refactor: Move PVMergeBlocksEnhanced plugin and create the VTK filter (#129)

* Modify isAttributeGlobal function to take into account recursive multiblockdataset

* Adding a VTK filter for enhanced merge block

* Move and update plugin

* Fix import format

* Typing

* Adding tests and data for test for merge blocks

* Documentation

* Fix docstring and typing

* error handling version

* adding Fails test

* Capture errors from VTK

* Update python-package.yml as order might matters

* CI is testing vtk9.5.1 so skip the failing test for now

* wrong version dispatch

* Clean and formatting

* Add child logger to prevent to much verbosity in parent logger

---------
Co-authored-by: Jacques Franc <[email protected]>

Author: paloma-martinez

.github/workflows/python-package.yml

@@ -1,4 +1,4 @@
-name: geosPythonPackages CI 
+name: geosPythonPackages CI
 on: pull_request
 
 # Cancels in-progress workflows for a PR when updated
@@ -9,13 +9,13 @@ concurrency:
 
 jobs:
   # Checks if PR title follows conventional semantics
-  semantic_pull_request:    
+  semantic_pull_request:
     permissions:
-      pull-requests: write  # for amannn/action-semantic-pull-request to analyze PRs and 
+      pull-requests: write  # for amannn/action-semantic-pull-request to analyze PRs and
       statuses: write  # for amannn/action-semantic-pull-request to mark status of analyzed PR
-      contents: read    
+      contents: read
     runs-on: ubuntu-latest
-    
+
     steps:
       - name: Check if the PR name has conventional semantics
         if: github.event_name == 'pull_request'
@@ -27,7 +27,7 @@ jobs:
           wip: true
           # Configure that a scope doesn't need to be provided.
           requireScope: false
-    
+
       - name: Skip the check on main branch
         if:  github.ref_name == 'main'
         run: |
@@ -40,14 +40,14 @@ jobs:
       max-parallel: 3
       matrix:
         python-version: ["3.10", "3.11", "3.12"]
-        package-name: 
+        package-name:
           - geos-ats
+          - geos-utils
           - geos-geomechanics
           - geos-mesh
           - geos-posp
           - geos-timehistory
           - geos-trame
-          - geos-utils
           - geos-xml-tools
           - geos-xml-viewer
           - hdf5-wrapper
@@ -77,7 +77,7 @@ jobs:
         python -m pip install --upgrade pip
         python -m pip install pytest yapf toml
 
-        DEPS="${{ matrix.dependencies || '' }}"  
+        DEPS="${{ matrix.dependencies || '' }}"
 
         if [ -n "$DEPS" ]; then
         echo "Installing additional dependencies: $DEPS"
@@ -114,20 +114,20 @@ jobs:
       uses: actions/checkout@v4
       with:
         fetch-depth: 0  # Fetch all history to compare with base branch
-    
+
     - name: Check if GEOS integration is required
       id: check_changes
       run: |
         echo "Analyzing changed files to determine if GEOS integration test is required..."
-        
+
         # Get list of changed files
         git fetch origin ${{ github.base_ref }}
         CHANGED_FILES=$(git diff --name-only origin/${{ github.base_ref }}...HEAD)
-        
+
         echo "Changed files:"
         echo "$CHANGED_FILES"
         echo ""
-        
+
         # Define packages that are integrated into GEOS (from GEOS/scripts/setupPythonEnvironment.bash)
         GEOS_INTEGRATED_PACKAGES=(
           "geos-utils"
@@ -137,7 +137,7 @@ jobs:
           "pygeos-tools"
           "geos-ats"
         )
-        
+
         # Define patterns that DON'T require GEOS integration testing
         SKIP_PATTERNS=(
           "^docs/"
@@ -158,21 +158,21 @@ jobs:
           "^geos-trame/"
           "^geos-xml-viewer/"
         )
-        
+
         # Check if label is present (overrides automatic detection)
         HAS_LABEL=$(echo '${{ toJson(github.event.pull_request.labels.*.name) }}' | grep -q "test-geos-integration" && echo "true" || echo "false")
-        
+
         if [[ "$HAS_LABEL" == "true" ]]; then
           echo "✓ Label 'test-geos-integration' found - GEOS integration test will run"
           echo "required=true" >> "$GITHUB_OUTPUT"
           echo "skip_reason=none" >> "$GITHUB_OUTPUT"
           exit 0
         fi
-        
+
         # Check if any changed file affects GEOS-integrated packages
         REQUIRES_GEOS_TEST=false
         AFFECTED_PACKAGES=""
-        
+
         for file in $CHANGED_FILES; do
           # Check if file matches any skip pattern
           SHOULD_SKIP=false
@@ -182,7 +182,7 @@ jobs:
               break
             fi
           done
-          
+
           if [[ "$SHOULD_SKIP" == "false" ]]; then
             # Check if file is in a GEOS-integrated package
             for package in "${GEOS_INTEGRATED_PACKAGES[@]}"; do
@@ -193,21 +193,21 @@ jobs:
                 fi
               fi
             done
-            
+
             # Check for CI workflow changes that affect GEOS integration
             if echo "$file" | grep -qE "^\.github/workflows/(python-package\.yml|test_geos_integration\.yml)$"; then
               REQUIRES_GEOS_TEST=true
               AFFECTED_PACKAGES="$AFFECTED_PACKAGES [CI-workflows]"
             fi
-            
+
             # Check for root-level scripts that might affect integration
             if echo "$file" | grep -qE "^install_packages\.sh$"; then
               REQUIRES_GEOS_TEST=true
               AFFECTED_PACKAGES="$AFFECTED_PACKAGES [install-scripts]"
             fi
           fi
         done
-        
+
         if [[ "$REQUIRES_GEOS_TEST" == "true" ]]; then
           echo "✓ GEOS integration test REQUIRED"
           echo "  Affected packages/components:$AFFECTED_PACKAGES"
@@ -240,11 +240,11 @@ jobs:
       run: |
         echo "Final CI Validation"
         echo "==================="
-        
+
         GEOS_REQUIRED="${{ needs.check_geos_integration_required.outputs.geos_integration_required }}"
         SKIP_REASON="${{ needs.check_geos_integration_required.outputs.skip_reason }}"
         GEOS_RESULT="${{ needs.geos_integration_test.result }}"
-        
+
         if [[ "$GEOS_REQUIRED" == "true" ]]; then
           echo "GEOS integration test was required and triggered"
           if [[ "$GEOS_RESULT" == "success" ]]; then

docs/geos_mesh_docs/processing.rst

@@ -45,10 +45,20 @@ geos.mesh.processing.SplitMesh filter
     :undoc-members:
     :show-inheritance:
 
+
+geos.mesh.processing.MergeBlockEnhanced filter
+------------------------------------------------
+
+.. automodule:: geos.mesh.processing.MergeBlockEnhanced
+    :members:
+    :undoc-members:
+    :show-inheritance:
+
+
 geos.mesh.processing.ClipToMainFrame filter
 --------------------------------------------
 
 .. automodule:: geos.mesh.processing.ClipToMainFrame
     :members:
     :undoc-members:
-    :show-inheritance:
+    :show-inheritance:

docs/geos_posp_docs/PVplugins.rst

@@ -64,11 +64,6 @@ PVGeomechanicsWorkflowVolumeWell plugin
 
 .. automodule:: PVplugins.PVGeomechanicsWorkflowVolumeWell
 
-PVplugins.PVMergeBlocksEnhanced module
---------------------------------------
-
-.. automodule:: PVplugins.PVMergeBlocksEnhanced
-
 
 PVMohrCirclePlot plugin
 ---------------------------------

docs/geos_pv_docs/processing.rst

@@ -24,6 +24,12 @@ PVSplitMesh
 .. automodule:: geos.pv.plugins.PVSplitMesh
 
 
+PVMergeBlocksEnhanced module
+--------------------------------------
+
+.. automodule:: geos.pv.plugins.PVMergeBlocksEnhanced
+
+
 PVClipToMainFrame
 ----------------------------------
 

geos-mesh/src/geos/mesh/processing/CreateConstantAttributePerRegion.py

@@ -2,8 +2,9 @@
 # SPDX-FileCopyrightText: Copyright 2023-2024 TotalEnergies.
 # SPDX-FileContributor: Romain Baville
 import numpy as np
-import numpy.typing as npt
+import logging
 
+import numpy.typing as npt
 from typing import Union, Any
 from typing_extensions import Self
 
@@ -13,7 +14,7 @@
     vtkDataSet,
 )
 
-from geos.utils.Logger import ( getLogger, Logger, logging, CountWarningHandler )
+from geos.utils.Logger import ( getLogger, Logger, CountWarningHandler )
 from geos.mesh.utils.arrayHelpers import ( getArrayInObject, getComponentNames, getNumberOfComponents,
                                            getVtkDataTypeInObject, isAttributeGlobal, getAttributePieceInfo,
                                            checkValidValuesInDataSet, checkValidValuesInMultiBlock )

geos-mesh/src/geos/mesh/processing/MergeBlockEnhanced.py

@@ -0,0 +1,128 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright 2023-2024 TotalEnergies.
+# SPDX-FileContributor: Paloma Martinez
+# ruff: noqa: E402 # disable Module level import not at top of file
+import logging
+
+from typing_extensions import Self
+
+from geos.utils.Logger import Logger, getLogger
+from geos.mesh.utils.multiblockModifiers import mergeBlocks
+
+from vtkmodules.vtkCommonDataModel import (
+    vtkMultiBlockDataSet,
+    vtkUnstructuredGrid,
+)
+
+__doc__ = """
+Merge Blocks Keeping Partial Attributes is a filter that allows to merge blocks from a multiblock dataset while keeping partial attributes.
+
+Input is a vtkMultiBlockDataSet and output is a vtkUnstructuredGrid.
+
+.. Note::
+    - You may encounter issues if two datasets of the input multiblock dataset have duplicated cell IDs.
+    - Partial attributes are filled with default values depending on their types.
+        - 0 for uint data.
+        - -1 for int data.
+        - nan for float data.
+
+
+To use it:
+
+.. code-block:: python
+
+    from geos.mesh.processing.MergeBlockEnhanced import MergeBlockEnhanced
+    import logging
+    from geos.utils.Errors import VTKError
+
+    # Define filter inputs
+    multiblockdataset: vtkMultiblockDataSet
+    speHandler: bool # optional
+
+    # Instantiate the filter
+    filter: MergeBlockEnhanced = MergeBlockEnhanced( multiblockdataset, speHandler )
+
+    # Use your own handler (if speHandler is True)
+    yourHandler: logging.Handler
+    filter.setLoggerHandler( yourHandler )
+
+    # Do calculations
+    try:
+        filter.applyFilter()
+    except VTKError:
+        logging.error("Something went wrong in VTK")
+
+    # Get the merged mesh
+    filter.getOutput()
+"""
+
+loggerTitle: str = "Merge Block Enhanced"
+
+
+class MergeBlockEnhanced:
+
+    def __init__(
+        self: Self,
+        inputMesh: vtkMultiBlockDataSet,
+        speHandler: bool = False,
+    ) -> None:
+        """Merge a multiblock dataset and keep the partial attributes in the output mesh.
+
+        Partial attributes are filled with default values depending on the data type such that:
+            - 0 for uint data.
+            - -1 for int data.
+            - nan for float data.
+
+        Args:
+            inputMesh (vtkMultiBlockDataSet): The input multiblock dataset to merge.
+            speHandler (bool, optional) : True to use a specific handler, False to use the internal handler.
+            Defaults to False.
+        """
+        self.inputMesh: vtkMultiBlockDataSet = inputMesh
+        self.outputMesh: vtkUnstructuredGrid = vtkUnstructuredGrid()
+
+        # Logger
+        self.logger: Logger
+        if not speHandler:
+            self.logger = getLogger( loggerTitle, True )
+        else:
+            self.logger = logging.getLogger( loggerTitle )
+            self.logger.setLevel( logging.INFO )
+
+    def setLoggerHandler( self: Self, handler: logging.Handler ) -> None:
+        """Set a specific handler for the filter logger.
+
+        In this filter 4 log levels are use, .info, .error, .warning and .critical, be sure to have at least the same 4 levels.
+
+        Args:
+            handler (logging.Handler): The handler to add.
+        """
+        if not self.logger.hasHandlers():
+            self.logger.addHandler( handler )
+        else:
+            self.logger.warning(
+                "The logger already has an handler, to use yours set the argument 'speHandler' to True during the filter initialization."
+            )
+
+    def applyFilter( self: Self ) -> None:
+        """Merge the blocks of a multiblock dataset mesh.
+
+        Returns:
+            bool: True if the blocks were successfully merged, False otherwise.
+
+        Raises:
+            VTKError (geos.utils.Errors) : error captured if any from the VTK log
+        """
+        self.logger.info( f"Applying filter { self.logger.name }." )
+
+        outputMesh: vtkUnstructuredGrid
+        outputMesh = mergeBlocks( self.inputMesh, keepPartialAttributes=True, logger=self.logger )
+        self.outputMesh = outputMesh
+
+    def getOutput( self: Self ) -> vtkUnstructuredGrid:
+        """Get the merged mesh.
+
+        Returns:
+            vtkUnstructuredGrid: The merged mesh.
+        """
+        return self.outputMesh

geos-mesh/src/geos/mesh/utils/arrayHelpers.py

@@ -677,7 +677,6 @@ def isAttributeGlobal( multiBlockDataSet: vtkMultiBlockDataSet, attributeName: s
         dataSet: vtkDataSet = vtkDataSet.SafeDownCast( multiBlockDataSet.GetDataSet( blockIndex ) )
         if not isAttributeInObjectDataSet( dataSet, attributeName, onPoints ):
             return False
-
     return True
 
 

geos-mesh/src/geos/mesh/utils/arrayModifiers.py

@@ -762,13 +762,13 @@ def transferAttributeToDataSetWithElementMap(
 
     for idElementTo in range( nbElementsTo ):
         valueToTransfer: Any = defaultValue
-        idElementFrom: int = elementMap[ flatIdDataSetTo ][ idElementTo ][ 1 ]
+        idElementFrom: int = int( elementMap[ flatIdDataSetTo ][ idElementTo ][ 1 ] )
         if idElementFrom != -1:
             dataFrom: Union[ vtkPointData, vtkCellData ]
             if isinstance( meshFrom, vtkDataSet ):
                 dataFrom = meshFrom.GetPointData() if onPoints else meshFrom.GetCellData()
             elif isinstance( meshFrom, vtkMultiBlockDataSet ):
-                flatIdDataSetFrom: int = elementMap[ flatIdDataSetTo ][ idElementTo ][ 0 ]
+                flatIdDataSetFrom: int = int( elementMap[ flatIdDataSetTo ][ idElementTo ][ 0 ] )
                 dataSetFrom: vtkDataSet = vtkDataSet.SafeDownCast( meshFrom.GetDataSet( flatIdDataSetFrom ) )
                 dataFrom = dataSetFrom.GetPointData() if onPoints else dataSetFrom.GetCellData()