fix: normalize search output

Author: gadomski

python/rustac/rustac.pyi

@@ -56,8 +56,8 @@ class DuckdbClient:
         filter: Optional[str | dict[str, Any]] = None,
         query: Optional[dict[str, Any]] = None,
         **kwargs: str,
-    ) -> dict[str, Any]:
-        """Search a stac-geoparquet file with duckdb, returning an item collection.
+    ) -> list[dict[str, Any]]:
+        """Search a stac-geoparquet file with duckdb, returning a list of items.
 
         Args:
             href: The stac-geoparquet file.
@@ -85,7 +85,7 @@ class DuckdbClient:
             kwargs: Additional parameters to pass in to the search.
 
         Returns:
-            A feature collection of STAC items.
+            A list of STAC items.
         """
 
     def search_to_arrow(
@@ -264,7 +264,7 @@ async def search(
     query: Optional[dict[str, Any]] = None,
     use_duckdb: Optional[bool] = None,
     **kwargs: str,
-) -> dict[str, Any]:
+) -> list[dict[str, Any]]:
     """
     Searches a STAC API server.
 
@@ -303,10 +303,10 @@ async def search(
         kwargs: Additional parameters to pass in to the search.
 
     Returns:
-        A feature collection of the returned STAC items.
+        STAC items
 
     Examples:
-        >>> item_collection = await rustac.search(
+        >>> items = await rustac.search(
         ...     "https://landsatlook.usgs.gov/stac-server",
         ...     collections=["landsat-c2l2-sr"],
         ...     intersects={"type": "Point", "coordinates": [-105.119, 40.173]},

src/duckdb.rs

@@ -59,7 +59,7 @@ impl DuckdbClient {
         filter: Option<StringOrDict>,
         query: Option<Bound<'py, PyDict>>,
         kwargs: Option<Bound<'py, PyDict>>,
-    ) -> Result<Bound<'py, PyDict>> {
+    ) -> Result<Bound<'py, PyList>> {
         let search = crate::search::build(
             intersects,
             ids,
@@ -81,9 +81,9 @@ impl DuckdbClient {
                 .map_err(|err| PyException::new_err(err.to_string()))?;
             client.search(&href, search)?
         };
-        let dict = pythonize::pythonize(py, &item_collection)?;
-        let dict = dict.extract()?;
-        Ok(dict)
+        let list = pythonize::pythonize(py, &item_collection.items)?;
+        let list = list.extract()?;
+        Ok(list)
     }
 
     #[pyo3(signature = (href, *, intersects=None, ids=None, collections=None, limit=None, bbox=None, datetime=None, include=None, exclude=None, sortby=None, filter=None, query=None, **kwargs))]

tests/test_duckdb.py

@@ -10,33 +10,24 @@ def client() -> DuckdbClient:
 
 
 def test_search(client: DuckdbClient) -> None:
-    item_collection = client.search("data/extended-item.parquet")
-    assert len(item_collection["features"]) == 1
+    items = client.search("data/extended-item.parquet")
+    assert len(items) == 1
 
 
 def test_search_offset(client: DuckdbClient) -> None:
-    item_collection = client.search(
-        "data/100-sentinel-2-items.parquet", offset=0, limit=1
-    )
-    assert (
-        item_collection["features"][0]["id"]
-        == "S2B_MSIL2A_20241203T174629_R098_T13TDE_20241203T211406"
-    )
-
-    item_collection = client.search(
-        "data/100-sentinel-2-items.parquet", offset=1, limit=1
-    )
-    assert (
-        item_collection["features"][0]["id"]
-        == "S2A_MSIL2A_20241201T175721_R141_T13TDE_20241201T213150"
-    )
+    items = client.search("data/100-sentinel-2-items.parquet", offset=0, limit=1)
+    assert items[0]["id"] == "S2B_MSIL2A_20241203T174629_R098_T13TDE_20241203T211406"
+
+    items = client.search("data/100-sentinel-2-items.parquet", offset=1, limit=1)
+    assert items[0]["id"] == "S2A_MSIL2A_20241201T175721_R141_T13TDE_20241201T213150"
 
 
 def test_get_collections(client: DuckdbClient) -> None:
     collections = client.get_collections("data/100-sentinel-2-items.parquet")
     assert len(collections) == 1
 
 
+@pytest.mark.skip("slow")
 def test_init_with_config() -> None:
     DuckdbClient(use_s3_credential_chain=True, use_hive_partitioning=True)