changelog, pagination fix

Author: jonhealy1

CHANGELOG.md

@@ -8,6 +8,10 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
 
 ## [Unreleased]
 
+### Added
+
+- GET `/collections` collection search sort extension ex. `/collections?sortby=+id`. [#456](https://github.com/stac-utils/stac-fastapi-elasticsearch-opensearch/pull/456)
+
 ### Changed
 
 - updated `numReturned` & `numMatched` fields in itemCollection return to `numberReturned` & `numberMatched`. [#446](https://github.com/stac-utils/stac-fastapi-elasticsearch-opensearch/pull/446)

stac_fastapi/elasticsearch/stac_fastapi/elasticsearch/database_logic.py

@@ -187,31 +187,30 @@ async def get_all_collections(
         Returns:
             A tuple of (collections, next pagination token if any).
         """
-        search_after = token
-
-        formatted_sort = None
+        # Format the sort parameter
+        formatted_sort = []
         if sort:
-            formatted_sort = {}
             for item in sort:
                 field = item.get("field")
                 direction = item.get("direction", "asc")
                 if field:
-                    formatted_sort[field] = {"order": direction}
+                    formatted_sort.append({field: {"order": direction}})
             # Always include id as a secondary sort to ensure consistent pagination
-            formatted_sort.setdefault("id", {"order": "asc"})
-
-        # Use a collections-specific default sort that doesn't rely on properties.datetime
-        collections_default_sort = {"id": {"order": "asc"}}
+            if not any("id" in item for item in formatted_sort):
+                formatted_sort.append({"id": {"order": "asc"}})
+        else:
+            # Use a collections-specific default sort that doesn't rely on properties.datetime
+            formatted_sort = [{"id": {"order": "asc"}}]
 
         # Build the search body step by step to avoid type errors
         body = {
-            "sort": formatted_sort or collections_default_sort,
+            "sort": formatted_sort,
             "size": limit,
         }
 
         # Only add search_after if we have a token
-        if search_after is not None:
-            body["search_after"] = search_after  # type: ignore
+        if token:
+            body["search_after"] = [token]  # search_after must be a list
 
         response = await self.client.search(
             index=COLLECTIONS_INDEX,
@@ -228,7 +227,10 @@ async def get_all_collections(
 
         next_token = None
         if len(hits) == limit:
-            next_token = hits[-1]["sort"][0]
+            # Ensure we have a valid sort value for next_token
+            next_token_values = hits[-1].get("sort")
+            if next_token_values:
+                next_token = next_token_values[0]
 
         return collections, next_token
 

stac_fastapi/opensearch/stac_fastapi/opensearch/database_logic.py

@@ -159,8 +159,7 @@ async def get_all_collections(
         request: Request,
         sort: Optional[List[Dict[str, Any]]] = None,
     ) -> Tuple[List[Dict[str, Any]], Optional[str]]:
-        """
-        Retrieve a list of collections from Opensearch, supporting pagination.
+        """Retrieve a list of collections from Elasticsearch, supporting pagination.
 
         Args:
             token (Optional[str]): The pagination token.
@@ -171,7 +170,7 @@ async def get_all_collections(
         Returns:
             A tuple of (collections, next pagination token if any).
         """
-        collections_default_sort = [{"id": {"order": "asc"}}]
+        # Format the sort parameter
         formatted_sort = []
         if sort:
             for item in sort:
@@ -183,21 +182,22 @@ async def get_all_collections(
             if not any("id" in item for item in formatted_sort):
                 formatted_sort.append({"id": {"order": "asc"}})
         else:
-            formatted_sort = collections_default_sort
+            # Use a collections-specific default sort that doesn't rely on properties.datetime
+            formatted_sort = [{"id": {"order": "asc"}}]
 
-        search_body = {
+        # Build the search body step by step to avoid type errors
+        body = {
             "sort": formatted_sort,
             "size": limit,
         }
 
-        # Only add search_after to the query if token is not None and not empty
+        # Only add search_after if we have a token
         if token:
-            search_after = [token]
-            search_body["search_after"] = search_after
+            body["search_after"] = [token]  # search_after must be a list
 
         response = await self.client.search(
             index=COLLECTIONS_INDEX,
-            body=search_body,
+            body=body,
         )
 
         hits = response["hits"]["hits"]