,11 @@ SELECT page, ids FROM pg_temp.collwalk('{"ids":["testcollection_1","testcollecti
     2 | ["testcollection_1"]
 
 select collection_search('{"filter":{"op":"eq", "args":[{"property":"title"},"My Test Collection."]},"limit":10, "sortby":[{"field":"datetime","direction":"desc"}]}');
- {"links": [{"rel": "next", "href": "./collections?token=next:MjAxMi0wMy0zMSAwMDowMDowMCswMB90ZXN0Y29sbGVjdGlvbl82MzkfY29sbGVjdGlvbnM=", "type": "application/json", "method": "GET"}], "collections": [{"id": "testcollection_648", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[170, 80, 180, 90]]}, "temporal": {"interval": [["2012-06-02 00:00:00+00", "2012-08-02 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_647", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[160, 80, 170, 90]]}, "temporal": {"interval": [["2012-05-26 00:00:00+00", "2012-07-26 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_646", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[150, 80, 160, 90]]}, "temporal": {"interval": [["2012-05-19 00:00:00+00", "2012-07-19 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_645", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[140, 80, 150, 90]]}, "temporal": {"interval": [["2012-05-12 00:00:00+00", "2012-07-12 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_644", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[130, 80, 140, 90]]}, "temporal": {"interval": [["2012-05-05 00:00:00+00", "2012-07-05 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_643", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[120, 80, 130, 90]]}, "temporal": {"interval": [["2012-04-28 00:00:00+00", "2012-06-28 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_642", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[110, 80, 120, 90]]}, "temporal": {"interval": [["2012-04-21 00:00:00+00", "2012-06-21 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_641", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[100, 80, 110, 90]]}, "temporal": {"interval": [["2012-04-14 00:00:00+00", "2012-06-14 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_640", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[90, 80, 100, 90]]}, "temporal": {"interval": [["2012-04-07 00:00:00+00", "2012-06-07 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_639", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[80, 80, 90, 90]]}, "temporal": {"interval": [["2012-03-31 00:00:00+00", "2012-05-31 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}], "numberMatched": 648, "numberReturned": 10}
+ {"links": [{"rel": "next", "href": "./collections?token=next:MjAxMi0wMy0zMSAwMDowMDowMCswMB9jb2xsZWN0aW9ucx90ZXN0Y29sbGVjdGlvbl82Mzk=", "type": "application/json", "method": "GET"}], "collections": [{"id": "testcollection_648", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[170, 80, 180, 90]]}, "temporal": {"interval": [["2012-06-02 00:00:00+00", "2012-08-02 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_647", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[160, 80, 170, 90]]}, "temporal": {"interval": [["2012-05-26 00:00:00+00", "2012-07-26 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_646", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[150, 80, 160, 90]]}, "temporal": {"interval": [["2012-05-19 00:00:00+00", "2012-07-19 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_645", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[140, 80, 150, 90]]}, "temporal": {"interval": [["2012-05-12 00:00:00+00", "2012-07-12 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_644", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[130, 80, 140, 90]]}, "temporal": {"interval": [["2012-05-05 00:00:00+00", "2012-07-05 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_643", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[120, 80, 130, 90]]}, "temporal": {"interval": [["2012-04-28 00:00:00+00", "2012-06-28 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_642", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[110, 80, 120, 90]]}, "temporal": {"interval": [["2012-04-21 00:00:00+00", "2012-06-21 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_641", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[100, 80, 110, 90]]}, "temporal": {"interval": [["2012-04-14 00:00:00+00", "2012-06-14 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_640", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[90, 80, 100, 90]]}, "temporal": {"interval": [["2012-04-07 00:00:00+00", "2012-06-07 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_639", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[80, 80, 90, 90]]}, "temporal": {"interval": [["2012-03-31 00:00:00+00", "2012-05-31 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}], "numberMatched": 648, "numberReturned": 10}
 
 select collection_search('{"datetime":["2012-01-01","2012-01-02"], "filter":{"op":"eq", "args":[{"property":"title"},"My Test Collection."]},"limit":10, "sortby":[{"field":"datetime","direction":"desc"}]}');
  {"links": [], "collections": [{"id": "testcollection_626", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[-50, 80, -40, 90]]}, "temporal": {"interval": [["2011-12-31 00:00:00+00", "2012-02-29 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_625", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[-60, 80, -50, 90]]}, "temporal": {"interval": [["2011-12-24 00:00:00+00", "2012-02-24 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_624", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[-70, 80, -60, 90]]}, "temporal": {"interval": [["2011-12-17 00:00:00+00", "2012-02-17 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_623", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[-80, 80, -70, 90]]}, "temporal": {"interval": [["2011-12-10 00:00:00+00", "2012-02-10 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_622", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[-90, 80, -80, 90]]}, "temporal": {"interval": [["2011-12-03 00:00:00+00", "2012-02-03 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_621", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[-100, 80, -90, 90]]}, "temporal": {"interval": [["2011-11-26 00:00:00+00", "2012-01-26 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_620", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[-110, 80, -100, 90]]}, "temporal": {"interval": [["2011-11-19 00:00:00+00", "2012-01-19 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_619", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[-120, 80, -110, 90]]}, "temporal": {"interval": [["2011-11-12 00:00:00+00", "2012-01-12 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}, {"id": "testcollection_618", "type": "Collection", "title": "My Test Collection.", "extent": {"spatial": {"bbox": [[-130, 80, -120, 90]]}, "temporal": {"interval": [["2011-11-05 00:00:00+00", "2012-01-05 00:00:00+00"]]}}, "description": "Description of my test collection.", "stac_extensions": []}], "numberMatched": 9, "numberReturned": 9}
 
 select collection_search('{"ids":["testcollection_1","testcollection_2"], "fields": {"include": ["title"]}}');
  {"links": [], "collections": [{"id": "testcollection_1", "title": "My Test Collection."}, {"id": "testcollection_2", "title": "My Test Collection."}], "numberMatched": 2, "numberReturned": 2}
+