pyogrio 0.9.0__cp39-cp39-macosx_12_0_x86_64.whl

pyogrio/tests/conftest.py

@@ -0,0 +1,204 @@
+from pathlib import Path
+from zipfile import ZIP_DEFLATED, ZipFile
+
+import pytest
+
+from pyogrio import (
+    __gdal_version_string__,
+    __version__,
+    list_drivers,
+)
+from pyogrio._compat import (
+    HAS_ARROW_API,
+    HAS_ARROW_WRITE_API,
+    HAS_GDAL_GEOS,
+    HAS_PYARROW,
+    HAS_SHAPELY,
+)
+from pyogrio.raw import read, write
+
+_data_dir = Path(__file__).parent.resolve() / "fixtures"
+
+# mapping of driver extension to driver name for well-supported drivers
+DRIVERS = {
+    ".fgb": "FlatGeobuf",
+    ".geojson": "GeoJSON",
+    ".geojsonl": "GeoJSONSeq",
+    ".geojsons": "GeoJSONSeq",
+    ".gpkg": "GPKG",
+    ".shp": "ESRI Shapefile",
+}
+
+# mapping of driver name to extension
+DRIVER_EXT = {driver: ext for ext, driver in DRIVERS.items()}
+
+ALL_EXTS = [".fgb", ".geojson", ".geojsonl", ".gpkg", ".shp"]
+
+
+def pytest_report_header(config):
+    drivers = ", ".join(
+        f"{driver}({capability})"
+        for driver, capability in sorted(list_drivers().items())
+    )
+    return (
+        f"pyogrio {__version__}\n"
+        f"GDAL {__gdal_version_string__}\n"
+        f"Supported drivers: {drivers}"
+    )
+
+
+# marks to skip tests if optional dependecies are not present
+requires_arrow_api = pytest.mark.skipif(not HAS_ARROW_API, reason="GDAL>=3.6 required")
+requires_pyarrow_api = pytest.mark.skipif(
+    not HAS_ARROW_API or not HAS_PYARROW, reason="GDAL>=3.6 and pyarrow required"
+)
+
+requires_arrow_write_api = pytest.mark.skipif(
+    not HAS_ARROW_WRITE_API or not HAS_PYARROW,
+    reason="GDAL>=3.8 required for Arrow write API",
+)
+
+requires_gdal_geos = pytest.mark.skipif(
+    not HAS_GDAL_GEOS, reason="GDAL compiled with GEOS required"
+)
+
+requires_shapely = pytest.mark.skipif(not HAS_SHAPELY, reason="Shapely >= 2.0 required")
+
+
+def prepare_testfile(testfile_path, dst_dir, ext):
+    if ext == testfile_path.suffix:
+        return testfile_path
+
+    dst_path = dst_dir / f"{testfile_path.stem}{ext}"
+    if dst_path.exists():
+        return dst_path
+
+    meta, _, geometry, field_data = read(testfile_path)
+
+    if ext == ".fgb":
+        # For .fgb, spatial_index=False to avoid the rows being reordered
+        meta["spatial_index"] = False
+        # allow mixed Polygons/MultiPolygons type
+        meta["geometry_type"] = "Unknown"
+
+    elif ext == ".gpkg":
+        # For .gpkg, spatial_index=False to avoid the rows being reordered
+        meta["spatial_index"] = False
+        meta["geometry_type"] = "MultiPolygon"
+
+    write(dst_path, geometry, field_data, **meta)
+    return dst_path
+
+
+@pytest.fixture(scope="session")
+def data_dir():
+    return _data_dir
+
+
+@pytest.fixture(scope="function")
+def naturalearth_lowres(tmp_path, request):
+    ext = getattr(request, "param", ".shp")
+    testfile_path = _data_dir / Path("naturalearth_lowres/naturalearth_lowres.shp")
+
+    return prepare_testfile(testfile_path, tmp_path, ext)
+
+
+@pytest.fixture(scope="function", params=ALL_EXTS)
+def naturalearth_lowres_all_ext(tmp_path, naturalearth_lowres, request):
+    return prepare_testfile(naturalearth_lowres, tmp_path, request.param)
+
+
+@pytest.fixture(scope="function")
+def naturalearth_lowres_vsi(tmp_path, naturalearth_lowres):
+    """Wrap naturalearth_lowres as a zip file for vsi tests"""
+
+    path = tmp_path / f"{naturalearth_lowres.name}.zip"
+    with ZipFile(path, mode="w", compression=ZIP_DEFLATED, compresslevel=5) as out:
+        for ext in ["dbf", "prj", "shp", "shx", "cpg"]:
+            filename = f"{naturalearth_lowres.stem}.{ext}"
+            out.write(naturalearth_lowres.parent / filename, filename)
+
+    return path, f"/vsizip/{path}/{naturalearth_lowres.name}"
+
+
+@pytest.fixture(scope="session")
+def test_fgdb_vsi():
+    return f"/vsizip/{_data_dir}/test_fgdb.gdb.zip"
+
+
+@pytest.fixture(scope="session")
+def test_gpkg_nulls():
+    return _data_dir / "test_gpkg_nulls.gpkg"
+
+
+@pytest.fixture(scope="session")
+def test_ogr_types_list():
+    return _data_dir / "test_ogr_types_list.geojson"
+
+
+@pytest.fixture(scope="session")
+def test_datetime():
+    return _data_dir / "test_datetime.geojson"
+
+
+@pytest.fixture(scope="session")
+def test_datetime_tz():
+    return _data_dir / "test_datetime_tz.geojson"
+
+
+@pytest.fixture(scope="function")
+def geojson_bytes(tmp_path):
+    """Extracts first 3 records from naturalearth_lowres and writes to GeoJSON,
+    returning bytes"""
+    meta, _, geometry, field_data = read(
+        _data_dir / Path("naturalearth_lowres/naturalearth_lowres.shp"), max_features=3
+    )
+
+    filename = tmp_path / "test.geojson"
+    write(filename, geometry, field_data, **meta)
+
+    with open(filename, "rb") as f:
+        bytes_buffer = f.read()
+
+    return bytes_buffer
+
+
+@pytest.fixture(scope="function")
+def geojson_filelike(tmp_path):
+    """Extracts first 3 records from naturalearth_lowres and writes to GeoJSON,
+    returning open file handle"""
+    meta, _, geometry, field_data = read(
+        _data_dir / Path("naturalearth_lowres/naturalearth_lowres.shp"), max_features=3
+    )
+
+    filename = tmp_path / "test.geojson"
+    write(filename, geometry, field_data, layer="test", **meta)
+
+    with open(filename, "rb") as f:
+        yield f
+
+
+@pytest.fixture(
+    scope="session",
+    params=[
+        # Japanese
+        ("CP932", "ﾎ"),
+        # Chinese
+        ("CP936", "中文"),
+        # Central European
+        ("CP1250", "Đ"),
+        # Latin 1 / Western European
+        ("CP1252", "ÿ"),
+        # Greek
+        ("CP1253", "Φ"),
+        # Arabic
+        ("CP1256", "ش"),
+    ],
+)
+def encoded_text(request):
+    """Return tuple with encoding name and very short sample text in that encoding
+    NOTE: it was determined through testing that code pages for MS-DOS do not
+    consistently work across all Python installations (in particular, fail with conda),
+    but ANSI code pages appear to work properly.
+    """
+    return request.param

pyogrio/tests/fixtures/README.md

@@ -0,0 +1,89 @@
+# Test datasets
+
+## Natural Earth lowres
+
+`naturalearth_lowres.shp` was copied from GeoPandas.
+
+## FGDB test dataset
+
+`test_fgdb.gdb.zip`
+Downloaded from http://trac.osgeo.org/gdal/raw-attachment/wiki/FileGDB/test_fgdb.gdb.zip
+
+### GPKG test dataset with null values
+
+`test_gpkg_nulls.gpkg` was created using Fiona backend to GeoPandas:
+
+```
+from collections import OrderedDict
+
+import fiona
+import geopandas as gp
+import numpy as np
+from pyogrio import write_dataframe
+
+filename = "test_gpkg_nulls.gpkg"
+
+df = gp.GeoDataFrame(
+    {
+        "col_bool": np.array([True, False, True], dtype="bool"),
+        "col_int8": np.array([1, 2, 3], dtype="int8"),
+        "col_int16": np.array([1, 2, 3], dtype="int16"),
+        "col_int32": np.array([1, 2, 3], dtype="int32"),
+        "col_int64": np.array([1, 2, 3], dtype="int64"),
+        "col_uint8": np.array([1, 2, 3], dtype="uint8"),
+        "col_uint16": np.array([1, 2, 3], dtype="uint16"),
+        "col_uint32": np.array([1, 2, 3], dtype="uint32"),
+        "col_uint64": np.array([1, 2, 3], dtype="uint64"),
+        "col_float32": np.array([1.5, 2.5, 3.5], dtype="float32"),
+        "col_float64": np.array([1.5, 2.5, 3.5], dtype="float64"),
+    },
+    geometry=gp.points_from_xy([0, 1, 2], [0, 1, 2]),
+    crs="EPSG:4326",
+)
+
+write_dataframe(df, filename)
+
+# construct row with null values
+# Note: np.nan can only be used for float values
+null_row = {
+    "type": "Fetaure",
+    "id": 4,
+    "properties": OrderedDict(
+        [
+            ("col_bool", None),
+            ("col_int8", None),
+            ("col_int16", None),
+            ("col_int32", None),
+            ("col_int64", None),
+            ("col_uint8", None),
+            ("col_uint16", None),
+            ("col_uint32", None),
+            ("col_uint64", None),
+            ("col_float32", np.nan),
+            ("col_float64", np.nan),
+        ]
+    ),
+    "geometry": {"type": "Point", "coordinates": (4.0, 4.0)},
+}
+
+# append row with nulls to GPKG
+with fiona.open(filename, "a") as c:
+    c.write(null_row)
+```
+
+NOTE: Reading boolean values into GeoPandas using Fiona backend treats those
+values as `None` and column dtype as `object`; Pyogrio treats those values as
+`np.nan` and column dtype as `float64`.
+
+### GPKG test with MultiSurface
+
+This was extracted from https://prd-tnm.s3.amazonaws.com/StagedProducts/Hydrography/NHDPlusHR/Beta/GDB/NHDPLUS_H_0308_HU4_GDB.zip
+`NHDWaterbody` layer using ogr2ogr:
+
+```bash
+ogr2ogr test_mixed_surface.gpkg NHDPLUS_H_0308_HU4_GDB.gdb NHDWaterbody -where '"NHDPlusID" = 15000300070477' -select "NHDPlusID"
+```
+
+### OSM PBF test
+
+This was downloaded from https://github.com/openstreetmap/OSM-binary/blob/master/resources/sample.pbf

pyogrio/tests/fixtures/naturalearth_lowres/naturalearth_lowres.cpg

@@ -0,0 +1 @@
+ISO-8859-1

pyogrio/tests/fixtures/naturalearth_lowres/naturalearth_lowres.prj

@@ -0,0 +1 @@
+GEOGCS["GCS_WGS_1984",DATUM["D_WGS_1984",SPHEROID["WGS_1984",6378137,298.257223563]],PRIMEM["Greenwich",0],UNIT["Degree",0.017453292519943295]]

pyogrio/tests/fixtures/test_datetime.geojson

@@ -0,0 +1,7 @@
+{
+"type": "FeatureCollection",
+"features": [
+{ "type": "Feature", "properties": { "col": "2020-01-01T09:00:00.123" }, "geometry": { "type": "Point", "coordinates": [ 1.0, 1.0 ] } },
+{ "type": "Feature", "properties": { "col": "2020-01-01T10:00:00" }, "geometry": { "type": "Point", "coordinates": [ 2.0, 2.0 ] } }
+]
+}

pyogrio/tests/fixtures/test_datetime_tz.geojson

@@ -0,0 +1,8 @@
+{
+"type": "FeatureCollection",
+"crs": { "type": "name", "properties": { "name": "urn:ogc:def:crs:OGC:1.3:CRS84" } },
+"features": [
+{ "type": "Feature", "properties": { "datetime_col": "2020-01-01T09:00:00.123-05:00" }, "geometry": { "type": "Point", "coordinates": [ 1.0, 1.0 ] } },
+{ "type": "Feature", "properties": { "datetime_col": "2020-01-01T10:00:00-05:00" }, "geometry": { "type": "Point", "coordinates": [ 2.0, 2.0 ] } }
+]
+}

pyogrio/tests/fixtures/test_nested.geojson

@@ -0,0 +1,18 @@
+{
+    "type": "FeatureCollection",
+    "features": [
+        {
+            "type": "Feature",
+            "geometry": {
+                "type": "Point",
+                "coordinates": [0, 0]
+            },
+            "properties": {
+                "top_level": "A",
+                "intermediate_level": {
+                    "bottom_level": "B"
+                }
+            }
+        }
+    ]
+}

pyogrio/tests/fixtures/test_ogr_types_list.geojson

@@ -0,0 +1,12 @@
+{
+"type": "FeatureCollection",
+"name": "test",
+"crs": { "type": "name", "properties": { "name": "urn:ogc:def:crs:OGC:1.3:CRS84" } },
+"features": [
+{ "type": "Feature", "properties": { "int64": 1, "list_int64": [ 0, 1 ] }, "geometry": { "type": "Point", "coordinates": [ 0.0, 2.0 ] } },
+{ "type": "Feature", "properties": { "int64": 2, "list_int64": [ 2, 3 ] }, "geometry": { "type": "Point", "coordinates": [ 1.0, 2.0 ] } },
+{ "type": "Feature", "properties": { "int64": 3, "list_int64": [ 4, 5 ] }, "geometry": { "type": "Point", "coordinates": [ 2.0, 2.0 ] } },
+{ "type": "Feature", "properties": { "int64": 4, "list_int64": [ 6, 7 ] }, "geometry": { "type": "Point", "coordinates": [ 3.0, 2.0 ] } },
+{ "type": "Feature", "properties": { "int64": 5, "list_int64": [ 8, 9 ] }, "geometry": { "type": "Point", "coordinates": [ 4.0, 2.0 ] } }
+]
+}