Fix tests (#620)

This PR fixes

1. [failing tests](https://github.com/rapidsai/cuxfilter/actions/runs/10303789190/job/28671747182?pr=617), which were caused by duplicate columns in the core_graph source_df. 

2. Failing [external tests](https://github.com/rapidsai/cuxfilter/actions/workflows/test-external.yaml)

Authors:
  - Ajay Thorve (https://github.com/AjayThorve)

Approvers:
  - Matthew Roeschke (https://github.com/mroeschke)
  - Bradley Dice (https://github.com/bdice)
  - Allan (https://github.com/exactlyallan)

URL: #620

ci/test_external.sh

@@ -7,9 +7,19 @@ rapids-logger "Create test_external conda environment"
 . /opt/conda/etc/profile.d/conda.sh
 
 # Install external dependencies into test_external conda environment
-rapids-conda-retry env update -f ./ci/utils/external_dependencies.yaml
+rapids-dependency-file-generator \
+  --output conda \
+  --file-key test_external \
+  --matrix "cuda=${RAPIDS_CUDA_VERSION%.*};arch=$(arch);py=${RAPIDS_PY_VERSION}" | tee env.yaml
 
+rapids-mamba-retry env create --yes -f env.yaml -n test_external
+
+# Temporarily allow unbound variables for conda activation.
+set +u
 conda activate test_external
+set -u
+
+rapids-print-env
 
 # Define input parameter
 PROJECT=$1
@@ -19,6 +29,11 @@ LIBRARIES=("datashader" "holoviews")
 # Change directory to /tmp
 pushd /tmp
 
+
+EXITCODE=0
+trap "EXITCODE=1" ERR
+set +e
+
 # Clone the specified Python libraries
 if [ "$PROJECT" = "all" ]; then
     # Loop through each library and install dependencies
@@ -33,7 +48,11 @@ if [ "$PROJECT" = "all" ]; then
         # Change directory to the library
         pushd $LIBRARY
         # Run setup.py with test dependencies
-        python -m pip install -e .[tests]
+        python -m pip install .[tests]
+
+        rapids-logger "Run GPU tests for $LIBRARY"
+
+        python -m pytest $LIBRARY/tests/ --numprocesses=8 --dist=worksteal --gpu
 
         popd
     done
@@ -53,42 +72,14 @@ else
     # Change directory to the specified project
     pushd $PROJECT
     # Run setup.py with test dependencies
-    python -m pip install -e .[tests]
-    popd
-fi
+    python -m pip install .[tests]
 
-FILES=""
-# Install and run tests
-if [ "$PROJECT" = "all" ]; then
-    # Loop through each library and install dependencies
-    for LIBRARY in "${LIBRARIES[@]}"
-    do
-        rapids-logger "gathering GPU tests for $LIBRARY"
-        TEST_DIR="$LIBRARY/$LIBRARY/tests"
-        # Find all Python scripts containing the keywords cudf or dask_cudf except test_quadmesh.py
-        FILES+=" $(grep -l -R -e 'cudf' --include='*.py' "$TEST_DIR" | grep -v test_quadmesh.py)"
-    done
-else
-    rapids-logger "gathering GPU tests for $PROJECT"
-    TEST_DIR="$PROJECT/$PROJECT/tests"
-    # Find all Python scripts containing the keywords cudf or dask_cudf
-    FILES+=$(grep -l -R -e 'cudf' --include='*.py' "$TEST_DIR")
-fi
 
-EXITCODE=0
-trap "EXITCODE=1" ERR
-set +e
+    rapids-logger "Run GPU tests for $LIBRARY"
 
-rapids-logger "running all gathered tests"
-DATASHADER_TEST_GPU=1 pytest \
-  --numprocesses=8 \
-  --dist=worksteal \
-  $FILES
+    python -m pytest $LIBRARY/tests/ --numprocesses=8 --dist=worksteal --gpu
 
-if [[ "$PROJECT" = "all" ]] || [[ "$PROJECT" = "datashader" ]]; then
-    # run test_quadmesh.py separately as dask.array tests fail with numprocesses
-    rapids-logger "running test_quadmesh.py"
-    DATASHADER_TEST_GPU=1 pytest datashader/datashader/tests/test_quadmesh.py
+    popd
 fi
 
 rapids-logger "Test script exiting with value: $EXITCODE"

dependencies.yaml

@@ -70,6 +70,10 @@ files:
       key: test
     includes:
       - test_python
+  test_external:
+    output: none
+    includes:
+      - test_external
 channels:
   - rapidsai
   - rapidsai-nightly
@@ -257,3 +261,21 @@ dependencies:
       - output_types: conda
         packages:
           - panel>=1.0,<=1.3.4
+  test_external:
+    common:
+      - output_types: [conda]
+        packages:
+          - coverage
+          - *cudf_unsuffixed
+          - cupy>=12.0.0
+          - *cuspatial_unsuffixed
+          - *dask_cudf_unsuffixed
+          - cuxfilter==24.10.*,>=0.0.0a0
+          - python>=3.9,<3.12
+          - xarray-spatial
+          - pycaret
+          - graphistry
+          - dash
+          - dask-sql
+          - pytest-benchmark
+          - pytest-xdist

python/cuxfilter/charts/core/non_aggregate/core_graph.py

@@ -50,14 +50,14 @@ def edge_columns(self):
 
     @property
     def node_columns(self):
-        if self.node_aggregate_col:
-            return [
-                self.node_id,
-                self.node_x,
-                self.node_y,
-                self.node_aggregate_col,
-            ]
-        return [self.node_id, self.node_x, self.node_y]
+        cols = [
+            self.node_id,
+            self.node_x,
+            self.node_y,
+        ]
+        if self.node_aggregate_col != self.node_id:
+            cols.append(self.node_aggregate_col)
+        return cols
 
     def __init__(
         self,