Merge pull request #5 from TomAugspurger/tom/fix/to_item_collection-types

Improved to_item_collection

Author: Tom Augspurger

.pre-commit-config.yaml

@@ -19,7 +19,7 @@ repos:
       - id: flake8
         language_version: python3
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v0.942
+    rev: v1.2.0
     hooks:
       - id: mypy
         # Override default --ignore-missing-imports

pyproject.toml

@@ -15,7 +15,8 @@ dependencies = [
     "geopandas",
     "pandas",
     "pyarrow",
-    "shapely"
+    "shapely",
+    "packaging",
 ]
 
 [project.optional-dependencies]
@@ -54,6 +55,8 @@ filterwarnings = [
 
 [tool.mypy]
 
+python_version = "3.10"
+
 [[tool.mypy.overrides]]
 module = [
     "shapely.*",

stac_geoparquet/__init__.py

@@ -1,7 +1,7 @@
 """stac-geoparquet"""
 __version__ = "0.2.0"
 
-from .stac_geoparquet import to_geodataframe
+from .stac_geoparquet import to_geodataframe, to_dict, to_item_collection
 
 
-__all__ = ["__version__", "to_geodataframe"]
+__all__ = ["__version__", "to_geodataframe", "to_dict", "to_item_collection"]

stac_geoparquet/_compat.py

@@ -0,0 +1,6 @@
+import pystac
+import packaging.version
+
+PYSTAC_1_7_0 = packaging.version.parse(pystac.__version__) >= packaging.version.Version(
+    "1.7.0"
+)

stac_geoparquet/stac_geoparquet.py

@@ -8,11 +8,22 @@
 import pystac
 import geopandas
 import pandas as pd
+import numpy as np
 import shapely.geometry
 
 from stac_geoparquet.utils import fix_empty_multipolygon
 
 
+def _fix_array(v):
+    if isinstance(v, np.ndarray):
+        v = v.tolist()
+
+    elif isinstance(v, dict):
+        v = {k: _fix_array(v2) for k, v2 in v.items()}
+
+    return v
+
+
 def to_geodataframe(items: Sequence[dict[str, Any]]) -> geopandas.GeoDataFrame:
     """
     Convert a sequence of STAC items to a :class:`geopandas.GeoDataFrame`.
@@ -94,6 +105,7 @@ def to_dict(record: dict) -> dict:
     }
     item = {}
     for k, v in record.items():
+        v = _fix_array(v)
 
         if k in top_level_keys:
             item[k] = v
@@ -107,6 +119,20 @@ def to_dict(record: dict) -> dict:
 
 
 def to_item_collection(df: geopandas.GeoDataFrame) -> pystac.ItemCollection:
+    """
+    Convert a GeoDataFrame of STAC items to an :class:`pystac.ItemCollection`.
+
+    Parameters
+    ----------
+    df : geopandas.GeoDataFrame
+        A GeoDataFrame with a schema similar to that exported by stac-geoparquet.
+
+    Returns
+    -------
+    item_collection : pystac.ItemCollection
+        The converted ItemCollection. There will be one record / feature per
+        row in the in the GeoDataFrame.
+    """
     df2 = df.copy()
     datelike = df2.select_dtypes(
         include=["datetime64[ns, UTC]", "datetime64[ns]"]
@@ -116,6 +142,5 @@ def to_item_collection(df: geopandas.GeoDataFrame) -> pystac.ItemCollection:
             df2[k].dt.strftime("%Y-%m-%dT%H:%M:%S.%fZ").fillna("").replace({"": None})
         )
 
-    return pystac.ItemCollection(
-        [to_dict(record) for record in df2.to_dict(orient="records")]
-    )
+    records = [to_dict(record) for record in df2.to_dict(orient="records")]
+    return pystac.ItemCollection(records)

tests/data/naip.parquet

@@ -9,6 +9,7 @@
 
 import stac_geoparquet.pgstac_reader
 from stac_geoparquet.utils import assert_equal
+from stac_geoparquet._compat import PYSTAC_1_7_0
 
 
 HERE = pathlib.Path(__file__).parent
@@ -96,7 +97,8 @@ def test_naip_item():
     ]
 
     cfg = stac_geoparquet.pgstac_reader.CollectionConfig(
-        collection_id="naip", render_config="assets=image&asset_bidx=image%7C1%2C2%2C3"
+        collection_id="naip",
+        render_config="assets=image&asset_bidx=image%7C1%2C2%2C3&format=png",
     )
     result = cfg.make_pgstac_items(records, base_item)[0]
     # shapely uses tuples instead of lists
@@ -106,6 +108,11 @@ def test_naip_item():
         "https://planetarycomputer.microsoft.com/api/stac/v1/collections/naip/items/pa_m_4108053_se_17_1_20150725_20151201"  # noqa: E501
     )
 
+    if PYSTAC_1_7_0:
+        # https://github.com/stac-utils/pystac/issues/1102
+        expected.remove_links(rel=pystac.RelType.SELF)
+        result.remove_links(rel=pystac.RelType.SELF)
+
     assert_equal(result, expected)
 
 
@@ -120,12 +127,17 @@ def test_sentinel2_l2a():
         partition_frequency=None,
         stac_api="https://planetarycomputer.microsoft.com/api/stac/v1",
         should_inject_dynamic_properties=True,
-        render_config="assets=visual&asset_bidx=visual%7C1%2C2%2C3&nodata=0",
+        render_config="assets=visual&asset_bidx=visual%7C1%2C2%2C3&nodata=0&format=png",
     )
     result = pystac.read_dict(config.make_pgstac_items([record], base_item)[0])
     expected = pystac.read_file(
         "https://planetarycomputer.microsoft.com/api/stac/v1/collections/sentinel-2-l2a/items/S2A_MSIL2A_20150704T101006_R022_T35XQA_20210411T133707"  # noqa: E501
     )
+    if PYSTAC_1_7_0:
+        # https://github.com/stac-utils/pystac/issues/1102
+        expected.remove_links(rel=pystac.RelType.SELF)
+        result.remove_links(rel=pystac.RelType.SELF)
+
     expected.remove_links(rel=pystac.RelType.LICENSE)
     assert_equal(result, expected)
 

tests/test_pgstac_reader.py

@@ -324,9 +324,11 @@ def test_s1_grd():
     ],
 )
 def test_smoke(collection_id):
-    items = requests.get(
+    r = requests.get(
         f"https://planetarycomputer.microsoft.com/api/stac/v1/collections/{collection_id}/items?limit=1"
-    ).json()["features"]
+    )
+    r.raise_for_status()
+    items = r.json()["features"]
     df = stac_geoparquet.to_geodataframe(items)
 
     result = to_item_collection(df)

tests/test_stac_geoparquet.py

@@ -0,0 +1,118 @@
+import pathlib
+
+import pytest
+import geopandas
+
+import stac_geoparquet
+
+
+HERE = pathlib.Path(__file__).parent
+
+
+@pytest.fixture
+def naip():
+    return geopandas.read_parquet(HERE / "data" / "naip.parquet")
+
+
+def test_to_dict(naip):
+    result = stac_geoparquet.to_item_collection(naip)
+    expected = {
+        "assets": {
+            "image": {
+                "eo:bands": [
+                    {"common_name": "red", "description": None, "name": "Red"},
+                    {"common_name": "green", "description": None, "name": "Green"},
+                    {"common_name": "blue", "description": None, "name": "Blue"},
+                    {
+                        "common_name": "nir",
+                        "description": "near-infrared",
+                        "name": "NIR",
+                    },
+                ],
+                "href": "https://naipeuwest.blob.core.windows.net/naip/v002/ok/2010/ok_100cm_2010/34099/m_3409901_nw_14_1_20100425.tif",  # noqa: E501
+                "roles": ["data"],
+                "title": "RGBIR COG tile",
+                "type": "image/tiff; application=geotiff; " "profile=cloud-optimized",
+            },
+            "rendered_preview": {
+                "href": "https://planetarycomputer.microsoft.com/api/data/v1/item/preview.png?collection=naip&item=ok_m_3409901_nw_14_1_20100425&assets=image&asset_bidx=image%7C1%2C2%2C3",  # noqa: E501
+                "rel": "preview",
+                "roles": ["overview"],
+                "title": "Rendered preview",
+                "type": "image/png",
+            },
+            "thumbnail": {
+                "href": "https://naipeuwest.blob.core.windows.net/naip/v002/ok/2010/ok_100cm_2010/34099/m_3409901_nw_14_1_20100425.200.jpg",  # noqa: E501
+                "roles": ["thumbnail"],
+                "title": "Thumbnail",
+                "type": "image/jpeg",
+            },
+            "tilejson": {
+                "href": "https://planetarycomputer.microsoft.com/api/data/v1/item/tilejson.json?collection=naip&item=ok_m_3409901_nw_14_1_20100425&assets=image&asset_bidx=image%7C1%2C2%2C3",  # noqa: E501
+                "roles": ["tiles"],
+                "title": "TileJSON with default rendering",
+                "type": "application/json",
+            },
+        },
+        "bbox": [-100.004084, 34.934259, -99.933454, 35.00323],
+        "collection": "naip",
+        "geometry": {
+            "coordinates": (
+                (
+                    (-99.933454, 34.934815),
+                    (-99.93423, 35.00323),
+                    (-100.004084, 35.002673),
+                    (-100.00325, 34.934259),
+                    (-99.933454, 34.934815),
+                ),
+            ),
+            "type": "Polygon",
+        },
+        "id": "ok_m_3409901_nw_14_1_20100425",
+        "links": [
+            {
+                "href": "https://planetarycomputer.microsoft.com/api/stac/v1/collections/naip",
+                "rel": "collection",
+                "type": "application/json",
+            },
+            {
+                "href": "https://planetarycomputer.microsoft.com/api/stac/v1/collections/naip",
+                "rel": "parent",
+                "type": "application/json",
+            },
+            {
+                "href": "https://planetarycomputer.microsoft.com/api/stac/v1/",
+                "rel": "root",
+                "title": "Microsoft Planetary Computer STAC API",
+                "type": "application/json",
+            },
+            {
+                "href": "https://planetarycomputer.microsoft.com/api/stac/v1/collections/naip/items/ok_m_3409901_nw_14_1_20100425",  # noqa: E501
+                "rel": "self",
+                "type": "application/geo+json",
+            },
+            {
+                "href": "https://planetarycomputer.microsoft.com/api/data/v1/item/map?collection=naip&item=ok_m_3409901_nw_14_1_20100425",  # noqa: E501
+                "rel": "preview",
+                "title": "Map of item",
+                "type": "text/html",
+            },
+        ],
+        "properties": {
+            "datetime": "2010-04-25T00:00:00Z",
+            "gsd": 1.0,
+            "naip:state": "ok",
+            "naip:year": "2010",
+            "proj:bbox": [408377.0, 3866212.0, 414752.0, 3873800.0],
+            "proj:epsg": 26914,
+            "proj:shape": [7588, 6375],
+            "proj:transform": [1.0, 0.0, 408377.0, 0.0, -1.0, 3873800.0, 0.0, 0.0, 1.0],
+        },
+        "stac_extensions": [
+            "https://stac-extensions.github.io/eo/v1.0.0/schema.json",
+            "https://stac-extensions.github.io/projection/v1.0.0/schema.json",
+        ],
+        "stac_version": "1.0.0",
+        "type": "Feature",
+    }
+    assert result[0].to_dict() == expected