feat: implement UserDefinedStratigraphyAlgorithm

Author: rabii-chaarani

m2l/processing/algorithms/user_defined_sorter.py

@@ -0,0 +1,112 @@
+from typing import Any, Optional
+from osgeo import gdal
+import numpy as np
+import json
+
+from PyQt5.QtCore import QVariant
+from qgis import processing
+from qgis.core import (
+    QgsFeatureSink,
+    QgsFields, 
+    QgsField, 
+    QgsFeature, 
+    QgsGeometry,
+    QgsRasterLayer,
+    QgsProcessing,
+    QgsProcessingAlgorithm,
+    QgsProcessingContext,
+    QgsProcessingException,
+    QgsProcessingFeedback,
+    QgsProcessingParameterEnum,
+    QgsProcessingParameterFileDestination,
+    QgsProcessingParameterFeatureSink,
+    QgsProcessingParameterFeatureSource,
+    QgsProcessingParameterField,
+    QgsProcessingParameterRasterLayer,
+    QgsProcessingParameterMatrix,
+    QgsCoordinateReferenceSystem,
+    QgsVectorLayer,
+    QgsWkbTypes,
+    QgsSettings
+)
+
+
+from qgis.core import (
+    QgsFields, QgsField, QgsFeature, QgsFeatureSink, QgsWkbTypes,
+    QgsCoordinateReferenceSystem, QgsProcessingAlgorithm, QgsProcessingContext,
+    QgsProcessingFeedback, QgsProcessingParameterFeatureSink, QgsProcessingParameterMatrix,
+    QgsSettings
+)
+from PyQt5.QtCore import QVariant
+import numpy as np
+
+class UserDefinedStratigraphyAlgorithm(QgsProcessingAlgorithm):
+    INPUT_STRATI_COLUMN = "INPUT_STRATI_COLUMN"
+    OUTPUT = "OUTPUT"
+
+    def name(self): return "loop_sorter_2"
+    def displayName(self): return "User-Defined Stratigraphic Column"
+    def group(self): return "Stratigraphy"
+    def groupId(self): return "Stratigraphy_Column"
+
+    def initAlgorithm(self, config=None):
+        strati_settings = QgsSettings()
+        last_strati_column = strati_settings.value("m2l/strati_column", "")
+        self.addParameter(
+            QgsProcessingParameterMatrix(
+                name=self.INPUT_STRATI_COLUMN,
+                description="Stratigraphic Order",
+                headers=["Unit"],
+                numberRows=0,
+                defaultValue=last_strati_column
+            )
+        )
+        self.addParameter(
+            QgsProcessingParameterFeatureSink(
+                self.OUTPUT,
+                "Stratigraphic column",
+            )
+        )
+
+    def processAlgorithm(self, parameters, context, feedback):
+        # 1) Read the matrix; it may be a list of lists (rows) or a flat list depending on input source.
+        matrix = self.parameterAsMatrix(parameters, self.INPUT_STRATI_COLUMN, context)
+
+        # Normalize to a list of unit strings (one column: "Unit")
+        units = []
+        for row in matrix:
+            if isinstance(row, (list, tuple)):
+                unit = row[0] if row else ""
+            else:
+                unit = row
+            unit = (unit or "").strip()
+            if unit:  # skip empty rows to avoid writing "" into fields
+                units.append(unit)
+
+        # 2) Build sequential order (1-based), cast to native int
+        order_vals = [int(i) for i in (np.arange(len(units)) + 1)]
+
+        # 3) Prepare sink
+        sink_fields = QgsFields()
+        sink_fields.append(QgsField("order", QVariant.Int))      # or QVariant.LongLong
+        sink_fields.append(QgsField("unit_name", QVariant.String))
+
+        crs = context.project().crs() if context and context.project() else QgsCoordinateReferenceSystem()
+        sink, dest_id = self.parameterAsSink(
+            parameters, self.OUTPUT, context,
+            sink_fields, QgsWkbTypes.NoGeometry, crs
+        )
+
+        # 4) Insert features
+        for pos, unit_name in zip(order_vals, units):
+            f = QgsFeature(sink_fields)
+            # Ensure correct types: int for "order", str for "unit_name"
+            f.setAttributes([int(pos), str(unit_name)])
+            ok = sink.addFeature(f, QgsFeatureSink.FastInsert)
+            if not ok:
+                feedback.reportError(f"Failed to add feature for unit '{unit_name}' (order={pos}).")
+
+        return {self.OUTPUT: dest_id}
+
+    def createInstance(self):
+        return __class__()