Cleaner data repro for point and polygon examples (#45)

* Point example data repro

* repro for polygon data

Author: kylebarron

examples/point/README.md

@@ -1,23 +1,18 @@
-todo: try portal again (had issue with mismatched deck.gl versions)
-
-```json
-"@geoarrow/deck.gl-layers": "portal:../../modules/deck.gl-geoarrow",
-```
-
-
-
-
-
-
-
-
-
+## Example: Use `@geoarrow/deck.gl-layers` with GeoArrow point data
 
+## Data for example:
 
+```
+wget https://ookla-open-data.s3.us-west-2.amazonaws.com/parquet/performance/type=mobile/year=2019/quarter=1/2019-01-01_performance_mobile_tiles.parquet
+poetry install
+poetry run python generate_data.py
+```
 
-## Example: Use deck.gl with react-map-gl and Webpack
+## Serve data
 
-Uses [Vite](https://vitejs.dev/) to bundle and serve files.
+```
+npx http-server --cors
+```
 
 ## Usage
 
@@ -30,9 +25,6 @@ yarn
 ```
 
 Commands:
+
 * `npm start` is the development target, to serve the app and hot reload.
 * `npm run build` is the production target, to create the final bundle and write to disk.
-
-### Basemap
-
-The basemap in this example is provided by [CARTO free basemap service](https://carto.com/basemaps). To use an alternative base map solution, visit [this guide](https://deck.gl/docs/get-started/using-with-map#using-other-basemap-services)

examples/point/app.tsx

@@ -6,7 +6,7 @@ import { GeoArrowScatterplotLayer } from "@geoarrow/deck.gl-layers";
 import * as arrow from "apache-arrow";
 
 const GEOARROW_POINT_DATA =
-  "http://localhost:8080/2019-01-01_performance_mobile_tiles_color.feather";
+  "http://localhost:8080/2019-01-01_performance_mobile_tiles.feather";
 
 const INITIAL_VIEW_STATE = {
   latitude: 20,
@@ -57,8 +57,8 @@ function Root() {
       new GeoArrowScatterplotLayer({
         id: "geoarrow-points",
         data: table,
-        getFillColor: [255, 0, 0],
-        // getFillColor: "colors",
+        // getFillColor: [255, 0, 0],
+        getFillColor: table.getChild("colors"),
         radiusMinPixels: 1,
         getPointRadius: 10,
         pointRadiusMinPixels: 0.8,

examples/point/generate_data.py

@@ -1,45 +1,45 @@
+from pathlib import Path
+
+import geopandas as gpd
 import pandas as pd
 import pyarrow as pa
 import pyarrow.feather as feather
 import shapely
+from lonboard.colormap import apply_continuous_cmap
+from lonboard.geoarrow.geopandas_interop import geopandas_to_geoarrow
+from palettable.colorbrewer.diverging import BrBG_10
 
+url = "https://ookla-open-data.s3.us-west-2.amazonaws.com/parquet/performance/type=mobile/year=2019/quarter=1/2019-01-01_performance_mobile_tiles.parquet"
 
-class PointGeometryType(pa.ExtensionType):
-    def __init__(self):
-        pa.ExtensionType.__init__(self, self._storage_type, self._extension_name)
-
-    _storage_type = pa.list_(pa.field("xy", pa.float64()), 2)
-    _extension_name = "geoarrow.point"
-
-    def __arrow_ext_serialize__(self):
-        # since we don't have a parameterized type, we don't need extra
-        # metadata to be deserialized
-        return b""
-
-    @classmethod
-    def __arrow_ext_deserialize__(cls, storage_type, serialized):
-        # return an instance of this subclass given the serialized
-        # metadata.
-        return PointGeometryType()
-
-
-# https://ookla-open-data.s3.us-west-2.amazonaws.com/parquet/performance/type=mobile/year=2019/quarter=1/2019-01-01_performance_mobile_tiles.parquet
+path = Path("2019-01-01_performance_mobile_tiles.parquet")
 
 
 def main():
-    df = pd.read_parquet("2019-01-01_performance_mobile_tiles.parquet")
+    if not path.exists():
+        msg = f"Please download file to this directory from {url=}."
+        raise ValueError(msg)
+
+    df = pd.read_parquet(path)
     centroids = shapely.centroid(shapely.from_wkt(df["tile"]))
 
     # Save space by using a smaller data type
     df_cols = ["avg_d_kbps", "avg_u_kbps", "avg_lat_ms"]
     for col in df_cols:
         df[col] = pd.to_numeric(df[col], downcast="unsigned")
 
-    table = pa.Table.from_pandas(df[df_cols])
-    coords = shapely.get_coordinates(centroids)
-    parr = pa.FixedSizeListArray.from_arrays(coords.flatten(), 2)
-    extension_arr = pa.ExtensionArray.from_storage(PointGeometryType(), parr)
-    table = table.append_column("geometry", extension_arr)
+    gdf = gpd.GeoDataFrame(df[df_cols], geometry=centroids)
+    table = geopandas_to_geoarrow(gdf, preserve_index=False)
+
+    min_bound = 5000
+    max_bound = 50000
+    download_speed = gdf["avg_d_kbps"]
+    normalized_download_speed = (download_speed - min_bound) / (max_bound - min_bound)
+
+    colors = apply_continuous_cmap(normalized_download_speed, BrBG_10)
+    table = table.append_column(
+        "colors", pa.FixedSizeListArray.from_arrays(colors.flatten("C"), 3)
+    )
+
     feather.write_feather(
         table, "2019-01-01_performance_mobile_tiles.feather", compression="uncompressed"
     )