vector2dggs 0.5.3__py3-none-any.whl → 0.6.0__py3-none-any.whl

vector2dggs/__init__.py

@@ -1 +1 @@
-__version__: str = "0.5.3"
+__version__: str = "0.6.0"

vector2dggs/h3.py

@@ -41,6 +41,7 @@ warnings.filterwarnings(
 DEFAULT_PARENT_OFFSET = 6
 DEFAULT_CHUNK_SIZE = 50
 
+
 class ParentResolutionException(Exception):
     pass
 
@@ -68,29 +69,45 @@ def polyfill(
     output_directory: str,
 ) -> None:
     """
-    Reads a geoparquet, performs H3 polyfilling,
-    and writes out to parquet.
+    Reads a geoparquet, performs H3 polyfilling (for polygons),
+    linetracing (for linestrings), and writes out to parquet.
     """
-    df = (
-        gpd.read_parquet(pq_in)
-        .reset_index()
-        .drop(columns=[spatial_sort_col])
-    )
+    df = gpd.read_parquet(pq_in).reset_index().drop(columns=[spatial_sort_col])
     if len(df.index) == 0:
         # Input is empty, nothing to polyfill
         return None
-    df = df.h3.polyfill_resample(resolution, return_geometry=False)
+
+    df_polygon = df[df.geom_type == "Polygon"]
+    if len(df_polygon.index) > 0:
+        df_polygon = df_polygon.h3.polyfill_resample(
+            resolution, return_geometry=False
+        ).drop(columns=["index"])
+
+    df_linestring = df[df.geom_type == "LineString"]
+    if len(df_linestring.index) > 0:
+        df_linestring = (
+            df_linestring.h3.linetrace(resolution)
+            .explode("h3_linetrace")
+            .set_index("h3_linetrace")
+        )
+        df_linestring = df_linestring[~df_linestring.index.duplicated(keep="first")]
+
+    df = pd.concat(
+        map(
+            lambda _df: pd.DataFrame(_df.drop(columns=[_df.geometry.name])),
+            [df_polygon, df_linestring],
+        )
+    )
+
     if len(df.index) == 0:
         # Polyfill resulted in empty output (e.g. large cell, small feature)
         return None
-    df = pd.DataFrame(df).drop(columns=["index", "geometry"])
+
     df.index.rename(f"h3_{resolution:02}", inplace=True)
     parent_res: int = _get_parent_res(parent_res, resolution)
     # Secondary (parent) H3 index, used later for partitioning
     df.h3.h3_to_parent(parent_res).to_parquet(
-        PurePath(output_directory, pq_in.name),
-        engine="auto",
-        compression="ZSTD"
+        PurePath(output_directory, pq_in.name), engine="auto", compression="ZSTD"
     )
     return None
 
@@ -102,7 +119,7 @@ def polyfill_star(args) -> None:
 def _parent_partitioning(
     input_dir: Path,
     output_dir: Path,
-    resolution,
+    resolution: int,
     parent_res: Union[None, int],
     **kwargs,
 ) -> None:
@@ -129,6 +146,25 @@ def _parent_partitioning(
     return
 
 
+def drop_condition(
+    df: pd.DataFrame,
+    drop_index: pd.Index,
+    log_statement: str,
+    warning_threshold: float = 0.01,
+):
+    LOGGER.info(log_statement)
+    _before = len(df)
+    df = df.drop(drop_index)
+    _after = len(df)
+    _diff = _before - _after
+    if _diff:
+        log_method = (
+            LOGGER.info if (_diff / float(_before)) < warning_threshold else LOGGER.warn
+        )
+        log_method(f"Dropped {_diff} rows ({_diff/float(_before)*100:.2f}%)")
+    return df
+
+
 def _index(
     input_file: Union[Path, str],
     output_directory: Union[Path, str],
@@ -193,11 +229,24 @@ def _index(
         .explode(index_parts=False)  # Explode from GeometryCollection
         .explode(index_parts=False)  # Explode multipolygons to polygons
     ).reset_index()
-    LOGGER.info("Dropping empty or null geometries")
-    df = (
-        df.drop(df[(df.geometry.is_empty|df.geometry.isna())].index)
-        .reset_index()
-    )
+
+    drop_conditions = [
+        {
+            "index": lambda frame: frame[
+                (frame.geometry.is_empty | frame.geometry.isna())
+            ],
+            "message": "Dropping empty or null geometries",
+        },
+        {
+            "index": lambda frame: frame[
+                (frame.geometry.geom_type != "Polygon")
+                & (frame.geometry.geom_type != "LineString")
+            ],  # NB currently points and other types are lost; in principle, these could be indexed
+            "message": "Dropping non-polygonal geometries",
+        },
+    ]
+    for condition in drop_conditions:
+        df = drop_condition(df, condition["index"](df).index, condition["message"])
 
     ddf = dgpd.from_geopandas(df, chunksize=max(1, chunksize), sort=True)
 
@@ -336,7 +385,7 @@ def _index(
     "--tempdir",
     default=tempfile.tempdir,
     type=click.Path(),
-    help="Temporary data is created during the execution of this program. This parameter allows you to control where this data will be written."
+    help="Temporary data is created during the execution of this program. This parameter allows you to control where this data will be written.",
 )
 @click.option("-o", "--overwrite", is_flag=True)
 @click.version_option(version=__version__)

vector2dggs/katana.py

@@ -26,7 +26,7 @@ def katana(geometry, threshold, count=0) -> GeometryCollection:
     if not geometry.is_valid:
         # print(explain_validity(geometry))
         geometry = make_valid(geometry)
-        if geometry.type == 'GeometryCollection':
+        if geometry.type == "GeometryCollection":
             geometry.normalize()
         geometry = geometry.buffer(0)
     bounds = geometry.bounds

{vector2dggs-0.5.3.dist-info → vector2dggs-0.6.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vector2dggs
-Version: 0.5.3
+Version: 0.6.0
 Summary: CLI DGGS indexer for vector geospatial data
 Home-page: https://github.com/manaakiwhenua/vector2dggs
 License: LGPL-3.0-or-later
@@ -21,12 +21,12 @@ Requires-Dist: click (>=8.1.3,<9.0.0)
 Requires-Dist: click-log (>=0.4.0,<0.5.0)
 Requires-Dist: dask (>=2023.3.0,<2024.0.0)
 Requires-Dist: dask-geopandas (>=0.3.0,<0.4.0)
-Requires-Dist: gdal (>=3.6.4,<4.0.0)
+Requires-Dist: gdal (>=3.8.0,<4.0.0)
 Requires-Dist: geopandas (>=0.12.2,<0.13.0)
-Requires-Dist: h3pandas (>=0.2.3,<0.3.0)
+Requires-Dist: h3pandas (>=0.2.6,<0.3.0)
 Requires-Dist: psycopg2 (>=2.9.6,<3.0.0)
-Requires-Dist: pyarrow (>=11.0.0,<12.0.0)
-Requires-Dist: pygeos (>=0.14,<0.15)
+Requires-Dist: pyarrow (>=14.0.1,<15.0.0)
+Requires-Dist: pygeos (>=0.13,<0.14)
 Requires-Dist: pyproj (>=3.5.0,<4.0.0)
 Requires-Dist: sqlalchemy (>=2.0.10,<3.0.0)
 Requires-Dist: tqdm (>=4.65.0,<5.0.0)
@@ -187,14 +187,14 @@ vector2dggs h3 -v DEBUG -id ogc_fid -r 9 -p 5 -t 4 --overwrite -tbl topo50_lake
   title={{vector2dggs}},
   author={Ardo, James and Law, Richard},
   url={https://github.com/manaakiwhenua/vector2dggs},
-  version={0.5.3},
+  version={0.6.0},
   date={2023-04-20}
 }
 ```
 
 APA/Harvard
 
-> Ardo, J., & Law, R. (2023). vector2dggs (0.5.3) [Computer software]. https://github.com/manaakiwhenua/vector2dggs
+> Ardo, J., & Law, R. (2023). vector2dggs (0.6.0) [Computer software]. https://github.com/manaakiwhenua/vector2dggs
 
 [![manaakiwhenua-standards](https://github.com/manaakiwhenua/vector2dggs/workflows/manaakiwhenua-standards/badge.svg)](https://github.com/manaakiwhenua/manaakiwhenua-standards)
 

vector2dggs-0.6.0.dist-info/RECORD

@@ -0,0 +1,10 @@
+vector2dggs/__init__.py,sha256=8xTECrwGH36hEfgoQ2Zcq4dfigWVZmIFK3OHqNOg-FQ,27
+vector2dggs/cli.py,sha256=tL4NJ99uQsqoVinwYadna1a4ko5v2sdZaFaeDAj6QNE,599
+vector2dggs/h3.py,sha256=kX3S630l-LOm04pe5YT-5g99DIV0t32GFYUEs0Hc5ZQ,14354
+vector2dggs/katana.py,sha256=xx5R9lDuraWLK5bfGkiDQDC2r2naj_sKZlYeB52_xwc,3320
+vector2dggs-0.6.0.dist-info/entry_points.txt,sha256=5h8LB9L2oOE5u_N7FRGtu4JDwa553iPs4u0XhcLeLZU,52
+vector2dggs-0.6.0.dist-info/WHEEL,sha256=vVCvjcmxuUltf8cYhJ0sJMRDLr1XsPuxEId8YDzbyCY,88
+vector2dggs-0.6.0.dist-info/METADATA,sha256=d8dvx5_wXFO-ZMKcdXmc2TuXQv5B5UFRo1hq4fghe8w,9777
+vector2dggs-0.6.0.dist-info/COPYING.LESSER,sha256=46mU2C5kSwOnkqkw9XQAJlhBL2JAf1_uCD8lVcXyMRg,7652
+vector2dggs-0.6.0.dist-info/COPYING,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
+vector2dggs-0.6.0.dist-info/RECORD,,

{vector2dggs-0.5.3.dist-info → vector2dggs-0.6.0.dist-info}/WHEEL

@@ -1,4 +1,4 @@
 Wheel-Version: 1.0
-Generator: poetry-core 1.6.1
+Generator: poetry-core 1.4.0
 Root-Is-Purelib: true
 Tag: py3-none-any

vector2dggs-0.5.3.dist-info/RECORD

@@ -1,10 +0,0 @@
-vector2dggs/__init__.py,sha256=ricvdCMTDRNJiioloPx-XsgIV3l9wyNaFT5PqZCvlLI,27
-vector2dggs/cli.py,sha256=tL4NJ99uQsqoVinwYadna1a4ko5v2sdZaFaeDAj6QNE,599
-vector2dggs/h3.py,sha256=iEZGDaLYfblXXvmjWWhveQiaIxZ1xYvXx0phYBFgkLE,12726
-vector2dggs/katana.py,sha256=rxmNvXTdfvFs7Q17BNGPin_5H-jSLWGt63RKrEgYGOo,3320
-vector2dggs-0.5.3.dist-info/COPYING,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
-vector2dggs-0.5.3.dist-info/COPYING.LESSER,sha256=46mU2C5kSwOnkqkw9XQAJlhBL2JAf1_uCD8lVcXyMRg,7652
-vector2dggs-0.5.3.dist-info/METADATA,sha256=wyof6tDZ7wY9-dJQwkaY0V6f5Btfu3FGT8we50mtgIU,9777
-vector2dggs-0.5.3.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
-vector2dggs-0.5.3.dist-info/entry_points.txt,sha256=5h8LB9L2oOE5u_N7FRGtu4JDwa553iPs4u0XhcLeLZU,52
-vector2dggs-0.5.3.dist-info/RECORD,,