feat: change minimal duckdb version (#153)

* feat: bump duckdb version

* chore: change readme

* chore: change string to constant

* ci: removed conftest.py from build process

* chore: replace unary_union calls and silence warnings

* feat: refactor duckdb parquet operations

* chore: remove loading parquet extension

* fix: change geometry reading for empty relations

* chore: change lock hash

* chore: add explicit geometry casting

* chore: simplify tags loading with new duckdb engine

CHANGELOG.md

@@ -7,6 +7,14 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+### Changed
+
+- Bumped minimal DuckDB version to `1.1.0`
+- Refactored geoparquet operations for compatibility with new DuckDB version
+- Excluded `conftest.py` file from the final library build
+- Replaced `unary_union` calls with `union_all()` on all GeoDataFrames
+- Silenced `pooch` library warnings regarding empty SHA hash
+
 ## [0.10.0] - 2024-09-23
 
 ### Changed

README.md

@@ -70,7 +70,7 @@ QuackOSM supports **Python >= 3.9**
 
 Required:
 
-- `duckdb (>=0.10.2, <1.1.0)`: For all DuckDB operations on PBF files
+- `duckdb (>=1.1.0)`: For all DuckDB operations on PBF files
 
 - `pyarrow (>=16.0.0)`: For parquet files wrangling
 
@@ -148,9 +148,7 @@ way/993121275      {'building': 'yes', 'name': ...  POLYGON ((7.43214 43.7481...
 ```python
 >>> import duckdb
 >>> duckdb.load_extension('spatial')
->>> duckdb.read_parquet(str(gpq_path)).project(
-...     "* REPLACE (ST_GeomFromWKB(geometry) AS geometry)"
-... ).order("feature_id")
+>>> duckdb.read_parquet(str(gpq_path)).order("feature_id")
 ┌──────────────────┬──────────────────────┬──────────────────────────────────────────────┐
 │    feature_id    │         tags         │                   geometry                   │
 │     varchar      │ map(varchar, varch…  │                   geometry                   │
@@ -270,9 +268,7 @@ way/998561139     {'barrier': 'bollard', 'bicyc...  LINESTRING (12.45828 41.9...
 ```python
 >>> import duckdb
 >>> duckdb.load_extension('spatial')
->>> duckdb.read_parquet(str(gpq_path)).project(
-...     "* REPLACE (ST_GeomFromWKB(geometry) AS geometry)"
-... ).order("feature_id")
+>>> duckdb.read_parquet(str(gpq_path)).order("feature_id")
 ┌──────────────────┬──────────────────────┬──────────────────────────────────────────────┐
 │    feature_id    │         tags         │                   geometry                   │
 │     varchar      │ map(varchar, varch…  │                   geometry                   │
@@ -388,9 +384,7 @@ relation/3256168   {'building': 'yes', 'type': ...  POLYGON ((12.46061 41.907...
 ```python
 >>> import duckdb
 >>> duckdb.load_extension('spatial')
->>> duckdb.read_parquet(str(gpq_path)).project(
-...     "* REPLACE (ST_GeomFromWKB(geometry) AS geometry)"
-... ).order("feature_id")
+>>> duckdb.read_parquet(str(gpq_path)).order("feature_id")
 ┌──────────────────┬────────────────────────────┬──────────────────────────────┐
 │    feature_id    │            tags            │           geometry           │
 │     varchar      │   map(varchar, varchar)    │           geometry           │

pyproject.toml

@@ -7,7 +7,7 @@ dependencies = [
     "geopandas>=0.6",
     "shapely>=2",
     "pyarrow>=16.0.0",
-    "duckdb>=0.10.2,<1.1.0",
+    "duckdb>=1.1.0",
     "geoarrow-pyarrow>=0.1.2",
     "geoarrow-pandas>=0.1.1",
     "typeguard>=3.0.0",
@@ -109,6 +109,9 @@ cli-dev = ["ipywidgets", "folium", "matplotlib>=3.2.0", "mapclassify"]
 [tool.pdm.scripts]
 post_install = "pre-commit install"
 
+[tool.pdm.build]
+excludes = ["quackosm/conftest.py"]
+
 [tool.black]
 line-length = 100
 target-version = ["py39", "py310", "py311", "py312"]

quackosm/cli.py

@@ -96,7 +96,7 @@ def convert(self, value, param=None, ctx=None):  # type: ignore
             import geopandas as gpd
 
             gdf = gpd.read_file(value)
-            return gdf.unary_union
+            return gdf.union_all()
         except Exception:
             raise typer.BadParameter("Cannot parse provided geo file") from None
 
@@ -140,7 +140,7 @@ def convert(self, value, param=None, ctx=None):  # type: ignore
                 geometries.append(
                     box(minx=bounds["w"], miny=bounds["s"], maxx=bounds["e"], maxy=bounds["n"])
                 )
-            return gpd.GeoSeries(geometries).unary_union
+            return gpd.GeoSeries(geometries).union_all()
         except Exception:
             raise typer.BadParameter(f"Cannot parse provided Geohash value: {geohash}") from None
 
@@ -165,7 +165,7 @@ def convert(self, value, param=None, ctx=None):  # type: ignore
                 geometries.append(
                     Polygon([coords[::-1] for coords in h3.cell_to_boundary(h3_cell.strip())])
                 )
-            return gpd.GeoSeries(geometries).unary_union
+            return gpd.GeoSeries(geometries).union_all()
         except Exception as ex:
             raise typer.BadParameter(f"Cannot parse provided H3 values: {value}") from ex
 
@@ -190,7 +190,7 @@ def convert(self, value, param=None, ctx=None):  # type: ignore
                 geometries.append(
                     Polygon(s2.s2_to_geo_boundary(s2_index.strip(), geo_json_conformant=True))
                 )
-            return gpd.GeoSeries(geometries).unary_union
+            return gpd.GeoSeries(geometries).union_all()
         except Exception:
             raise typer.BadParameter(f"Cannot parse provided S2 value: {s2_index}") from None
 

quackosm/conftest.py

@@ -9,6 +9,7 @@
 import duckdb
 import pandas
 import pytest
+from pooch import get_logger as get_pooch_logger
 from pooch import retrieve
 
 from quackosm.osm_extracts.extract import OsmExtractSource
@@ -53,13 +54,15 @@ def add_pbf_files(doctest_namespace):  # type: ignore
         shutil.copy(pbf_file_path, geofabrik_pbf_file_path)
 
 
-
 @pytest.fixture(autouse=True, scope="session")
 def download_osm_extracts_indexes():  # type: ignore
     """Download OSM extract indexes files to cache."""
     download_directory = Path("cache")
     download_directory.mkdir(parents=True, exist_ok=True)
 
+    logger = get_pooch_logger()
+    logger.setLevel("WARNING")
+
     for osm_extract in OsmExtractSource:
         if osm_extract == OsmExtractSource.any:
             continue
@@ -81,6 +84,7 @@ def install_spatial_extension():  # type: ignore
     """Install duckdb spatial extension."""
     duckdb.install_extension("spatial")
 
+
 @pytest.fixture(autouse=True, scope="session")  # type: ignore
 def pandas_terminal_width() -> None:
     """Change pandas dataframe display options."""