feat: add real PDF ingestion workflow

services/api/app/repositories/postgres.py

@@ -0,0 +1,84 @@
+from __future__ import annotations
+
+import json
+from functools import lru_cache
+
+import psycopg
+from psycopg.types.json import Jsonb
+
+from app.domain.models import ProjectSnapshot
+from app.repositories.memory import create_demo_project
+
+
+class PostgresProjectRepository:
+    def __init__(self, database_url: str) -> None:
+        self.database_url = database_url
+        self.ensure_schema()
+        if self.get("demo-apartment") is None:
+            self.save(create_demo_project())
+
+    def _connect(self):
+        return psycopg.connect(self.database_url)
+
+    def ensure_schema(self) -> None:
+        with self._connect() as connection, connection.cursor() as cursor:
+            cursor.execute(
+                """
+                CREATE TABLE IF NOT EXISTS projects (
+                    project_id TEXT PRIMARY KEY,
+                    name TEXT NOT NULL,
+                    stage TEXT NOT NULL,
+                    revision INTEGER NOT NULL,
+                    snapshot JSONB NOT NULL,
+                    updated_at TIMESTAMPTZ NOT NULL
+                )
+                """
+            )
+
+    def get(self, project_id: str) -> ProjectSnapshot | None:
+        with self._connect() as connection, connection.cursor() as cursor:
+            cursor.execute("SELECT snapshot FROM projects WHERE project_id = %s", (project_id,))
+            row = cursor.fetchone()
+        if row is None:
+            return None
+        payload = json.loads(row[0]) if isinstance(row[0], str) else row[0]
+        return ProjectSnapshot.model_validate(payload)
+
+    def list(self) -> list[ProjectSnapshot]:
+        with self._connect() as connection, connection.cursor() as cursor:
+            cursor.execute("SELECT snapshot FROM projects ORDER BY updated_at DESC")
+            rows = cursor.fetchall()
+        return [
+            ProjectSnapshot.model_validate(json.loads(row[0]) if isinstance(row[0], str) else row[0])
+            for row in rows
+        ]
+
+    def save(self, project: ProjectSnapshot) -> ProjectSnapshot:
+        payload = project.model_dump(mode="json")
+        with self._connect() as connection, connection.cursor() as cursor:
+            cursor.execute(
+                """
+                INSERT INTO projects (project_id, name, stage, revision, snapshot, updated_at)
+                VALUES (%s, %s, %s, %s, %s, %s)
+                ON CONFLICT (project_id) DO UPDATE SET
+                    name = EXCLUDED.name,
+                    stage = EXCLUDED.stage,
+                    revision = EXCLUDED.revision,
+                    snapshot = EXCLUDED.snapshot,
+                    updated_at = EXCLUDED.updated_at
+                """,
+                (
+                    project.project_id,
+                    project.name,
+                    project.stage.value,
+                    project.revision,
+                    Jsonb(payload),
+                    project.updated_at,
+                ),
+            )
+        return project
+
+
+@lru_cache(maxsize=4)
+def postgres_repository(database_url: str) -> PostgresProjectRepository:
+    return PostgresProjectRepository(database_url)