fix: permissive extent parsing for collections (#787)

* fix: permissive extent parsing for collections

* fix: cli collections dump, no tests

* chore: update changelog

Author: gadomski

CHANGELOG.md

@@ -15,6 +15,7 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
 ### Fixed
 
 - `Client.get_collection` for static catalogs [#782](https://github.com/stac-utils/pystac-client/pull/782)
+- Permissive collection reading [#787](https://github.com/stac-utils/pystac-client/pull/787)
 
 ## [v0.8.5] - 2024-10-23
 

pystac_client/cli.py

@@ -124,7 +124,7 @@ def collections(
             else:
                 raise KeyError("'matched' is not supported for this catalog")
         else:
-            collections_dicts = [c.to_dict() for c in result.collections()]
+            collections_dicts = list(result.collections_as_dicts())
             if save:
                 with open(save, "w") as f:
                     f.write(json.dumps(collections_dicts))
@@ -257,8 +257,7 @@ def parse_args(args: list[str]) -> dict[str, Any]:
     )
     collections_group.add_argument(
         "--datetime",
-        help="Single datetime or begin and end datetime "
-        "(e.g., 2017-01-01/2017-02-15)",
+        help="Single datetime or begin and end datetime (e.g., 2017-01-01/2017-02-15)",
     )
     collections_group.add_argument("--q", help="Free-text search query")
     collections_group.add_argument(
@@ -318,8 +317,7 @@ def parse_args(args: list[str]) -> dict[str, Any]:
     )
     search_group.add_argument(
         "--datetime",
-        help="Single datetime or begin and end datetime "
-        "(e.g., 2017-01-01/2017-02-15)",
+        help="Single datetime or begin and end datetime (e.g., 2017-01-01/2017-02-15)",
     )
     search_group.add_argument(
         "--query",

pystac_client/collection_search.py

@@ -8,7 +8,7 @@
     Optional,
 )
 
-from pystac import Collection, TemporalExtent
+from pystac import Collection, Extent
 
 from pystac_client._utils import Modifiable, call_modifier
 from pystac_client.conformance import ConformanceClasses
@@ -26,7 +26,7 @@
     SortbyLike,
 )
 from pystac_client.stac_api_io import StacApiIO
-from pystac_client.warnings import DoesNotConformTo
+from pystac_client.warnings import DoesNotConformTo, PystacClientWarning
 
 if TYPE_CHECKING:
     from pystac_client import client as _client
@@ -54,26 +54,21 @@ def bboxes_overlap(bbox1: BBox, bbox2: BBox) -> bool:
     return xmin1 <= xmax2 and xmin2 <= xmax1 and ymin1 <= ymax2 and ymin2 <= ymax1
 
 
-def collection_matches(
-    collection_dict: dict[str, Any],
+def _extent_matches(
+    extent: Extent,
     bbox: BBox | None = None,
     temporal_interval_str: Datetime | None = None,
-    q: str | None = None,
-) -> bool:
-    # check for overlap between provided bbox and the collection's spatial extent
-    collection_bboxes = collection_dict["extent"]["spatial"]["bbox"]
+) -> tuple[bool, bool]:
     bbox_overlaps = not bbox or (
         any(
-            bboxes_overlap(bbox, collection_bbox)
-            for collection_bbox in collection_bboxes
+            bboxes_overlap(bbox, tuple(collection_bbox))
+            for collection_bbox in extent.spatial.bboxes
         )
     )
 
     # check for overlap between the provided temporal interval and the collection's
     # temporal extent
-    collection_temporal_extent = TemporalExtent.from_dict(
-        collection_dict["extent"]["temporal"]
-    )
+    collection_temporal_extent = extent.temporal
 
     # process the user-provided temporal interval
     search_temporal_interval = (
@@ -110,6 +105,29 @@ def collection_matches(
             for collection_temporal_interval in collection_temporal_extent.intervals
         )
     )
+    return bbox_overlaps, datetime_overlaps
+
+
+def collection_matches(
+    collection_dict: dict[str, Any],
+    bbox: BBox | None = None,
+    temporal_interval_str: Datetime | None = None,
+    q: str | None = None,
+) -> bool:
+    # check for overlap between provided bbox and the collection's spatial extent
+    try:
+        extent = Extent.from_dict(collection_dict.get("extent", {}))
+    except Exception:
+        warnings.warn(
+            f"Unable to parse extent from collection={collection_dict.get('id', None)}",
+            PystacClientWarning,
+        )
+        bbox_overlaps = True
+        datetime_overlaps = True
+    else:
+        bbox_overlaps, datetime_overlaps = _extent_matches(
+            extent, bbox, temporal_interval_str
+        )
 
     # check for overlap between the provided free-text search query (q) and the
     # collection's title, description, and keywords

tests/data/invalid-collection.json

@@ -0,0 +1,37 @@
+{
+    "id": "planet",
+    "title": "Planet",
+    "extent": {},
+    "license": "proprietary",
+    "description": "Planet",
+    "stac_version": "1.0.0",
+    "item_type_accessor": "pl:item_type",
+    "links": [
+        {
+            "rel": "items",
+            "type": "application/geo+json",
+            "href": "https://csdap.earthdata.nasa.gov/stac/collections/planet/items"
+        },
+        {
+            "rel": "parent",
+            "type": "application/json",
+            "href": "https://csdap.earthdata.nasa.gov/stac/"
+        },
+        {
+            "rel": "root",
+            "type": "application/json",
+            "href": "https://csdap.earthdata.nasa.gov/stac/"
+        },
+        {
+            "rel": "self",
+            "type": "application/json",
+            "href": "https://csdap.earthdata.nasa.gov/stac/collections/planet"
+        },
+        {
+            "rel": "http://www.opengis.net/def/rel/ogc/1.0/queryables",
+            "type": "application/schema+json",
+            "title": "Queryables",
+            "href": "https://csdap.earthdata.nasa.gov/stac/collections/planet/queryables"
+        }
+    ]
+}

tests/test_collection_search.py

@@ -1,3 +1,4 @@
+import json
 from datetime import datetime
 from math import ceil
 from typing import Any
@@ -12,8 +13,10 @@
 from pystac_client.collection_search import (
     CollectionSearch,
     bboxes_overlap,
+    collection_matches,
     temporal_intervals_overlap,
 )
+from pystac_client.warnings import PystacClientWarning
 
 from .helpers import STAC_URLS, read_data_file
 
@@ -387,3 +390,10 @@ def test_temporal_intervals_overlap() -> None:
             None,
         ),
     )
+
+
+def test_invalid_collection() -> None:
+    # https://github.com/stac-utils/pystac-client/issues/786
+    data = json.loads(read_data_file("invalid-collection.json"))
+    with pytest.warns(PystacClientWarning):
+        collection_matches(data)