feat: 百度通用文字识别提取供应商/日期（config api_key 持久化）

app/core/db/order_metadata_db.py

@@ -0,0 +1,122 @@
+"""单据元信息 SQLite 存储。
+
+独立于 ProductDatabase，与 product_cache.db 共用文件。
+提供：save / get / list_all / count。
+"""
+from __future__ import annotations
+
+import os
+import sqlite3
+import logging
+from datetime import datetime
+from typing import Dict, List, Optional
+
+logger = logging.getLogger(__name__)
+
+
+class OrderMetadataDB:
+    """单据元信息库（共用 product_cache.db 文件，独立表 order_metadata）。"""
+
+    SCHEMA = """
+    CREATE TABLE IF NOT EXISTS order_metadata (
+        file_hash         TEXT PRIMARY KEY,
+        supplier          TEXT DEFAULT '',
+        bill_date          TEXT DEFAULT '',
+        total_amount       REAL DEFAULT 0.0,
+        raw_supplier_text  TEXT DEFAULT '',
+        source_image       TEXT DEFAULT '',
+        created_at         TEXT DEFAULT '',
+        updated_at         TEXT DEFAULT ''
+    );
+    """
+
+    def __init__(self, db_path: str):
+        self.db_path = db_path
+        self._ensure_table()
+
+    def _connect(self) -> sqlite3.Connection:
+        return sqlite3.connect(self.db_path)
+
+    def _ensure_table(self):
+        """幂等创建表。"""
+        try:
+            os.makedirs(os.path.dirname(self.db_path) or '.', exist_ok=True)
+        except OSError:
+            pass
+        conn = self._connect()
+        try:
+            conn.execute(self.SCHEMA)
+            conn.commit()
+        finally:
+            conn.close()
+
+    def save(self, file_hash: str, supplier: str = '', bill_date: str = '',
+             total_amount: float = 0.0, raw_supplier_text: str = '',
+             source_image: str = '') -> bool:
+        """保存或更新单据元信息（PK = file_hash）。"""
+        if not file_hash:
+            return False
+        now = datetime.now().isoformat(timespec='seconds')
+        conn = self._connect()
+        try:
+            # 使用 UPSERT 语义
+            conn.execute("""
+                INSERT INTO order_metadata
+                    (file_hash, supplier, bill_date, total_amount,
+                     raw_supplier_text, source_image, created_at, updated_at)
+                VALUES (?, ?, ?, ?, ?, ?, ?, ?)
+                ON CONFLICT(file_hash) DO UPDATE SET
+                    supplier = excluded.supplier,
+                    bill_date = excluded.bill_date,
+                    total_amount = excluded.total_amount,
+                    raw_supplier_text = excluded.raw_supplier_text,
+                    source_image = excluded.source_image,
+                    updated_at = excluded.updated_at
+            """, (file_hash, supplier, bill_date, total_amount,
+                  raw_supplier_text, source_image, now, now))
+            conn.commit()
+            return True
+        except Exception as e:
+            logger.error(f"save_order_metadata 失败: {e}")
+            return False
+        finally:
+            conn.close()
+
+    def get(self, file_hash: str) -> Optional[Dict]:
+        conn = self._connect()
+        try:
+            conn.row_factory = sqlite3.Row
+            row = conn.execute(
+                "SELECT * FROM order_metadata WHERE file_hash=?",
+                (file_hash,)).fetchone()
+            return dict(row) if row else None
+        finally:
+            conn.close()
+
+    def list_all(self, limit: int = 1000) -> List[Dict]:
+        """列出所有元信息，按 updated_at 倒序。"""
+        conn = self._connect()
+        try:
+            conn.row_factory = sqlite3.Row
+            rows = conn.execute(
+                "SELECT * FROM order_metadata ORDER BY updated_at DESC LIMIT ?",
+                (limit,)).fetchall()
+            return [dict(r) for r in rows]
+        finally:
+            conn.close()
+
+    def count(self) -> int:
+        conn = self._connect()
+        try:
+            return conn.execute("SELECT COUNT(*) FROM order_metadata").fetchone()[0]
+        finally:
+            conn.close()
+
+    def delete(self, file_hash: str) -> bool:
+        conn = self._connect()
+        try:
+            conn.execute("DELETE FROM order_metadata WHERE file_hash=?", (file_hash,))
+            conn.commit()
+            return True
+        finally:
+            conn.close()