feat: Add MySQL to PostgreSQL migration tool with JSONB transformation

Implement comprehensive migration solution with:
- Full and incremental migration modes
- JSONB schema transformation for RAWDATACOR and ELABDATADISP tables
- Native PostgreSQL partitioning (2014-2031)
- Optimized GIN indexes for JSONB queries
- Rich logging with progress tracking
- Complete benchmark system for MySQL vs PostgreSQL comparison
- CLI interface with multiple commands (setup, migrate, benchmark)
- Configuration management via .env file
- Error handling and retry logic
- Batch processing for performance (configurable batch size)

Database transformations:
- RAWDATACOR: 16 Val columns + units → single JSONB measurements
- ELABDATADISP: 25+ measurement fields → structured JSONB with categories

🤖 Generated with Claude Code

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

src/transformers/data_transformer.py

@@ -0,0 +1,178 @@
+"""Data transformation from MySQL to PostgreSQL format."""
+from typing import Dict, Any, List
+from datetime import datetime
+from config import (
+    RAWDATACOR_COLUMNS,
+    ELABDATADISP_FIELD_MAPPING,
+    TABLE_CONFIGS,
+)
+from src.utils.logger import get_logger
+
+logger = get_logger(__name__)
+
+
+class DataTransformer:
+    """Transform MySQL data to PostgreSQL format."""
+
+    @staticmethod
+    def transform_rawdatacor_row(mysql_row: Dict[str, Any]) -> Dict[str, Any]:
+        """Transform a RAWDATACOR row from MySQL to PostgreSQL format.
+
+        Args:
+            mysql_row: Row dictionary from MySQL
+
+        Returns:
+            Transformed row dictionary for PostgreSQL
+        """
+        # Create measurements JSONB
+        measurements = {}
+
+        # Map Val0-ValF with their units
+        for i, val_col in enumerate(RAWDATACOR_COLUMNS["val_columns"]):
+            unit_col = RAWDATACOR_COLUMNS["unit_columns"][i]
+
+            value = mysql_row.get(val_col)
+            unit = mysql_row.get(unit_col)
+
+            # Only add to JSONB if value is not None
+            if value is not None:
+                measurements[str(i)] = {
+                    "value": str(value),
+                    "unit": unit if unit else None,
+                }
+
+        # Create PostgreSQL row
+        pg_row = {
+            "id": mysql_row["id"],
+            "unit_name": mysql_row.get("UnitName"),
+            "tool_name_id": mysql_row["ToolNameID"],
+            "node_num": mysql_row["NodeNum"],
+            "event_date": mysql_row["EventDate"],
+            "event_time": mysql_row["EventTime"],
+            "bat_level": mysql_row["BatLevel"],
+            "temperature": mysql_row["Temperature"],
+            "measurements": measurements,
+            "created_at": mysql_row.get("created_at"),
+            "bat_level_module": mysql_row.get("BatLevelModule"),
+            "temperature_module": mysql_row.get("TemperatureModule"),
+            "rssi_module": mysql_row.get("RssiModule"),
+        }
+
+        return pg_row
+
+    @staticmethod
+    def transform_elabdatadisp_row(mysql_row: Dict[str, Any]) -> Dict[str, Any]:
+        """Transform an ELABDATADISP row from MySQL to PostgreSQL format.
+
+        Args:
+            mysql_row: Row dictionary from MySQL
+
+        Returns:
+            Transformed row dictionary for PostgreSQL
+        """
+        # Create measurements JSONB with structured categories
+        measurements = {
+            "shifts": {},
+            "coordinates": {},
+            "kinematics": {},
+            "sensors": {},
+            "calculated": {},
+        }
+
+        # Map all measurement fields using the configuration
+        for mysql_col, (category, pg_key) in ELABDATADISP_FIELD_MAPPING.items():
+            value = mysql_row.get(mysql_col)
+            if value is not None:
+                measurements[category][pg_key] = float(value) if isinstance(value, str) else value
+
+        # Remove empty categories
+        measurements = {
+            k: v for k, v in measurements.items() if v
+        }
+
+        # Create PostgreSQL row
+        pg_row = {
+            "id_elab_data": mysql_row["idElabData"],
+            "unit_name": mysql_row.get("UnitName"),
+            "tool_name_id": mysql_row["ToolNameID"],
+            "node_num": mysql_row["NodeNum"],
+            "event_date": mysql_row["EventDate"],
+            "event_time": mysql_row["EventTime"],
+            "state": mysql_row.get("State"),
+            "calc_err": mysql_row.get("calcerr", 0),
+            "measurements": measurements,
+            "created_at": mysql_row.get("created_at"),
+            "updated_at": mysql_row.get("updated_at"),
+        }
+
+        return pg_row
+
+    @staticmethod
+    def transform_batch(
+        table: str,
+        rows: List[Dict[str, Any]]
+    ) -> List[Dict[str, Any]]:
+        """Transform a batch of rows from MySQL to PostgreSQL format.
+
+        Args:
+            table: Table name ('RAWDATACOR' or 'ELABDATADISP')
+            rows: List of row dictionaries from MySQL
+
+        Returns:
+            List of transformed row dictionaries for PostgreSQL
+        """
+        if table == "RAWDATACOR":
+            return [
+                DataTransformer.transform_rawdatacor_row(row)
+                for row in rows
+            ]
+        elif table == "ELABDATADISP":
+            return [
+                DataTransformer.transform_elabdatadisp_row(row)
+                for row in rows
+            ]
+        else:
+            raise ValueError(f"Unknown table: {table}")
+
+    @staticmethod
+    def get_column_order(table: str) -> List[str]:
+        """Get the column order for inserting into PostgreSQL.
+
+        Args:
+            table: PostgreSQL table name
+
+        Returns:
+            List of column names in order
+        """
+        if table == "rawdatacor":
+            return [
+                "id",
+                "unit_name",
+                "tool_name_id",
+                "node_num",
+                "event_date",
+                "event_time",
+                "bat_level",
+                "temperature",
+                "measurements",
+                "created_at",
+                "bat_level_module",
+                "temperature_module",
+                "rssi_module",
+            ]
+        elif table == "elabdatadisp":
+            return [
+                "id_elab_data",
+                "unit_name",
+                "tool_name_id",
+                "node_num",
+                "event_date",
+                "event_time",
+                "state",
+                "calc_err",
+                "measurements",
+                "created_at",
+                "updated_at",
+            ]
+        else:
+            raise ValueError(f"Unknown table: {table}")