core-lens 0.1.dev129__tar.gz → 0.1.dev130__tar.gz

{core_lens-0.1.dev129 → core_lens-0.1.dev130}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: core-lens
-Version: 0.1.dev129
+Version: 0.1.dev130
 Summary: Query, analyse, and visualise CoreStack's microwatershed and Earth science data through a clean, composable Python API.
 Project-URL: Homepage, https://github.com/ApoorvaKashyap/core-lens
 Project-URL: Issues, https://github.com/ApoorvaKashyap/core-lens/issues

{core_lens-0.1.dev129 → core_lens-0.1.dev130}/src/core_lens/_version.py

@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
 commit_id: str | None
 __commit_id__: str | None
 
-__version__ = version = '0.1.dev129'
-__version_tuple__ = version_tuple = (0, 1, 'dev129')
+__version__ = version = '0.1.dev130'
+__version_tuple__ = version_tuple = (0, 1, 'dev130')
 
 __commit_id__ = commit_id = None

{core_lens-0.1.dev129 → core_lens-0.1.dev130}/src/core_lens/base/result.py

@@ -100,10 +100,12 @@ class Result:
                 "Call .with_geometry() first to join the static geometry column before extracting GeoDataFrame."
             )
         import geopandas as gpd
-        import shapely.wkb as wkb
+        import shapely
 
         geometry_col = self.entity.geometry_col
-        geometries = [wkb.loads(b) for b in self.data[geometry_col].to_list()]
+        # Vectorised C-level decode — shapely.from_wkb operates on the whole
+        # numpy array at once, avoiding a Python-loop per row.
+        geometries = shapely.from_wkb(self.data[geometry_col].to_numpy())
         geo_series = gpd.GeoSeries(geometries, crs="EPSG:4326")
         return gpd.GeoDataFrame(
             self.data.drop(geometry_col).to_pandas(),

{core_lens-0.1.dev129 → core_lens-0.1.dev130}/src/core_lens/export/formats.py

@@ -162,12 +162,17 @@ def geoparquet(result: "Result", path: str | pathlib.Path, **kwargs: Any) -> Non
 def geojson(result: "Result", path: str | pathlib.Path, **kwargs: Any) -> None:
     """Export the result data to a GeoJSON file.
 
-    This function uses DuckDB to write the data frame with spatial extensions.
+    This function uses GeoPandas to write the data frame.  It prefers the
+    ``pyogrio`` engine (faster, C-backed) when available and falls back to
+    ``fiona`` automatically.  The ``engine`` kwarg can override this:
+    ``engine="fiona"`` or ``engine="pyogrio"``.
 
     Args:
         result (Result): The Result object to export.
         path (str | pathlib.Path): Destination path for the GeoJSON file.
-        **kwargs (Any): Additional options to pass to DuckDB's COPY statement.
+        **kwargs (Any): Additional options forwarded to
+            ``GeoDataFrame.to_file()`` (e.g. ``driver="GeoJSON"``,
+            ``engine="pyogrio"``).
 
     Example:
         >>> from core_lens.export import geojson
@@ -182,37 +187,20 @@ def geojson(result: "Result", path: str | pathlib.Path, **kwargs: Any) -> None:
             "Call .with_geometry() first to join the static geometry column before exporting to geojson."
         )
 
-    df = result.df()
-    geom_col = result.entity.geometry_col
-
-    cols = []
-    for c in df.columns:
-        if c == geom_col:
-            cols.append(f'ST_GeomFromWKB("{c}") AS "{c}"')
-        else:
-            cols.append(f'"{c}"')
-
-    select_clause = ", ".join(cols)
-
-    conn = _get_duckdb_conn()
+    gdf = result.gdf()
 
-    path_str = str(path).replace("'", "''")
+    # Resolve write engine: prefer pyogrio (C-backed, no GDAL Python overhead),
+    # fall back to fiona if not installed.
+    engine = kwargs.pop("engine", None)
+    if engine is None:
+        try:
+            import pyogrio as _  # type: ignore[import-untyped]  # noqa: F401
 
-    options = ["FORMAT GDAL", "DRIVER 'GeoJSON'"]
-    for k, v in kwargs.items():
-        if isinstance(v, bool):
-            options.append(f"{k.upper()} {'TRUE' if v else 'FALSE'}")
-        elif isinstance(v, str):
-            options.append(f"{k.upper()} '{v}'")
-        else:
-            options.append(f"{k.upper()} {v}")
+            engine = "pyogrio"
+        except ModuleNotFoundError:
+            engine = "fiona"
 
-    options_str = ", ".join(options)
+    # Always write as GeoJSON unless the caller overrides driver.
+    kwargs.setdefault("driver", "GeoJSON")
 
-    query = f"""
-    COPY (
-        SELECT {select_clause}
-        FROM df
-    ) TO '{path_str}' ({options_str});
-    """
-    conn.execute(query)
+    gdf.to_file(path, engine=engine, **kwargs)

{core_lens-0.1.dev129 → core_lens-0.1.dev130}/tests/unit/test_export.py

@@ -152,8 +152,10 @@ def test_geoparquet_options_partition(
 
 
 def test_geojson_options(sample_result: Result, tmp_path: pathlib.Path) -> None:
+    """Unknown kwargs are forwarded to the write engine; unrecognised ones raise
+    ValueError (pyogrio) rather than silently being ignored.  The test simply
+    verifies the call path is exercised without an unexpected exception type."""
     result_with_geom = sample_result.with_geometry()
-    import duckdb
 
     out_path = tmp_path / "out_opts.json"
     try:
@@ -165,5 +167,6 @@ def test_geojson_options(sample_result: Result, tmp_path: pathlib.Path) -> None:
             flag=False,
             number=10,
         )
-    except duckdb.BinderException:
+    except ValueError:
+        # pyogrio rejects unrecognised driver options — that's expected.
         pass