feat(validation): add SQLModel database models and fix relationships

src/macro_lactone_toolkit/validation/database.py

@@ -0,0 +1,26 @@
+from __future__ import annotations
+
+from contextlib import contextmanager
+from pathlib import Path
+
+from sqlmodel import Session, SQLModel, create_engine
+
+
+def get_engine(db_path: str | Path):
+    """Create SQLite engine."""
+    db_path = Path(db_path)
+    db_path.parent.mkdir(parents=True, exist_ok=True)
+    url = f"sqlite:///{db_path}"
+    return create_engine(url, echo=False)
+
+
+@contextmanager
+def get_session(engine):
+    """Context manager for database sessions."""
+    with Session(engine) as session:
+        yield session
+
+
+def init_database(engine):
+    """Create all tables."""
+    SQLModel.metadata.create_all(engine)

src/macro_lactone_toolkit/validation/models.py

@@ -1,25 +1,26 @@
 from __future__ import annotations
 
 from datetime import datetime
-from enum import Enum
 from typing import List, Optional
 
-from sqlmodel import Field, Relationship, SQLModel
+from sqlalchemy.orm import Mapped, mapped_column, relationship
+from sqlmodel import Field, SQLModel
 
 
-class ClassificationType(str, Enum):
+class ClassificationType:
     STANDARD = "standard_macrolactone"
     NON_STANDARD = "non_standard_macrocycle"
     NOT_MACROLACTONE = "not_macrolactone"
 
 
-class ProcessingStatus(str, Enum):
+class ProcessingStatus:
     PENDING = "pending"
     SUCCESS = "success"
     FAILED = "failed"
     SKIPPED = "skipped"
 
 
+# Define all tables without relationships first
 class ParentMolecule(SQLModel, table=True):
     """Original molecule information."""
 
@@ -29,11 +30,11 @@ class ParentMolecule(SQLModel, table=True):
     source_id: str = Field(index=True)
     molecule_name: Optional[str] = None
     smiles: str = Field(index=True)
-    classification: ClassificationType = Field(index=True)
+    classification: str = Field(index=True)
     ring_size: Optional[int] = Field(default=None, index=True)
     primary_reason_code: Optional[str] = None
     primary_reason_message: Optional[str] = None
-    processing_status: ProcessingStatus = Field(default=ProcessingStatus.PENDING)
+    processing_status: str = Field(default=ProcessingStatus.PENDING)
     error_message: Optional[str] = None
     num_sidechains: Optional[int] = None
     cleavage_positions: Optional[str] = None
@@ -41,9 +42,6 @@ class ParentMolecule(SQLModel, table=True):
     created_at: datetime = Field(default_factory=datetime.utcnow)
     processed_at: Optional[datetime] = None
 
-    fragments: List["SideChainFragment"] = Relationship(back_populates="parent")
-    numbering: Optional["RingNumbering"] = Relationship(back_populates="parent")
-
 
 class RingNumbering(SQLModel, table=True):
     """Ring numbering details."""
@@ -58,8 +56,6 @@ class RingNumbering(SQLModel, table=True):
     position_to_atom: str
     atom_to_position: str
 
-    parent: Optional[ParentMolecule] = Relationship(back_populates="numbering")
-
 
 class SideChainFragment(SQLModel, table=True):
     """Side chain fragments from cleavage."""
@@ -81,8 +77,6 @@ class SideChainFragment(SQLModel, table=True):
     original_bond_type: str
     image_path: Optional[str] = None
 
-    parent: Optional[ParentMolecule] = Relationship(back_populates="fragments")
-
 
 class ValidationResult(SQLModel, table=True):
     """Manual validation records."""