RavenGIS

.github/workflows/main.yml

@@ -24,32 +24,47 @@ jobs:
         run: tox -e ${{ matrix.tox-env }}
 
   pip:
-    name: Pip with Python${{ matrix.python-version }}
+    name: Pip with Python${{ matrix.python-version }} (GDAL=${{ matrix.uses-gdal }})
     needs: black
     runs-on: ubuntu-latest
     strategy:
       matrix:
         include:
           - tox-env: py37
             python-version: 3.7
+            uses-gdal: true
           - tox-env: py38
             python-version: 3.8
+            uses-gdal: true
+          - tox-env: py38
+            python-version: 3.8
+            uses-gdal: false
           - tox-env: py39
             python-version: 3.9
+            uses-gdal: true
     steps:
     - uses: actions/checkout@v2
     - name: Set up Python ${{ matrix.python-version }}
       uses: actions/setup-python@v2
       with:
         python-version: ${{ matrix.python-version }}
+    - name: Install NetCDF and GEOS
+      run: |
+        sudo apt-get update
+        sudo apt-get install libnetcdf-dev libgeos-dev
     - name: Install GDAL
+      if: ${{ matrix.uses-gdal }}
       run: |
         sudo apt-get update
         sudo apt-get install libgdal-dev
     - name: Install tox
       run: pip install tox
-    - name: Test with tox
-      run: env GDAL_VERSION="$(gdal-config --version)" tox -e ${{ matrix.tox-env }}
+    - name: Test with tox (GDAL)
+      if: ${{ matrix.uses-gdal }}
+      run: env GDAL_VERSION="$(gdal-config --version)" tox -e ${{ matrix.tox-env }}-gdal-slow
+    - name: Test with tox (no GDAL)
+      if: ${{ !matrix.uses-gdal }}
+      run: tox -e ${{ matrix.tox-env }}-slow
 
   conda:
     name: Conda

MANIFEST.in

@@ -5,7 +5,6 @@ include LICENSE
 include README.rst
 include requirements_dev.txt
 include requirements_docs.txt
-include requirements_gis.txt
 
 recursive-include tests *
 recursive-exclude * __pycache__

environment.yml

@@ -6,14 +6,10 @@ dependencies:
   - raven-hydro ==3.0.4.322
   - ostrich ==21.03.16
   - python >=3.7
-  - affine
   - cf_xarray
   - click
   - climpred >=2.1
   - dask
-  - fiona
-  - gdal >=3.0.0
-  - geopandas >=0.9
   - haversine
   - holoviews
   - hvplot
@@ -25,16 +21,15 @@ dependencies:
   - pip
   - pre-commit
   - pydantic
+  - pygml
   - pyproj>=3
-  - rasterio
   - requests
-  - rioxarray
   - scikit-learn ==0.24.2
   - scipy
   - setuptools >50.0
   - shapely
   - statsmodels
   - xarray >=0.18
-  - xclim >=0.26.1,<0.29.0
+  - xclim >=0.28.1,<0.29.0
   - xskillscore
   - wheel

ravenpy/cli/collect_subbasins_upstream_of_gauge.py

@@ -2,7 +2,10 @@
 from pathlib import Path
 
 import click
-import geopandas as gpd
+
+from ravenpy.utilities.vector import archive_sniffer, vector_to_dataframe
+
+# import geopandas as gpd
 
 
 @click.command()
@@ -22,11 +25,16 @@ def collect_subbasins_upstream_of_gauge(
 
     GAUGE_ID: ID of the target gauge, to be found in the "Obs_NM" column (e.g. "02LE024").
     """
-    if Path(input_file).suffix == ".zip":
-        input_file = f"zip://{input_file}"
-    gdf = gpd.read_file(input_file)
+    input_file = Path(input_file)
 
-    if gauge_id not in gdf.Obs_NM.values:
+    if input_file.suffix == ".zip":
+        input_file = next(iter(archive_sniffer(input_file)))
+    if input_file.suffix.lower() == ".shp":
+        df = vector_to_dataframe(input_file.as_posix())
+    else:
+        raise FileNotFoundError()
+
+    if gauge_id not in df.Obs_NM.values:
         raise click.ClickException(f"Cannot find gauge {gauge_id} in `Obs_NM` column")
 
     if not output:
@@ -37,11 +45,11 @@ def collect_subbasins_upstream_of_gauge(
 
     downsubid_to_subids = defaultdict(set)
 
-    for _, r in gdf.iterrows():
+    for _, r in df.iterrows():
         downsubid_to_subids[r.DowSubId].add(r.SubId)
 
     # Starting from the SubId of the gauge, iteratively expand the set of upstream subbasin SubIds
-    upstream_subids = {gdf[gdf.Obs_NM == gauge_id].iloc[0].SubId}
+    upstream_subids = {df[df.Obs_NM == gauge_id].iloc[0].SubId}
     prev = upstream_subids.copy()
     while True:
         curr = set()
@@ -53,10 +61,12 @@ def collect_subbasins_upstream_of_gauge(
         else:
             break
 
-    gdf_upstream = gdf[gdf.SubId.isin(upstream_subids)]
+    df_upstream = df[df.SubId.isin(upstream_subids)]
+
+    df_upstream.to_json(output_file)
 
-    gpd.GeoDataFrame(gdf_upstream).to_file(output_file)
+    # pd.GeoDataFrame(gdf_upstream).to_file(output_file)
 
     click.echo(
-        f"Found {len(gdf_upstream)} upstream subbasins, saved them in {output_file}"
+        f"Found {len(df_upstream)} upstream subbasins, saved them in {output_file}"
     )

ravenpy/cli/generate_grid_weights.py

@@ -4,6 +4,8 @@
 
 from ravenpy.extractors.routing_product import RoutingProductGridWeightExtractor
 
+WGS84 = 4326
+
 
 @click.command()
 @click.argument("input-file", type=click.Path(exists=True))
@@ -74,6 +76,22 @@
     type=click.Path(),
     help="Text field that will contain the results as a single :GridWeights Raven command containing the weights.",
 )
+@click.option(
+    "-i",
+    "--input-shape-crs",
+    type=int,
+    default=WGS84,
+    show_default=True,
+    help="Coordinate reference system (EPSG) of the shapefile INPUT_FILE.",
+)
+@click.option(
+    "-r",
+    "--routing-shape-crs",
+    type=int,
+    default=WGS84,
+    show_default=True,
+    help="Coordinate reference system (EPSG) of the shapefile ROUTING_FILE.",
+)
 def generate_grid_weights(
     input_file,
     routing_file,
@@ -85,6 +103,8 @@ def generate_grid_weights(
     sub_ids,
     area_error_threshold,
     output,
+    input_shape_crs,
+    routing_shape_crs,
 ):
     """
     Generate grid weights in various formats.
@@ -119,6 +139,8 @@ def generate_grid_weights(
         gauge_ids,
         sub_ids,
         area_error_threshold,
+        input_shape_crs,
+        routing_shape_crs,
     )
     gw_cmd = extractor.extract()