Update: Refactor to absolute imports, fix QPS limits, and enhance Gitea sync with SQLite support

app/config/settings.py

@@ -9,8 +9,8 @@ import configparser
 from typing import Dict, List, Optional, Any
 
 from dotenv import load_dotenv
-from ..core.utils.log_utils import get_logger
-from .defaults import DEFAULT_CONFIG
+from app.core.utils.log_utils import get_logger
+from app.config.defaults import DEFAULT_CONFIG
 
 # 加载 .env 文件
 load_dotenv()

app/core/db/product_db.py

@@ -16,9 +16,9 @@ from typing import Dict, List, Optional, Tuple, Callable
 
 import pandas as pd
 
-from ..utils.log_utils import get_logger
-from ..utils.file_utils import smart_read_excel
-from ...core.handlers.column_mapper import ColumnMapper
+from app.core.utils.log_utils import get_logger
+from app.core.utils.file_utils import smart_read_excel
+from app.core.handlers.column_mapper import ColumnMapper
 
 logger = get_logger(__name__)
 
@@ -43,6 +43,13 @@ class ProductDatabase:
         max_price REAL DEFAULT 0.0,
         price_count INTEGER DEFAULT 0
     );
+    CREATE TABLE IF NOT EXISTS missing_barcodes (
+        barcode     TEXT PRIMARY KEY,
+        name        TEXT DEFAULT '',
+        last_seen   TEXT,
+        source_file TEXT,
+        count       INTEGER DEFAULT 1
+    );
     """
 
     _NEW_COLUMNS = {
@@ -90,16 +97,68 @@ class ProductDatabase:
     def _migrate_schema(self):
         conn = self._connect()
         try:
+            # 迁移 products 表
             cursor = conn.execute("PRAGMA table_info(products)")
             existing_cols = {row[1] for row in cursor.fetchall()}
             for col_name, col_type in self._NEW_COLUMNS.items():
                 if col_name not in existing_cols:
                     conn.execute(f"ALTER TABLE products ADD COLUMN {col_name} {col_type}")
                     logger.info(f"数据库迁移: 添加列 {col_name}")
+            
+            # 确保 missing_barcodes 表存在
+            conn.execute("""
+                CREATE TABLE IF NOT EXISTS missing_barcodes (
+                    barcode     TEXT PRIMARY KEY,
+                    name        TEXT DEFAULT '',
+                    last_seen   TEXT,
+                    source_file TEXT,
+                    count       INTEGER DEFAULT 1
+                )
+            """)
             conn.commit()
         finally:
             conn.close()
 
+    # ══════════════════════════════════════════════════════════════
+    # 缺失条码记录
+    # ══════════════════════════════════════════════════════════════
+
+    def record_missing_barcode(self, barcode: str, name: str = '', source_file: str = ''):
+        """记录缺失条码。"""
+        barcode = str(barcode).strip()
+        if not barcode:
+            return
+        now = datetime.now().isoformat(timespec='seconds')
+        conn = self._connect()
+        try:
+            conn.execute("""
+                INSERT INTO missing_barcodes (barcode, name, last_seen, source_file, count)
+                VALUES (?, ?, ?, ?, 1)
+                ON CONFLICT(barcode) DO UPDATE SET
+                    name = CASE WHEN excluded.name != '' THEN excluded.name ELSE name END,
+                    last_seen = excluded.last_seen,
+                    source_file = excluded.source_file,
+                    count = count + 1
+            """, (barcode, name, now, os.path.basename(source_file)))
+            conn.commit()
+            logger.info(f"已记录缺失条码: {barcode} ({name})")
+        except Exception as e:
+            logger.error(f"记录缺失条码失败: {e}")
+        finally:
+            conn.close()
+
+    def get_missing_barcodes(self, limit: int = 100) -> List[Dict]:
+        """获取缺失条码列表。"""
+        conn = self._connect()
+        conn.row_factory = sqlite3.Row
+        try:
+            rows = conn.execute(
+                "SELECT * FROM missing_barcodes ORDER BY last_seen DESC LIMIT ?",
+                (limit,)).fetchall()
+            return [dict(r) for r in rows]
+        finally:
+            conn.close()
+
     # ══════════════════════════════════════════════════════════════
     # 导入
     # ══════════════════════════════════════════════════════════════

app/core/excel/converter.py

@@ -10,13 +10,13 @@ import os
 import json
 from typing import Dict, Tuple, Optional, Any, List, Union
 
-from ..utils.log_utils import get_logger
-from .handlers.barcode_mapper import BarcodeMapper
-from .handlers.unit_converter_handlers import (
+from app.core.utils.log_utils import get_logger
+from app.core.excel.handlers.barcode_mapper import BarcodeMapper
+from app.core.excel.handlers.unit_converter_handlers import (
     JianUnitHandler, BoxUnitHandler, TiHeUnitHandler, 
     GiftUnitHandler, UnitHandler
 )
-from .validators import ProductValidator
+from app.core.excel.validators import ProductValidator
 
 logger = get_logger(__name__)
 

app/core/excel/handlers/__init__.py

@@ -7,5 +7,5 @@
 from typing import Dict, Any
 
 # 导出所有处理程序类
-from .barcode_mapper import BarcodeMapper
-from .unit_converter_handlers import JianUnitHandler, BoxUnitHandler, TiHeUnitHandler, GiftUnitHandler, UnitHandler 
+from app.core.excel.handlers.barcode_mapper import BarcodeMapper
+from app.core.excel.handlers.unit_converter_handlers import JianUnitHandler, BoxUnitHandler, TiHeUnitHandler, GiftUnitHandler, UnitHandler 

app/core/excel/handlers/barcode_mapper.py

@@ -7,7 +7,7 @@
 import logging
 from typing import Dict, Optional, Any
 
-from ...utils.log_utils import get_logger
+from app.core.utils.log_utils import get_logger
 
 logger = get_logger(__name__)
 

app/core/excel/handlers/unit_converter_handlers.py

@@ -8,7 +8,7 @@ import logging
 from typing import Dict, Optional, Any, Tuple, Protocol
 from abc import ABC, abstractmethod
 
-from ...utils.log_utils import get_logger
+from app.core.utils.log_utils import get_logger
 
 logger = get_logger(__name__)
 

app/core/excel/merger.py

@@ -14,17 +14,17 @@ from xlutils.copy import copy as xlcopy
 from typing import Dict, List, Optional, Tuple, Union, Any, Callable
 from datetime import datetime
 
-from ...config.settings import ConfigManager
-from ..utils.log_utils import get_logger
-from ..handlers.column_mapper import ColumnMapper
-from ..utils.file_utils import (
+from app.config.settings import ConfigManager
+from app.core.utils.log_utils import get_logger
+from app.core.handlers.column_mapper import ColumnMapper
+from app.core.utils.file_utils import (
     ensure_dir,
     get_file_extension,
     get_files_by_extensions,
     load_json,
     save_json
 )
-from ..utils.string_utils import (
+from app.core.utils.string_utils import (
     clean_string,
     clean_barcode,
     format_barcode

app/core/excel/processor.py

@@ -14,23 +14,23 @@ from xlutils.copy import copy as xlcopy
 from typing import Dict, List, Optional, Tuple, Union, Any, Callable
 from datetime import datetime
 
-from ...config.settings import ConfigManager
-from ..utils.log_utils import get_logger
-from ..utils.file_utils import (
+from app.config.settings import ConfigManager
+from app.core.utils.log_utils import get_logger
+from app.core.utils.file_utils import (
     ensure_dir,
     get_file_extension,
     get_latest_file,
     load_json,
     save_json
 )
-from ..utils.string_utils import (
+from app.core.utils.string_utils import (
     clean_string,
     extract_number,
     format_barcode,
     parse_monetary_string
 )
-from .converter import UnitConverter
-from ..handlers.column_mapper import ColumnMapper
+from app.core.excel.converter import UnitConverter
+from app.core.handlers.column_mapper import ColumnMapper
 
 logger = get_logger(__name__)
 
@@ -40,15 +40,19 @@ class ExcelProcessor:
     提取条码、单价和数量，并按照采购单模板的格式填充
     """
     
-    def __init__(self, config, product_db=None):
+    def __init__(self, config, product_db=None, missing_barcodes_cb: Optional[Callable[[List[str]], None]] = None):
         """
         初始化Excel处理器
 
         Args:
             config: 配置信息
             product_db: 商品数据库实例（可选，由外部传入以共享）
+            missing_barcodes_cb: 缺失条码的回调函数，接收条码列表
         """
         self.config = config
+        self.missing_barcodes_cb = missing_barcodes_cb
+        self.current_missing_barcodes = []  # 记录当前文件处理中缺失的条码
+        self.current_file_path = ""        # 记录当前处理的文件路径
         
         # 修复ConfigParser对象没有get_path方法的问题
         try:
@@ -62,7 +66,7 @@ class ExcelProcessor:
                 logger.warning(f"模板文件不存在: {self.template_path}")
             
             # 设置缓存文件路径
-            self.cache_file = os.path.join(self.output_dir, "processed_files.json")
+            self.cache_file = os.path.join(self.output_dir, "excel_process_records.json")
             self.processed_files = self._load_processed_files()
             
             # 确保目录存在
@@ -80,7 +84,7 @@ class ExcelProcessor:
             if product_db is not None:
                 self.product_db = product_db
             else:
-                from ..db.product_db import ProductDatabase
+                from app.core.db.product_db import ProductDatabase
                 db_path = config.get_path('Paths', 'product_db', fallback='data/product_cache.db') if hasattr(config, 'get_path') else 'data/product_cache.db'
                 tpl_folder = config.get('Paths', 'template_folder', fallback='templates')
                 item_data = config.get('Templates', 'item_data', fallback='商品资料.xlsx')
@@ -220,6 +224,21 @@ class ExcelProcessor:
                 # 跳过空条码行
                 if not product['barcode']:
                     continue
+
+                # 检查条码是否存在于数据库（商品资料）
+                bc = product['barcode']
+                mem = self.product_db.get_memory(bc)
+                if not mem or mem.get('confidence', 0) < 50:
+                    # 如果不存在，或者置信度低（说明不是来自商品资料模板），记录为缺失
+                    if bc not in self.current_missing_barcodes:
+                        self.current_missing_barcodes.append(bc)
+                        # 记录到数据库
+                        self.product_db.record_missing_barcode(
+                            bc, 
+                            product.get('name', ''), 
+                            self.current_file_path
+                        )
+                        logger.warning(f"条码缺失: {bc} ({product.get('name', '')})")
                 
                 # 检查备注列，过滤换货、退货、作废等非采购行
                 skip_row = False
@@ -606,6 +625,9 @@ class ExcelProcessor:
         if not os.path.exists(file_path):
             logger.error(f"文件不存在: {file_path}")
             return None
+
+        self.current_missing_barcodes = []  # 重置缺失列表
+        self.current_file_path = file_path # 设置当前处理文件路径
         
         try:
             # 读取Excel文件时不立即指定表头
@@ -672,6 +694,14 @@ class ExcelProcessor:
                 
                 # 不再自动打开输出目录
                 logger.info(f"采购单已保存到: {output_file}")
+
+                # 处理完成，如果有缺失条码，触发回调
+                if self.current_missing_barcodes and self.missing_barcodes_cb:
+                    try:
+                        self.missing_barcodes_cb(self.current_missing_barcodes)
+                    except Exception as e:
+                        logger.error(f"触发缺失条码回调失败: {e}")
+
                 if progress_cb:
                     try:
                         progress_cb(100)

app/core/excel/validators.py

@@ -8,8 +8,8 @@ import re
 import logging
 from typing import Dict, Any, Optional, List, Tuple, Union
 
-from ..utils.log_utils import get_logger
-from ..utils.string_utils import parse_monetary_string
+from app.core.utils.log_utils import get_logger
+from app.core.utils.string_utils import parse_monetary_string
 
 logger = get_logger(__name__)
 

app/core/handlers/__init__.py

@@ -2,8 +2,8 @@
 数据处理handlers模块初始化文件
 """
 
-from .data_cleaner import DataCleaner
-from .column_mapper import ColumnMapper
-from .calculator import DataCalculator
+from app.core.handlers.data_cleaner import DataCleaner
+from app.core.handlers.column_mapper import ColumnMapper
+from app.core.handlers.calculator import DataCalculator
 
 __all__ = ['DataCleaner', 'ColumnMapper', 'DataCalculator']

app/core/handlers/calculator.py

@@ -7,7 +7,7 @@
 import pandas as pd
 import numpy as np
 from typing import Dict, Any, Optional, List, Union
-from ...core.utils.log_utils import get_logger
+from app.core.utils.log_utils import get_logger
 
 logger = get_logger(__name__)
 

app/core/handlers/column_mapper.py

@@ -7,7 +7,7 @@
 import re
 import pandas as pd
 from typing import Dict, Any, Optional, List, Union
-from ...core.utils.log_utils import get_logger
+from app.core.utils.log_utils import get_logger
 
 logger = get_logger(__name__)
 
@@ -22,7 +22,7 @@ class ColumnMapper:
     STANDARD_COLUMNS = {
         'barcode': [
             '条码', '条形码', '商品条码', '商品条形码', '产品条码', '商品编码',
-            '商品编号', '条码（必填）', '电脑条码', '条码ID',
+            '商品编号', '条码（必填）', '电脑条码', '条码ID', '单品条码',
             'barcode', 'Barcode', 'BarCode', 'code', '编码',
         ],
         'name': [
@@ -363,7 +363,7 @@ class ColumnMapper:
         """
         header_keywords = [
             '条码', '条形码', '商品条码', '商品名称', '名称', '规格',
-            '单价', '数量', '金额', '单位', '必填', '编码',
+            '单价', '数量', '金额', '单位', '必填', '编码', '单品条码', '序号',
         ]
 
         best_row = -1

app/core/handlers/data_cleaner.py

@@ -6,7 +6,7 @@
 
 import pandas as pd
 from typing import Dict, Any, Optional, List, Union
-from ...core.utils.log_utils import get_logger
+from app.core.utils.log_utils import get_logger
 
 logger = get_logger(__name__)
 

app/core/ocr/baidu_ocr.py

@@ -9,7 +9,7 @@ import base64
 import requests
 from typing import Dict, Optional, Union, List
 
-from ..utils.log_utils import get_logger
+from app.core.utils.log_utils import get_logger
 
 logger = get_logger(__name__)
 
@@ -229,10 +229,19 @@ class BaiduOCRClient:
                     logger.debug(f"百度OCR API返回结果: {result}")
                     
                     if 'error_code' in result:
+                        error_code = result.get('error_code')
                         error_msg = result.get('error_msg', '未知错误')
+                        
+                        # 如果是 QPS 限制 (18)，增加延迟并重试
+                        if error_code == 18:
+                            wait_time = self.retry_delay * (attempt + 1)
+                            logger.warning(f"触发 QPS 限制，将在 {wait_time} 秒后重试 (尝试 {attempt+1}/{self.max_retries})")
+                            time.sleep(wait_time)
+                            continue
+                            
                         logger.error(f"百度OCR API错误: {error_msg}")
                         # 如果是授权错误，尝试刷新令牌
-                        if result.get('error_code') in [110, 111]:  # 授权相关错误码
+                        if error_code in [110, 111]:  # 授权相关错误码
                             logger.info("尝试刷新访问令牌...")
                             self.token_manager.refresh_token()
                         return None
@@ -292,8 +301,18 @@ class BaiduOCRClient:
                 if response.status_code == 200:
                     result = response.json()
                     if 'error_code' in result:
-                        logger.warning(f"通用识别错误: {result.get('error_msg')}")
-                        if result.get('error_code') in (110, 111):
+                        error_code = result.get('error_code')
+                        error_msg = result.get('error_msg', '未知错误')
+                        
+                        # 如果是 QPS 限制 (18)，增加延迟并重试
+                        if error_code == 18:
+                            wait_time = self.retry_delay * (attempt + 1)
+                            logger.warning(f"触发 QPS 限制 (General)，将在 {wait_time} 秒后重试 (尝试 {attempt+1}/{self.max_retries})")
+                            time.sleep(wait_time)
+                            continue
+
+                        logger.warning(f"通用识别错误: {error_msg}")
+                        if error_code in (110, 111):
                             self.token_manager.refresh_token()
                         return None
                     words_list = result.get('words_result') or []
@@ -306,6 +325,110 @@ class BaiduOCRClient:
                 time.sleep(self.retry_delay * (2 ** attempt))
         logger.error("通用识别失败")
         return None
+
+    def recognize_ticket(self, image_data: Union[str, bytes]) -> Optional[List[Dict]]:
+        """通用票务识别：用于精准捕获表头抬头、日期等关键信息。
+        使用用户指定的 URL: https://aip.baidubce.com/rest/2.0/ocr/v1/general_ocr
+
+        Returns:
+            [{'words': '...'}, ...]
+        """
+        access_token = self.token_manager.get_token()
+        if not access_token:
+            logger.error("无法获取访问令牌，无法进行票务识别")
+            return None
+
+        if isinstance(image_data, str):
+            image_data = self.read_image(image_data)
+            if image_data is None:
+                return None
+
+        # 使用用户指定的通用票务识别接口
+        url = self.config.get('API', 'ticket_ocr_url',
+                              fallback='https://aip.baidubce.com/rest/2.0/ocr/v1/general_ocr')
+        url = f"{url}?access_token={access_token}"
+        image_base64 = base64.b64encode(image_data).decode('utf-8')
+
+        payload = {
+            'image': image_base64,
+            'detect_direction': 'true',  # 开启方向检测，应对旋转的图片
+        }
+        headers = {'Content-Type': 'application/x-www-form-urlencoded',
+                   'Accept': 'application/json'}
+
+        for attempt in range(self.max_retries):
+            try:
+                response = requests.post(url, data=payload, headers=headers,
+                                         timeout=self.timeout)
+                if response.status_code == 200:
+                    result = response.json()
+                    if 'error_code' in result:
+                        error_code = result.get('error_code')
+                        error_msg = result.get('error_msg', '未知错误')
+
+                        # 如果是 QPS 限制 (18)，增加延迟并重试
+                        if error_code == 18:
+                            wait_time = self.retry_delay * (attempt + 1)
+                            logger.warning(f"触发 QPS 限制 (Ticket)，将在 {wait_time} 秒后重试 (尝试 {attempt+1}/{self.max_retries})")
+                            time.sleep(wait_time)
+                            continue
+
+                        # 如果是权限错误 (6)，记录详细日志告知用户
+                        if error_code == 6:
+                            logger.error("权限错误 (6): 您的百度 API Key 未开启【通用卡证票据识别】(General OCR) 服务。请前往百度云控制台手动开启该服务，否则无法精准识别表头。")
+                        
+                        logger.warning(f"票务识别错误: {error_msg}")
+                        if error_code in (110, 111):
+                            self.token_manager.refresh_token()
+                        return None
+                    
+                    # 1. 优先处理 results 结构 (通用卡证票据识别的新结构)
+                    if 'results' in result and isinstance(result['results'], dict):
+                        extracted = []
+                        # 通常只有一个结果 "0"
+                        for res_id, res_content in result['results'].items():
+                            if not isinstance(res_content, dict): continue
+                            for k, v_list in res_content.items():
+                                if isinstance(v_list, list):
+                                    for item in v_list:
+                                        if isinstance(item, dict) and 'words' in item:
+                                            words_val = item['words']
+                                            if isinstance(words_val, list):
+                                                for w in words_val:
+                                                    extracted.append({'words': f"{k}: {w}"})
+                                            else:
+                                                extracted.append({'words': f"{k}: {words_val}"})
+                        return extracted
+
+                    # 2. 通用票据/票务识别通常也返回 words_result
+                    words_list = result.get('words_result')
+                    
+                    # 如果返回的是结构化字典，提取其中的文字行
+                    if isinstance(words_list, dict):
+                        extracted = []
+                        for k, v in words_list.items():
+                            if isinstance(v, dict) and 'words' in v:
+                                words_val = v['words']
+                                if isinstance(words_val, list):
+                                    for w in words_val:
+                                        extracted.append({'words': f"{k}: {w}"})
+                                else:
+                                    extracted.append({'words': f"{k}: {words_val}"})
+                            elif isinstance(v, str):
+                                extracted.append({'words': f"{k}: {v}"})
+                        return extracted
+                    
+                    return words_list or []
+                
+                logger.warning(f"票务识别请求失败 (尝试 {attempt+1}): {response.text[:200]}")
+            except Exception as e:
+                logger.warning(f"票务识别异常 (尝试 {attempt+1}): {e}")
+            
+            if attempt < self.max_retries - 1:
+                time.sleep(self.retry_delay * (2 ** attempt))
+        
+        logger.error("票务识别失败")
+        return None
     
     def get_excel_result(self, request_id_or_result: Union[str, Dict]) -> Optional[bytes]:
         """

app/core/ocr/metadata_extractor.py

@@ -16,6 +16,14 @@ from typing import List, Optional, Tuple
 SUPPLIER_KEYWORDS = (
     "供货单", "供应商", "供货方", "批发", "酒行", "商行",
     "经销", "专卖店", "配送单", "送货单", "采购单", "订单",
+    "销售单", "出库单", "经营部", "商贸", "有限公司", "发证机构",
+)
+
+# 供应商名称中需要剔除的冗余噪声词
+SUPPLIER_NOISE_WORDS = (
+    "标题", "采购单", "销售单", "入库单", "出库单", "送货单", 
+    "单据", "订单", "配送单", "清单", "供货单", "预览", "详情",
+    "打印", "副本", "记账", "联", "存根",
 )
 
 # 总金额关键词（命中后取该行最近一个金额数字）
@@ -24,6 +32,11 @@ AMOUNT_KEYWORDS = (
     "合计", "总计", "总计金额",
 )
 
+# 日期关键词（用于辅助定位日期）
+DATE_KEYWORDS = (
+    "单据日期", "下单时间", "日期", "时间", "开单日期", "制单日期", "打印时间", "业务日期", "下单日期",
+)
+
 # 日期正则：4 种格式
 DATE_PATTERNS = [
     # 2026年07月17日 / 2026年7月17日
@@ -52,6 +65,16 @@ class OrderMetadata:
         return asdict(self)
 
 
+# 排除关键词：包含这些词的行绝对不是供应商抬头
+EXCLUDE_KEYWORDS = (
+    "购货单位", "客户名称", "收货地址", "联系电话", "经手人", 
+    "地址", "电话", "传真", "邮编", "网址", "开户行", "账号", 
+    "税号", "业务员", "联系人", "单据编号", "流水号", "页码",
+    "四川省", "成都市", "武侯区", "双流区", "高新区", "金牛区",
+    "成华区", "锦江区", "龙泉驿区", "青羊区", "新都区", "温江区",
+    "街道", "社区", "楼", "号", "室", "路", "巷", "段",
+)
+
 class OrderMetadataExtractor:
     """单据元信息识别器。"""
 
@@ -61,20 +84,35 @@ class OrderMetadataExtractor:
     # 供应商名最大长度
     MAX_SUPPLIER_LEN = 50
 
-    def extract(self, ocr_text: str, ocr_rows: Optional[List[List[str]]] = None) -> OrderMetadata:
+    def extract(self, ocr_text: str, ocr_rows: Optional[List[List[str]]] = None, general_text: Optional[str] = None) -> OrderMetadata:
         """从 OCR 原始文本和/或解析后的二维数组提取三字段。
 
         Args:
-            ocr_text: OCR 原始字符串全文（带换行）
-            ocr_rows: 解析后的二维数组（可选），用于 row-level 精确匹配
+            ocr_text: OCR 原始字符串全文（从表格 OCR 提取）
+            ocr_rows: 解析后的二维数组（可选）
+            general_text: 从通用票据识别接口 (/v1/general_ocr) 提取的文本，优先级最高
 
         Returns:
             OrderMetadata
         """
-        text = ocr_text or ''
-        supplier, raw_supplier = self._extract_supplier(text, ocr_rows)
-        bill_date = self._extract_bill_date(text)
-        total_amount = self._extract_total_amount(text, ocr_rows)
+        # 优先使用通用票据识别的文本进行供应商和日期提取
+        supplier_source = general_text if general_text else ocr_text
+        date_source = general_text if general_text else ocr_text
+        
+        # 金额通常在表格内，优先使用 ocr_text
+        amount_source = ocr_text or general_text or ''
+
+        supplier, raw_supplier = self._extract_supplier(supplier_source, ocr_rows)
+        bill_date = self._extract_bill_date(date_source)
+        total_amount = self._extract_total_amount(amount_source, ocr_rows)
+
+        # 如果通用票据识别没抓到供应商，尝试用表格 OCR 的文本补位
+        if not supplier and general_text and ocr_text:
+            supplier, raw_supplier = self._extract_supplier(ocr_text, ocr_rows)
+            
+        # 如果通用票据识别没抓到日期，尝试用表格 OCR 的文本补位
+        if not bill_date and general_text and ocr_text:
+            bill_date = self._extract_bill_date(ocr_text)
 
         return OrderMetadata(
             supplier=supplier,
@@ -95,9 +133,15 @@ class OrderMetadataExtractor:
             cleaned = self._clean_supplier_line(line)
             if not cleaned:
                 continue
+            
+            # 严格排除：购货单位、地址、电话等干扰项
+            if any(k in cleaned for k in EXCLUDE_KEYWORDS):
+                continue
+
             for kw in SUPPLIER_KEYWORDS:
                 if kw in cleaned:
-                    return self._truncate(cleaned), cleaned
+                    final_name = self._final_cleanup_supplier(cleaned)
+                    return self._truncate(final_name), cleaned
 
         # 2) 兜底：取顶部第一个"含中文且无数字行号"且长度 ≥ 4 的非空行
         # 但排除"纯日期行"（避免把日期当供应商）
@@ -105,6 +149,11 @@ class OrderMetadataExtractor:
             cleaned = self._clean_supplier_line(line)
             if not cleaned:
                 continue
+            
+            # 兜底也要严格排除地址和电话行
+            if any(k in cleaned for k in EXCLUDE_KEYWORDS):
+                continue
+
             if not (re.search(r'[\u4e00-\u9fa5]', cleaned) and len(cleaned) >= 4):
                 continue
             # 排除：纯日期（YYYY-MM-DD / YYYY/MM/DD / YYYY年MM月DD日 / YYYYMMDD）
@@ -117,16 +166,63 @@ class OrderMetadataExtractor:
             # 排除：以"单据"开头的行
             if re.match(r'^\s*单据[:：]?', cleaned):
                 continue
-            return self._truncate(cleaned), cleaned
+            
+            final_name = self._final_cleanup_supplier(cleaned)
+            if final_name:
+                return self._truncate(final_name), cleaned
 
         return '', ''
 
+    @staticmethod
+    def _final_cleanup_supplier(s: str) -> str:
+        """最后的供应商名称深度清理：去除“标题”、“采购单”等噪声。"""
+        if not s:
+            return ''
+        
+        # 1. 统一处理全角和常见符号
+        s = s.replace('：', ':').replace('（', '(').replace('）', ')')
+        
+        # 2. 去除“标题:”或“名称:”这类前缀
+        s = re.sub(r'^(标题|名称|供应商|供货方|单位|商户)[:\s]*', '', s)
+        
+        # 3. 循环去除噪声词
+        noise_sorted = sorted(SUPPLIER_NOISE_WORDS, key=len, reverse=True)
+        
+        changed = True
+        while changed:
+            original = s
+            for noise in noise_sorted:
+                # 简单替换所有匹配的噪声词
+                s = s.replace(noise, '')
+            
+            # 去除括号及其中间的噪声词（如 (采购单) ）
+            s = re.sub(r'\(\s*\)', '', s)
+            
+            # 去除首尾残留标点和空白
+            s = re.sub(r'^[:：\s\-_\|\.\(\)]+', '', s)
+            s = re.sub(r'[:：\s\-_\|\.\(\)]+$', '', s)
+            
+            changed = (s != original)
+            
+        return s.strip()
+
     @staticmethod
     def _clean_supplier_line(line: str) -> str:
         """清理一行文本：去前后空白、去首尾日期/编号/电话/标点。"""
         s = line.strip()
         if not s:
             return ''
+        
+        # 统一全角冒号
+        s = s.replace('：', ':')
+        
+        # 如果是“标题: 新双利采购单”，保留冒号后面的部分进行初步处理
+        if ':' in s:
+            parts = s.split(':', 1)
+            # 如果冒号前面是“标题”、“名称”等词，则取后面
+            if any(k in parts[0] for k in ("标题", "名称", "供应商", "商户")):
+                s = parts[1].strip()
+
         # 去行首日期/编号前缀
         s = re.sub(r'^[\s\d\-\.\/年月日:：]+', '', s)
         # 去行尾标点
@@ -142,13 +238,54 @@ class OrderMetadataExtractor:
 
     def _extract_bill_date(self, text: str) -> str:
         """提取单据日期，标准化为 YYYYMMDD。"""
+        if not text:
+            return ''
+            
+        lines = text.splitlines()
+        
+        # 1. 优先尝试关键词定位逻辑（如用户建议：搜索“时间”或“日期”）
+        for i, line in enumerate(lines):
+            line_clean = line.strip().replace('：', ':')
+            
+            # 检查是否包含日期相关关键词
+            for kw in DATE_KEYWORDS:
+                if kw in line_clean:
+                    # a) 尝试在当前行找日期
+                    for pat in DATE_PATTERNS:
+                        m = pat.search(line_clean)
+                        if m:
+                            y, mo, d = m.group(1), m.group(2), m.group(3)
+                            if self._is_valid_date(y, mo, d):
+                                return f"{y}{int(mo):02d}{int(d):02d}"
+                    
+                    # b) 如果当前行没找到，尝试在下一行找（针对表格 OCR 错位情况）
+                    if i + 1 < len(lines):
+                        next_line = lines[i+1].strip()
+                        for pat in DATE_PATTERNS:
+                            m = pat.search(next_line)
+                            if m:
+                                y, mo, d = m.group(1), m.group(2), m.group(3)
+                                if self._is_valid_date(y, mo, d):
+                                    return f"{y}{int(mo):02d}{int(d):02d}"
+        
+        # 2. 兜底策略：全文正则匹配
         for pat in DATE_PATTERNS:
-            m = pat.search(text)
-            if m:
+            # 优先找包含 202 开头的年份（更像当前日期）
+            matches = pat.finditer(text)
+            for m in matches:
                 y, mo, d = m.group(1), m.group(2), m.group(3)
                 if self._is_valid_date(y, mo, d):
-                    return f"{y}{int(mo):02d}{int(d):02d}"
-        return ''
+                    # 如果年份以 202 开头，优先返回
+                    if y.startswith('202'):
+                        return f"{y}{int(mo):02d}{int(d):02d}"
+                    # 否则记录下来作为候选
+                    last_valid = f"{y}{int(mo):02d}{int(d):02d}"
+            
+        # 如果没有 202 开头的，返回最后一个有效的
+        try:
+            return last_valid
+        except NameError:
+            return ''
 
     @staticmethod
     def _is_valid_date(y: str, m: str, d: str) -> bool:

app/core/ocr/table_ocr.py

@@ -10,8 +10,8 @@ import base64
 from concurrent.futures import ThreadPoolExecutor
 from typing import Dict, List, Optional, Tuple, Callable
 
-from ..utils.log_utils import get_logger
-from ..utils.file_utils import (
+from app.core.utils.log_utils import get_logger
+from app.core.utils.file_utils import (
     ensure_dir, 
     get_file_extension, 
     get_files_by_extensions, 
@@ -20,6 +20,7 @@ from ..utils.file_utils import (
     load_json,
     save_json
 )
+from app.config.settings import ConfigManager
 from .baidu_ocr import BaiduOCRClient
 
 logger = get_logger(__name__)
@@ -102,14 +103,16 @@ class OCRProcessor:
     OCR处理器，负责协调OCR识别和结果处理
     """
     
-    def __init__(self, config):
+    def __init__(self, config: Optional[ConfigManager] = None):
         """
         初始化OCR处理器
         
         Args:
-            config: 配置信息
+            config: 配置管理器
         """
-        self.config = config
+        self.config = config or ConfigManager()
+        self.ocr_client = None
+        self._ensure_ocr_client()
         
         # 修复ConfigParser对象没有get_path方法的问题
         try:
@@ -348,9 +351,10 @@ class OCRProcessor:
 
         if max_workers is None:
             try:
-                max_workers = self.config.getint('Performance', 'max_workers', fallback=4)
+                # 强制设为 1 以严格遵守百度 QPS=2 的限制（一张图调两个接口就满了）
+                max_workers = self.config.getint('Performance', 'max_workers', fallback=1)
             except Exception:
-                max_workers = 4
+                max_workers = 1
     
         # 获取未处理的图片
         unprocessed_images = self.get_unprocessed_images()

app/core/processors/__init__.py

@@ -2,8 +2,8 @@
 处理器模块初始化文件
 """
 
-from .base import BaseProcessor
-from .ocr_processor import OCRProcessor
-from .tobacco_processor import TobaccoProcessor
+from app.core.processors.base import BaseProcessor
+from app.core.processors.ocr_processor import OCRProcessor
+from app.core.processors.tobacco_processor import TobaccoProcessor
 
 __all__ = ['BaseProcessor', 'OCRProcessor', 'TobaccoProcessor']

app/core/processors/base.py

@@ -10,7 +10,7 @@ from pathlib import Path
 import logging
 import pandas as pd
 
-from ...core.utils.log_utils import get_logger
+from app.core.utils.log_utils import get_logger
 
 logger = get_logger(__name__)
 

app/core/processors/ocr_processor.py

@@ -8,10 +8,10 @@ import os
 from pathlib import Path
 from typing import Optional, Dict, Any, List
 
-from .base import BaseProcessor
-from ...services.ocr_service import OCRService
-from ...services.order_service import OrderService
-from ...core.utils.log_utils import get_logger
+from app.core.processors.base import BaseProcessor
+from app.services.ocr_service import OCRService
+from app.services.order_service import OrderService
+from app.core.utils.log_utils import get_logger
 
 logger = get_logger(__name__)
 

app/core/processors/supplier_processors/__init__.py

@@ -2,6 +2,6 @@
 供应商处理器模块初始化文件
 """
 
-from .generic_supplier_processor import GenericSupplierProcessor
+from app.core.processors.supplier_processors.generic_supplier_processor import GenericSupplierProcessor
 
 __all__ = ['GenericSupplierProcessor']

app/core/processors/supplier_processors/generic_supplier_processor.py

@@ -9,12 +9,12 @@ import pandas as pd
 from typing import Optional, Dict, Any, List
 from pathlib import Path
 
-from ..base import BaseProcessor
-from ...utils.log_utils import get_logger
-from ...handlers.rule_engine import apply_rules
-from ...handlers.column_mapper import ColumnMapper
-from ...handlers.data_cleaner import DataCleaner
-from ...handlers.calculator import DataCalculator
+from app.core.processors.base import BaseProcessor
+from app.core.utils.log_utils import get_logger
+from app.core.handlers.rule_engine import apply_rules
+from app.core.handlers.column_mapper import ColumnMapper
+from app.core.handlers.data_cleaner import DataCleaner
+from app.core.handlers.calculator import DataCalculator
 
 logger = get_logger(__name__)
 

app/core/processors/tobacco_processor.py

@@ -14,10 +14,10 @@ from openpyxl import load_workbook
 from typing import Optional, Dict, Any, List, Tuple
 from pathlib import Path
 
-from .base import BaseProcessor
-from ...core.utils.log_utils import get_logger
-from ...core.utils.string_utils import parse_monetary_string
-from ...core.utils.dialog_utils import show_custom_dialog
+from app.core.processors.base import BaseProcessor
+from app.core.utils.log_utils import get_logger
+from app.core.utils.string_utils import parse_monetary_string
+from app.core.utils.dialog_utils import show_custom_dialog
 
 logger = get_logger(__name__)
 

app/core/utils/cloud_sync.py

@@ -6,7 +6,7 @@ from typing import Optional, Tuple
 
 import requests
 
-from .log_utils import get_logger
+from app.core.utils.log_utils import get_logger
 
 logger = get_logger(__name__)
 

app/core/utils/dialog_utils.py

@@ -13,7 +13,7 @@ import tkinter as tk
 from tkinter import messagebox, ttk, simpledialog
 from datetime import datetime
 
-from .cloud_sync import GiteaSync
+from app.core.utils.cloud_sync import GiteaSync
 from app.config.settings import ConfigManager
 
 def create_custom_dialog(title="提示", message="", result_file=None, time_info=None, 
@@ -82,7 +82,7 @@ def create_custom_dialog(title="提示", message="", result_file=None, time_info
             file_size = os.path.getsize(result_file)
             file_time = datetime.fromtimestamp(os.path.getmtime(result_file))
             
-            from .file_utils import format_file_size
+            from app.core.utils.file_utils import format_file_size
             size_text = format_file_size(file_size)
             
             tk.Label(file_frame, text=f"文件大小: {size_text}", font=("Arial", 10)).pack(anchor=tk.W, padx=10, pady=2)
@@ -831,10 +831,10 @@ SYNC_FILES = [
         "type": "binary",
     },
     {
-        "name": "商品记忆库",
-        "remote": "product_memory.json",
-        "local": "data/product_memory.json",
-        "type": "json",
+        "name": "商品记忆库 (DB)",
+        "remote": "product_cache.db",
+        "local": "data/product_cache.db",
+        "type": "binary",
     },
 ]
 

app/core/utils/file_utils.py

@@ -12,7 +12,7 @@ from datetime import datetime
 from pathlib import Path
 from typing import Dict, List, Optional, Union, Any
 
-from .log_utils import get_logger
+from app.core.utils.log_utils import get_logger
 
 logger = get_logger(__name__)
 

app/services/batch_service.py

@@ -11,8 +11,8 @@ import os
 from pathlib import Path
 from typing import Any, Callable, Dict, List, Optional
 
-from ..config.settings import ConfigManager
-from ..core.utils.log_utils import get_logger
+from app.config.settings import ConfigManager
+from app.core.utils.log_utils import get_logger
 from .ocr_service import OCRService
 from .order_service import OrderService
 
@@ -25,10 +25,11 @@ IMAGE_EXTS = {'.png', '.jpg', '.jpeg', '.bmp', '.webp', '.tif', '.tiff'}
 class BatchService:
     """批量处理服务。"""
 
-    def __init__(self, config: Optional[ConfigManager] = None):
+    def __init__(self, config: Optional[ConfigManager] = None, missing_barcodes_cb: Optional[Callable[[List[str]], None]] = None):
         self.config = config or ConfigManager()
+        self.missing_barcodes_cb = missing_barcodes_cb
         self.ocr_service = OCRService(self.config)
-        self.order_service = OrderService(self.config)
+        self.order_service = OrderService(self.config, missing_barcodes_cb=self.missing_barcodes_cb)
         self._input_folder = self.config.get_path(
             'Paths', 'input_folder', fallback='data/input', create=True
         )

app/services/ocr_service.py

@@ -7,9 +7,9 @@ OCR服务模块
 from typing import Dict, List, Optional, Tuple, Union, Any, Callable
 import os
 
-from ..config.settings import ConfigManager
-from ..core.utils.log_utils import get_logger
-from ..core.ocr.table_ocr import OCRProcessor
+from app.config.settings import ConfigManager
+from app.core.utils.log_utils import get_logger
+from app.core.ocr.table_ocr import OCRProcessor
 
 logger = get_logger(__name__)
 
@@ -62,16 +62,10 @@ class OCRService:
             if not self._is_valid_image(image_path):
                 logger.error(f"不支持的文件类型: {image_path}")
                 return None
-                
-            # 检查是否已处理
-            excel_file = self._get_excel_path(image_path)
-            if os.path.exists(excel_file):
-                logger.info(f"文件已处理过，跳过OCR识别: {image_path}")
-                # 即使 xlsx 已存在，仍补写 meta.json（首次 OCR 后可能未生成）
-                meta_path = str(excel_file).replace('.xlsx', '.meta.json')
-                if not os.path.exists(meta_path):
-                    self._write_meta_json(str(excel_file), {}, image_path)
-                return excel_file
+
+            # 不再做 xlsx 已存在就跳过的判断（xlsx 可能被业务层重命名过），
+            # 跳过逻辑交给 OCRProcessor 内部的 record_manager.is_processed(image_path)
+            # —— 由它读 processed_files.json（业务层会同步更新 xlsx 路径）
                 
             # 执行OCR识别
             result = self.ocr_processor.process_image(image_path)
@@ -212,24 +206,36 @@ class OCRService:
             base = Path(excel_path)
             meta_path = base.with_suffix('.meta.json')
 
-            # 0) 优先：调百度通用文字识别（/accurate），覆盖全图文字（含手写抬头/日期）
+            # 0) 优先：调百度通用票务识别（/general_ocr），精准捕获表头（含手写抬头/日期）
             general_text = ''
             general_lines = []
             try:
-                # OCRService.ocr_processor = core.ocr.table_ocr.OCRProcessor
-                # 其 .ocr_client = BaiduOCRClient (process_image 时初始化)
                 client = getattr(self.ocr_processor, 'ocr_client', None)
                 if client is None:
-                    # 显式触发一次 process_image 准备流程 (不会重复 OCR)
                     try:
                         self.ocr_processor._ensure_ocr_client()
                     except Exception:
                         pass
                     client = getattr(self.ocr_processor, 'ocr_client', None)
-                if client and hasattr(client, 'recognize_general') and image_path and os.path.exists(str(image_path)):
+                
+                # 关键：在进行第二次 OCR (票务识别) 前稍微等待，避免触发 QPS 限制
+                import time
+                time.sleep(0.5)
+
+                # 优先使用票务识别 (Ticket OCR)
+                if client and hasattr(client, 'recognize_ticket') and image_path and os.path.exists(str(image_path)):
+                    logger.info("使用通用票务识别捕获表头元数据...")
+                    words = client.recognize_ticket(str(image_path))
+                    if words:
+                        general_lines = [(w.get('words') or '').strip() for w in words if (w.get('words') or '').strip()]
+                        general_text = '\n'.join(general_lines)
+                        logger.info(f"票务识别获取 {len(general_lines)} 行文字")
+                
+                # 如果票务识别无结果或不可用，则兜底使用高精度文字识别
+                if not general_lines and client and hasattr(client, 'recognize_general') and image_path and os.path.exists(str(image_path)):
+                    logger.info("尝试高精度文字识别作为兜底...")
                     words = client.recognize_general(str(image_path))
                     if words:
-                        # 按 location.top 排序（顶部先），方便后续提取供应商/日期
                         def _top(w):
                             loc = w.get('location') or {}
                             try:
@@ -239,9 +245,9 @@ class OCRService:
                         words_sorted = sorted(words, key=_top)
                         general_lines = [(w.get('words') or '').strip() for w in words_sorted if (w.get('words') or '').strip()]
                         general_text = '\n'.join(general_lines)
-                        logger.info(f"通用识别获取 {len(general_lines)} 行文字")
+                        logger.info(f"通用高精度识别获取 {len(general_lines)} 行文字")
             except Exception as e:
-                logger.warning(f"通用识别失败（不影响主流程）: {e}")
+                logger.warning(f"获取元数据识别失败（不影响主流程）: {e}")
 
             # 1) 表格识别 header/body 拼表内文字（已在前面逻辑处理）
             ocr_text = ''

app/services/order_service.py

@@ -12,13 +12,13 @@ from datetime import datetime
 from pathlib import Path
 from typing import Dict, List, Optional, Tuple, Union, Any, Callable
 
-from ..config.settings import ConfigManager
-from ..core.utils.log_utils import get_logger
-from ..core.excel.processor import ExcelProcessor
-from ..core.excel.merger import PurchaseOrderMerger
-from ..core.db.product_db import ProductDatabase
-from ..core.db.order_metadata_db import OrderMetadataDB
-from ..core.ocr.metadata_extractor import OrderMetadataExtractor, sanitize_for_filename
+from app.config.settings import ConfigManager
+from app.core.utils.log_utils import get_logger
+from app.core.excel.processor import ExcelProcessor
+from app.core.excel.merger import PurchaseOrderMerger
+from app.core.db.product_db import ProductDatabase
+from app.core.db.order_metadata_db import OrderMetadataDB
+from app.core.ocr.metadata_extractor import OrderMetadataExtractor, sanitize_for_filename
 
 logger = get_logger(__name__)
 
@@ -27,15 +27,17 @@ class OrderService:
     订单服务：协调Excel处理和订单合并流程
     """
     
-    def __init__(self, config: Optional[ConfigManager] = None):
+    def __init__(self, config: Optional[ConfigManager] = None, missing_barcodes_cb: Optional[Callable[[List[str]], None]] = None):
         """
         初始化订单服务
         
         Args:
             config: 配置管理器，如果为None则创建新的
+            missing_barcodes_cb: 缺失条码的回调函数
         """
         logger.info("初始化OrderService")
         self.config = config or ConfigManager()
+        self.missing_barcodes_cb = missing_barcodes_cb
 
         # 创建共享的商品数据库实例
         db_path = self.config.get_path('Paths', 'product_db', fallback='data/product_cache.db') if hasattr(self.config, 'get_path') else 'data/product_cache.db'
@@ -45,7 +47,11 @@ class OrderService:
         self.product_db = ProductDatabase(db_path, tpl_path)
 
         # 创建Excel处理器和采购单合并器
-        self.excel_processor = ExcelProcessor(self.config, product_db=self.product_db)
+        self.excel_processor = ExcelProcessor(
+            self.config, 
+            product_db=self.product_db,
+            missing_barcodes_cb=self.missing_barcodes_cb
+        )
         self.order_merger = PurchaseOrderMerger(self.config)
 
         # 元信息识别器 + 单据元信息库
@@ -291,15 +297,16 @@ class OrderService:
             if meta_path.exists():
                 try:
                     payload = json.loads(meta_path.read_text(encoding='utf-8'))
-                    # 优先用通用识别文本（含手写抬头/日期）
-                    ocr_text = payload.get('general_text') or payload.get('ocr_text', '') or ''
+                    # 区分表格 OCR 文本和通用识别文本
+                    ocr_text = payload.get('ocr_text', '') or ''
+                    general_text = payload.get('general_text', '') or ''
                     ocr_rows = payload.get('ocr_rows', []) or []
                     source_image = payload.get('image_path', '') or ''
                 except Exception as e:
                     logger.warning(f"读 meta.json 失败: {e}")
 
             # 兜底：从 xlsx 拼文本（与 OCRService._write_meta_json 的兜底一致）
-            if not ocr_text:
+            if not ocr_text and not general_text:
                 try:
                     import xlrd
                     rb = xlrd.open_workbook(str(ocr_excel_path))
@@ -313,7 +320,7 @@ class OrderService:
                 except Exception as e:
                     logger.debug(f"从 xlsx 拼 OCR 文本失败: {e}")
 
-            meta = self.extractor.extract(ocr_text, ocr_rows)
+            meta = self.extractor.extract(ocr_text, ocr_rows, general_text=general_text)
             self.metadata_db.save(
                 file_hash=file_hash,
                 supplier=meta.supplier,
@@ -334,49 +341,77 @@ class OrderService:
     def _apply_metadata_to_filenames(self, result_path: str,
                                       ocr_excel_path: str,
                                       meta) -> str:
-        """应用新文件名规则：
-        - result: 采购单_{YYYYMMDD}_{供应商}_{hash}.xls
-        - 原图: {原stem}_{YYYYMMDD}_{供应商}_{hash}.{ext}
+        """应用新文件名规则（按供应商+日期）：
+        - result (xls)：{供应商}_{日期}.xls
+        - output xlsx：  {供应商}_{日期}.xlsx
+        - 原图：       {供应商}_{日期}.{ext}
+
+        冲突时加 _2 / _3 ...；任一步骤失败不影响整体。
 
-        任一步骤失败不影响 result 文件本身。
         Returns:
             新 result 路径（无论重命名是否成功都返回；失败时返回原路径）
         """
         try:
-            file_hash = Path(ocr_excel_path).stem
+            # 1. 提取并清理元数据
             supplier_clean = sanitize_for_filename(meta.supplier) or '未知供应商'
             date_part = meta.bill_date or '未知日期'
 
-            # ── 1. result 重命名 ──
-            new_result_name = f"采购单_{date_part}_{supplier_clean}_{file_hash}.xls"
-            result_dir = Path(result_path).parent
-            new_result_path = result_dir / new_result_name
-            try:
-                # 冲突时加 _N
-                if new_result_path.exists() and str(new_result_path) != str(result_path):
-                    new_result_path = self._dedup_path(new_result_path)
-                if str(new_result_path) != str(result_path):
-                    os.rename(result_path, str(new_result_path))
-                    logger.info(f"result 重命名: {result_path} -> {new_result_path}")
-                    result_path = str(new_result_path)
-            except Exception as e:
-                logger.warning(f"result 重命名失败: {e}")
+            # 2. 构造新的基础文件名：[供应商]_[日期]
+            # 按照用户要求：按照供应商名称加日期进行修改，且不含“采购单”等冗余字眼
+            base_name = f"{supplier_clean}_{date_part}"
 
-            # ── 2. 原图重命名 ──
+            file_hash = Path(ocr_excel_path).stem  # 仅用作 SQLite PK，不入文件名
+            logger.info(f"生成标准化文件名: {base_name}")
+
+            # ── 1. result 重命名 ──
+            new_result_path = self._safe_rename(
+                result_path, f"{base_name}.xls"
+            )
+
+            # ── 2. output xlsx 重命名（保证 output/ 与 result/ 文件名一致）──
+            new_output_path = self._safe_rename(
+                ocr_excel_path, f"{base_name}.xlsx"
+            )
+            # 同步 meta.json
+            old_meta = Path(ocr_excel_path).with_suffix('.meta.json')
+            if old_meta.exists():
+                new_meta = Path(new_output_path).with_suffix('.meta.json')
+                if str(new_meta) != str(old_meta):
+                    try:
+                        if new_meta.exists(): os.remove(new_meta)
+                        os.rename(old_meta, new_meta)
+                    except Exception as e:
+                        logger.warning(f"meta.json 重命名失败: {e}")
+
+            # ── 2.5 同步 processed_files.json（避免下次 OCR 重新识别）──
+            if str(new_output_path) != str(ocr_excel_path):
+                self._sync_processed_record(Path(ocr_excel_path), Path(new_output_path))
+                # 也要同步 ExcelProcessor 的记录
+                self._update_excel_process_record(Path(ocr_excel_path), Path(new_output_path))
+
+            # ── 3. 原图重命名 ──
             try:
-                meta_row = self.metadata_db.get(file_hash)
-                src_image = (meta_row or {}).get('source_image', '')
+                # 尝试从 processed_files.json 反查原图路径
+                src_image = self._find_input_image_for_output(str(ocr_excel_path))
+                
+                # 如果没找到，再从 DB 拿
+                if not src_image:
+                    meta_row = self.metadata_db.get(file_hash)
+                    src_image = (meta_row or {}).get('source_image', '')
+                
                 if src_image and os.path.exists(src_image):
                     src_p = Path(src_image)
-                    stem = src_p.stem
                     ext = src_p.suffix
-                    new_image_name = f"{stem}_{date_part}_{supplier_clean}_{file_hash}{ext}"
-                    new_image_path = src_p.parent / new_image_name
-                    # 不覆盖已重命名的图片
-                    if str(new_image_path) != str(src_p) and not new_image_path.exists():
-                        os.rename(src_p, new_image_path)
-                        logger.info(f"原图重命名: {src_p.name} -> {new_image_path.name}")
-                        # 更新 source_image 路径
+                    # 严禁包含原文件名，统一格式: 采购单_YYYYMMDD_供应商.ext
+                    new_image_name = f"{base_name}{ext}"
+                    new_image_path = self._safe_rename(src_image, new_image_name)
+                    
+                    # ── 3.5 同步 processed_files.json 的 Key (原图路径) ──
+                    if str(new_image_path) != str(src_p):
+                        self._sync_processed_key(src_p, Path(new_image_path))
+                        
+                    # 更新 source_image 路径到 DB
+                    if str(new_image_path) != str(src_p):
                         self.metadata_db.save(
                             file_hash=file_hash,
                             supplier=meta.supplier,
@@ -385,14 +420,47 @@ class OrderService:
                             raw_supplier_text=meta.raw_supplier_text,
                             source_image=str(new_image_path),
                         )
+                else:
+                    logger.warning(f"找不到原图，跳过重命名: {src_image}")
             except Exception as e:
                 logger.warning(f"原图重命名失败: {e}")
 
-            return result_path
+            return str(new_result_path)
         except Exception as e:
             logger.error(f"_apply_metadata_to_filenames 失败: {e}")
             return result_path
 
+    @staticmethod
+    def _safe_rename(src: str, new_basename: str) -> str:
+        """把 src 重命名为 src.parent / new_basename。
+
+        - 目标已存在时加 _2 / _3 ... 后缀避免覆盖
+        - 失败返回原路径，不抛异常
+        """
+        try:
+            src_p = Path(src)
+            if not src_p.exists():
+                return src
+            new_p = src_p.parent / new_basename
+            if str(new_p) == str(src_p):
+                return src
+            if new_p.exists():
+                # 冲突去重
+                stem, suffix = new_p.stem, new_p.suffix
+                n = 2
+                while True:
+                    cand = src_p.parent / f"{stem}_{n}{suffix}"
+                    if not cand.exists():
+                        new_p = cand
+                        break
+                    n += 1
+            os.rename(src_p, new_p)
+            logger.info(f"重命名: {src_p.name} -> {new_p.name}")
+            return str(new_p)
+        except Exception as e:
+            logger.warning(f"重命名失败 {src} -> {new_basename}: {e}")
+            return src
+
     @staticmethod
     def _dedup_path(p: Path) -> Path:
         """路径冲突时加 _N 后缀。"""
@@ -404,3 +472,113 @@ class OrderService:
             if not cand.exists():
                 return cand
             n += 1
+
+    def _find_input_image_for_output(self, output_xlsx: str) -> Optional[str]:
+        """从 processed_files.json 反查生成该 output_xlsx 的原图路径。"""
+        try:
+            output_dir = self.config.get_path('Paths', 'output_folder', fallback='data/output')
+            record_file = os.path.join(output_dir, 'processed_files.json')
+            
+            if not os.path.exists(record_file):
+                logger.debug(f"记录文件不存在: {record_file}")
+                return None
+                
+            with open(record_file, 'r', encoding='utf-8') as f:
+                data = json.load(f)
+            
+            target_output = os.path.normpath(output_xlsx)
+            found_img = None
+            
+            # 1. 尝试直接匹配和规范化路径匹配
+            for img_path, out_path in data.items():
+                if os.path.normpath(out_path) == target_output:
+                    found_img = img_path
+                    break
+            
+            if found_img:
+                if os.path.exists(found_img):
+                    return found_img
+                
+                # 2. 模糊匹配：如果记录中的路径不存在，尝试在同目录下找 stem 匹配的文件
+                # (解决因之前重命名导致的路径不一致问题)
+                img_p = Path(found_img)
+                parent = img_p.parent
+                if parent.exists():
+                    stem = img_p.stem
+                    logger.debug(f"尝试模糊匹配原图: stem={stem[:20]}... in {parent}")
+                    # 尝试寻找以原 stem 开头的文件
+                    for cand in parent.iterdir():
+                        if cand.is_file() and cand.stem.startswith(stem[:20]): 
+                            logger.info(f"模糊匹配成功: {cand.name}")
+                            return str(cand)
+                    logger.debug("模糊匹配失败")
+            else:
+                logger.debug(f"在 processed_files.json 中未找到输出文件 {output_xlsx} 对应的原图记录")
+        except Exception as e:
+            logger.debug(f"反查原图路径失败: {e}")
+        return None
+
+    def _sync_processed_record(self, old_output: Path, new_output: Path) -> None:
+        """重命名 output xlsx 后，把 processed_files.json 里所有 value 为旧路径的项改成新路径。"""
+        self._update_processed_json(old_path=old_output, new_path=new_output, is_key=False)
+
+    def _sync_processed_key(self, old_input: Path, new_input: Path) -> None:
+        """重命名原图后，把 processed_files.json 里的 key 从旧路径改为新路径。"""
+        self._update_processed_json(old_path=old_input, new_path=new_input, is_key=True)
+
+    def _update_excel_process_record(self, old_xlsx: Path, new_xlsx: Path) -> None:
+        """重命名 xlsx 后，同步 excel_process_records.json (ExcelProcessor 使用的记录)。"""
+        self._update_processed_json(
+            old_path=old_xlsx, 
+            new_path=new_xlsx, 
+            is_key=True, 
+            filename='excel_process_records.json'
+        )
+
+    def _update_processed_json(self, old_path: Path, new_path: Path, is_key: bool = False, filename: str = 'processed_files.json') -> None:
+        """更新处理记录 JSON。
+        
+        Args:
+            old_path: 旧路径
+            new_path: 新路径
+            is_key: True 更新 key, False 更新 value
+            filename: JSON 文件名
+        """
+        try:
+            # 确定文件路径
+            output_dir = self.config.get_path('Paths', 'output_folder', fallback='data/output')
+            record_file = os.path.join(output_dir, filename)
+            
+            if not os.path.exists(record_file):
+                return
+                
+            with open(record_file, 'r', encoding='utf-8') as f:
+                data = json.load(f)
+            
+            old_str = str(old_path)
+            new_str = str(new_path)
+            changed = False
+            
+            # 规范化路径以便匹配
+            old_norm = os.path.normpath(old_str)
+            
+            if is_key:
+                # 更新 Key
+                for k in list(data.keys()):
+                    if k == old_str or os.path.normpath(k) == old_norm:
+                        data[new_str] = data.pop(k)
+                        changed = True
+                        break
+            else:
+                # 更新 Value
+                for k, v in list(data.items()):
+                    if v == old_str or os.path.normpath(v) == old_norm:
+                        data[k] = new_str
+                        changed = True
+            
+            if changed:
+                with open(record_file, 'w', encoding='utf-8') as f:
+                    json.dump(data, f, ensure_ascii=False, indent=2)
+                logger.info(f"{filename} 已同步 ({'Key' if is_key else 'Value'}): {old_str} -> {new_str}")
+        except Exception as e:
+            logger.warning(f"更新 {filename} 失败: {e}")

app/services/processor_service.py

@@ -8,10 +8,10 @@ import logging
 from typing import Dict, Any, Optional, List
 from pathlib import Path
 
-from ..core.processors.base import BaseProcessor
-from ..core.processors.tobacco_processor import TobaccoProcessor
-from ..core.processors.ocr_processor import OCRProcessor
-from ..core.utils.log_utils import get_logger
+from app.core.processors.base import BaseProcessor
+from app.core.processors.tobacco_processor import TobaccoProcessor
+from app.core.processors.ocr_processor import OCRProcessor
+from app.core.utils.log_utils import get_logger
 
 logger = get_logger(__name__)
 
@@ -65,7 +65,7 @@ class ProcessorService:
 
             for supplier_config in supplier_configs:
                 try:
-                    from ..core.processors.supplier_processors.generic_supplier_processor import GenericSupplierProcessor
+                    from app.core.processors.supplier_processors.generic_supplier_processor import GenericSupplierProcessor
                     processor = GenericSupplierProcessor(self.config, supplier_config)
                     self.processors.append(processor)
                     logger.info(f"加载供应商处理器: {processor.name}")

app/services/special_suppliers_service.py

@@ -7,7 +7,7 @@ import time
 import pandas as pd
 from typing import Optional, Callable
 
-from ..core.utils.log_utils import get_logger
+from app.core.utils.log_utils import get_logger
 
 logger = get_logger(__name__)
 

app/services/tobacco_service.py

@@ -20,7 +20,7 @@ from typing import Optional, Dict, Any, List, Tuple
 from app.core.utils.log_utils import get_logger
 from app.core.utils.string_utils import parse_monetary_string
 from app.core.utils.dialog_utils import show_custom_dialog  # 导入自定义弹窗工具
-from ..config.settings import ConfigManager
+from app.config.settings import ConfigManager
 
 logger = get_logger(__name__)
 

app/ui/action_handlers.py

@@ -27,6 +27,27 @@ from .command_runner import get_running_task, set_running_task
 from .file_operations import select_file, select_excel_file, validate_unit_price_against_item_data
 
 
+def _get_missing_barcodes_callback(log_widget):
+    """创建并返回缺失条码的回调函数"""
+    def callback(missing_barcodes):
+        if not missing_barcodes:
+            return
+        
+        msg = f"发现以下条码不在商品资料中，已记录到数据库：\n\n" + "\n".join([f"• {bc}" for bc in missing_barcodes])
+        # 在主线程中弹出对话框
+        def show_msg():
+            messagebox.showwarning("发现缺失条码", msg)
+        
+        # 尝试使用 log_widget 的 winfo_toplevel() 来调用 after，确保在 UI 线程执行
+        try:
+            log_widget.after(0, show_msg)
+        except Exception:
+            # 降级处理
+            show_msg()
+            
+    return callback
+
+
 def _ask_and_merge_purchase_orders(order_service, log_widget, add_to_recent=False):
     """弹窗询问是否合并采购单，返回合并结果路径或 None。
 
@@ -138,108 +159,23 @@ def process_single_image_with_status(log_widget, status_bar):
 
 
 def run_pipeline_directly(log_widget, status_bar):
-    """直接运行完整处理流程"""
+    """运行完整处理流程：先选择图片，再执行 OCR+Excel 处理"""
     if get_running_task() is not None:
         messagebox.showinfo("任务进行中", "请等待当前任务完成后再执行新的操作。")
         return
 
-    def run_in_thread():
-        set_running_task("pipeline")
+    # 先选择图片
+    file_path = select_file(
+        log_widget, 
+        [("支持文件", "*.jpg *.jpeg *.png *.bmp *.xlsx *.xls"), ("图片文件", "*.jpg *.jpeg *.png *.bmp"), ("Excel文件", "*.xlsx *.xls"), ("所有文件", "*.*")], 
+        "选择待处理文件"
+    )
+    if not file_path:
+        add_to_log(log_widget, "未选择文件，一键处理已取消\n", "warning")
+        return
 
-        if status_bar:
-            status_bar.set_running(True)
-            status_bar.set_status("开始完整处理流程...")
-
-        start_time = datetime.datetime.now()
-        start_perf = time.perf_counter()
-        log_widget.configure(state=tk.NORMAL)
-        log_widget.delete(1.0, tk.END)
-        log_widget.insert(tk.END, "执行命令: 完整处理流程\n", "command")
-        log_widget.insert(tk.END, f"开始时间: {start_time.strftime('%Y-%m-%d %H:%M:%S')}\n", "time")
-        log_widget.insert(tk.END, "=" * 50 + "\n\n", "separator")
-        log_widget.configure(state=tk.DISABLED)
-
-        try:
-            config = ConfigManager()
-
-            gui_handler = init_gui_logger(log_widget)
-
-            ocr_service = OCRService(config)
-            order_service = OrderService(config)
-
-            reporter = ProgressReporter(status_bar)
-            reporter.running()
-            reporter.set("开始OCR批量处理...", 10)
-
-            total, success = ocr_service.batch_process(progress_cb=lambda p: reporter.set("OCR处理中...", p))
-            if total == 0:
-                add_to_log(log_widget, "没有找到需要处理的图片\n", "warning")
-                if status_bar:
-                    status_bar.set_status("未找到图片文件")
-                return
-            elif success == 0:
-                add_to_log(log_widget, "OCR处理没有成功处理任何新文件\n", "warning")
-            else:
-                add_to_log(log_widget, f"OCR处理完成，共处理 {success}/{total} 个文件\n", "success")
-                try:
-                    processed_map = {}
-                    config = ConfigManager()
-                    pjson = config.get('Paths', 'processed_record', fallback='data/processed_files.json')
-                    if os.path.exists(pjson):
-                        with open(pjson, 'r', encoding='utf-8') as f:
-                            processed_map = json.load(f)
-                    outputs = list(processed_map.values())
-                    for p in outputs[-10:]:
-                        if p:
-                            add_recent_file(os.path.abspath(p))
-                except Exception as e:
-                    logger.debug(f"加载已处理文件记录失败: {e}")
-            reporter.set("开始Excel处理...", 92)
-
-            add_to_log(log_widget, "开始Excel处理...\n", "info")
-            result = order_service.process_excel()
-
-            if not result:
-                add_to_log(log_widget, "Excel处理失败\n", "error")
-            else:
-                add_to_log(log_widget, "Excel处理完成\n", "success")
-                try:
-                    add_recent_file(result)
-                except Exception as e:
-                    logger.debug(f"添加最近文件失败: {e}")
-                try:
-                    validate_unit_price_against_item_data(result, log_widget)
-                except Exception as e:
-                    logger.debug(f"单价校验失败: {e}")
-
-            reporter.set("检查是否需要合并采购单...", 80)
-            _ask_and_merge_purchase_orders(order_service, log_widget, add_to_recent=True)
-
-            end_time = datetime.datetime.now()
-            duration_sec = max(0.0, time.perf_counter() - start_perf)
-
-            add_to_log(log_widget, f"\n{'=' * 50}\n", "separator")
-            add_to_log(log_widget, "完整处理流程执行完毕！\n", "success")
-            add_to_log(log_widget, f"结束时间: {end_time.strftime('%Y-%m-%d %H:%M:%S')}\n", "time")
-            add_to_log(log_widget, f"耗时: {duration_sec:.2f} 秒\n", "time")
-            reporter.set("处理完成", 100)
-
-        except Exception as e:
-            add_to_log(log_widget, f"执行过程中发生错误: {str(e)}\n", "error")
-            import traceback
-            add_to_log(log_widget, f"详细错误信息: {traceback.format_exc()}\n", "error")
-        finally:
-            dispose_gui_logger()
-            reporter.done()
-
-            set_running_task(None)
-            if status_bar:
-                status_bar.set_running(False)
-                status_bar.set_status("就绪")
-
-    thread = Thread(target=run_in_thread)
-    thread.daemon = True
-    thread.start()
+    # 复用拖拽处理的逻辑，实现“先选图，后全流程”
+    process_dropped_file(log_widget, status_bar, file_path)
 
 
 def batch_ocr_with_status(log_widget, status_bar):
@@ -318,7 +254,7 @@ def batch_process_all_inputs(log_widget, status_bar):
             from .memory_editor import show_memory_editor  # noqa: F401  触发 import 顺序
             cfg = ConfigManager()
 
-            svc = BatchService(cfg)
+            svc = BatchService(cfg, missing_barcodes_cb=_get_missing_barcodes_callback(log_widget))
 
             def progress(done, total, entry):
                 pct = int(done / total * 100) if total else 100
@@ -404,7 +340,7 @@ def batch_process_orders_with_status(log_widget, status_bar):
 
             init_gui_logger(log_widget)
 
-            order_service = OrderService()
+            order_service = OrderService(missing_barcodes_cb=_get_missing_barcodes_callback(log_widget))
 
             add_to_log(log_widget, "开始Excel处理...\n", "info")
             try:
@@ -461,7 +397,7 @@ def merge_orders_with_status(log_widget, status_bar):
 
             init_gui_logger(log_widget)
 
-            order_service = OrderService()
+            order_service = OrderService(missing_barcodes_cb=_get_missing_barcodes_callback(log_widget))
 
             result = order_service.merge_all_purchase_orders(progress_cb=lambda p: reporter.set("合并处理中...", p))
 
@@ -511,7 +447,7 @@ def process_excel_file_with_status(log_widget, status_bar):
 
             init_gui_logger(log_widget)
 
-            order_service = OrderService()
+            order_service = OrderService(missing_barcodes_cb=_get_missing_barcodes_callback(log_widget))
 
             if file_path:
                 try:
@@ -589,7 +525,7 @@ def process_dropped_file(log_widget, status_bar, file_path):
 
                     # 步骤2: Excel处理
                     reporter.set("Excel处理中...", 40)
-                    order_service = OrderService()
+                    order_service = OrderService(missing_barcodes_cb=_get_missing_barcodes_callback(log_widget))
                     result = order_service.process_excel(excel_path, progress_cb=lambda p: reporter.set("Excel处理中...", p))
                     if not result:
                         add_to_log(log_widget, "Excel处理失败\n", "error")
@@ -622,7 +558,7 @@ def process_dropped_file(log_widget, status_bar, file_path):
                     reporter = ProgressReporter(status_bar)
                     reporter.running()
                     init_gui_logger(log_widget)
-                    order_service = OrderService()
+                    order_service = OrderService(missing_barcodes_cb=_get_missing_barcodes_callback(log_widget))
                     add_to_log(log_widget, f"开始一键处理Excel文件: {file_path}\n", "info")
                     try:
                         add_recent_file(file_path)

app/ui/barcode_editor.py

@@ -7,7 +7,7 @@ from tkinter import messagebox
 from app.core.excel.converter import UnitConverter
 from app.core.utils.dialog_utils import show_barcode_mapping_dialog
 
-from .logging_ui import add_to_log
+from app.ui.logging_ui import add_to_log
 
 
 def edit_barcode_mappings(log_widget):

app/ui/command_runner.py

@@ -12,8 +12,8 @@ import tkinter as tk
 from tkinter import messagebox
 from threading import Thread
 
-from .logging_ui import LogRedirector
-from .result_previews import show_result_preview
+from app.ui.logging_ui import LogRedirector
+from app.ui.result_previews import show_result_preview
 
 # 任务状态跟踪
 _RUNNING_TASK = None

app/ui/config_dialog.py

@@ -8,8 +8,8 @@ from tkinter import messagebox, filedialog, ttk
 
 from app.config.settings import ConfigManager
 
-from .user_settings import load_user_settings, save_user_settings
-from .ui_widgets import center_window
+from app.ui.user_settings import load_user_settings, save_user_settings
+from app.ui.ui_widgets import center_window
 from app.core.utils.dialog_utils import show_cloud_sync_dialog
 
 

app/ui/db_viewer.py

@@ -0,0 +1,264 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+"""数据库内容查看器模块"""
+
+import os
+import tkinter as tk
+from tkinter import ttk, messagebox
+import sqlite3
+from datetime import datetime
+from typing import Optional, Dict, List, Any
+
+from app.config.settings import ConfigManager
+from app.ui.ui_widgets import center_window
+from app.ui.theme import THEMES, get_theme_mode
+
+class DatabaseViewer:
+    """通用数据库查看器，支持多表切换"""
+    
+    def __init__(self, root, config: Optional[ConfigManager] = None):
+        self.root = root
+        self.config = config or ConfigManager()
+        self.db_path = self.config.get_path('Paths', 'product_db', fallback='data/product_cache.db')
+        
+        # 确保路径是绝对路径
+        if not os.path.isabs(self.db_path):
+            app_root = getattr(self.config, 'app_root', os.getcwd())
+            self.db_path = os.path.join(app_root, self.db_path)
+            
+        self.dlg = tk.Toplevel(root)
+        self.dlg.title("数据库内容查看器")
+        self.dlg.geometry("1000x600")
+        center_window(self.dlg)
+        
+        theme = THEMES[get_theme_mode()]
+        self.dlg.configure(bg=theme["bg"])
+        
+        # 使用 Notebook 实现多表切换
+        self.notebook = ttk.Notebook(self.dlg)
+        self.notebook.pack(fill=tk.BOTH, expand=True, padx=10, pady=10)
+        
+        # 定义要查看的表及其列信息
+        self.table_configs = {
+            "order_metadata": {
+                "title": "订单记录",
+                "columns": {
+                    "bill_date": ("单据日期", 100),
+                    "supplier": ("供应商", 200),
+                    "total_amount": ("总金额", 80),
+                    "updated_at": ("处理时间", 150),
+                    "source_image": ("原始图片", 250),
+                    "file_hash": ("Hash", 120)
+                },
+                "query": "SELECT * FROM order_metadata ORDER BY updated_at DESC"
+            },
+            "missing_barcodes": {
+                "title": "缺失条码",
+                "columns": {
+                    "barcode": ("条码", 150),
+                    "name": ("商品名称", 200),
+                    "count": ("出现次数", 80),
+                    "last_seen": ("最后发现", 150),
+                    "source_file": ("来源文件", 250)
+                },
+                "query": "SELECT * FROM missing_barcodes ORDER BY last_seen DESC"
+            },
+            "products": {
+                "title": "商品记忆库",
+                "columns": {
+                    "barcode": ("条码", 120),
+                    "name": ("名称", 180),
+                    "specification": ("规格", 80),
+                    "unit": ("单位", 50),
+                    "price": ("单价", 70),
+                    "confidence": ("置信度", 60),
+                    "usage_count": ("使用次数", 70),
+                    "last_seen": ("最后使用", 140)
+                },
+                "query": "SELECT * FROM products ORDER BY last_seen DESC"
+            }
+        }
+        
+        self.trees = {}
+        self._init_tabs()
+        
+        # 底部按钮
+        btn_frame = ttk.Frame(self.dlg)
+        btn_frame.pack(fill=tk.X, padx=10, pady=(0, 10))
+        
+        ttk.Button(btn_frame, text="刷新当前表", command=self.refresh_current_tab).pack(side=tk.LEFT, padx=5)
+        ttk.Button(btn_frame, text="清空记录 (慎用)", command=self.clear_current_table).pack(side=tk.LEFT, padx=5)
+        ttk.Button(btn_frame, text="关闭", command=self.dlg.destroy).pack(side=tk.RIGHT, padx=5)
+
+    def _init_tabs(self):
+        """初始化各个标签页"""
+        for table_id, config in self.table_configs.items():
+            frame = ttk.Frame(self.notebook)
+            self.notebook.add(frame, text=config["title"])
+            
+            # 搜索栏
+            search_frame = ttk.Frame(frame)
+            search_frame.pack(fill=tk.X, padx=5, pady=5)
+            
+            ttk.Label(search_frame, text="搜索:").pack(side=tk.LEFT)
+            search_var = tk.StringVar()
+            search_entry = ttk.Entry(search_frame, textvariable=search_var, width=30)
+            search_entry.pack(side=tk.LEFT, padx=5)
+            
+            # Treeview
+            cols = list(config["columns"].keys())
+            tree = ttk.Treeview(frame, columns=cols, show="headings")
+            
+            for col, (text, width) in config["columns"].items():
+                tree.heading(col, text=text)
+                tree.column(col, width=width, anchor="center")
+            
+            scrollbar = ttk.Scrollbar(frame, orient=tk.VERTICAL, command=tree.yview)
+            tree.configure(yscrollcommand=scrollbar.set)
+            
+            tree.pack(side=tk.LEFT, fill=tk.BOTH, expand=True)
+            scrollbar.pack(side=tk.RIGHT, fill=tk.Y)
+            
+            # 双击复制单元格内容
+            tree.bind("<Double-1>", lambda e, tid=table_id: self.copy_cell_value(e, tid))
+            
+            self.trees[table_id] = {
+                "tree": tree,
+                "search_var": search_var,
+                "config": config
+            }
+            
+            # 绑定搜索事件
+            search_var.trace_add("write", lambda *args, tid=table_id: self.load_table_data(tid))
+            
+            # 初始加载数据
+            self.load_table_data(table_id)
+
+    def load_table_data(self, table_id):
+        """加载指定表的数据"""
+        if not os.path.exists(self.db_path):
+            return
+            
+        info = self.trees[table_id]
+        tree = info["tree"]
+        config = info["config"]
+        search_text = info["search_var"].get().lower()
+        
+        # 清空现有数据
+        for item in tree.get_children():
+            tree.delete(item)
+            
+        try:
+            conn = sqlite3.connect(self.db_path)
+            conn.row_factory = sqlite3.Row
+            cursor = conn.cursor()
+            
+            # 检查表是否存在
+            cursor.execute("SELECT name FROM sqlite_master WHERE type='table' AND name=?", (table_id,))
+            if not cursor.fetchone():
+                conn.close()
+                return
+                
+            cursor.execute(config["query"])
+            rows = cursor.fetchall()
+            
+            cols = list(config["columns"].keys())
+            for row in rows:
+                vals = [row[c] if c in row.keys() else "" for c in cols]
+                
+                # 简单搜索过滤
+                if search_text:
+                    match = False
+                    for val in vals:
+                        if search_text in str(val).lower():
+                            match = True
+                            break
+                    if not match:
+                        continue
+                
+                # 格式化金额
+                if "total_amount" in row.keys():
+                    idx = cols.index("total_amount")
+                    try:
+                        vals[idx] = f"{float(vals[idx]):.2f}"
+                    except:
+                        pass
+                
+                tree.insert("", tk.END, values=vals)
+                
+            conn.close()
+            
+            # 自动调整列宽
+            self.auto_resize_columns(table_id)
+            
+        except Exception as e:
+            print(f"加载表 {table_id} 数据失败: {e}")
+
+    def auto_resize_columns(self, table_id):
+        """根据内容自动调整列宽"""
+        info = self.trees[table_id]
+        tree = info["tree"]
+        config = info["config"]
+        
+        for col in list(config["columns"].keys()):
+            # 获取表头宽度
+            header_text = config["columns"][col][0]
+            max_w = len(header_text) * 12 + 20
+            
+            # 获取内容宽度（检查前 20 行）
+            for item in tree.get_children()[:20]:
+                val = str(tree.set(item, col))
+                w = len(val) * 8 + 20
+                if w > max_w:
+                    max_w = w
+            
+            # 限制最大宽度
+            max_w = min(max_w, 400)
+            tree.column(col, width=max_w)
+
+    def copy_cell_value(self, event, table_id):
+        """双击复制单元格内容到剪贴板"""
+        tree = self.trees[table_id]["tree"]
+        region = tree.identify_region(event.x, event.y)
+        if region == "cell":
+            column = tree.identify_column(event.x)
+            item = tree.identify_row(event.y)
+            value = tree.set(item, column)
+            
+            self.root.clipboard_clear()
+            self.root.clipboard_append(value)
+            messagebox.showinfo("成功", f"内容已复制到剪贴板：\n{value}")
+
+    def refresh_current_tab(self):
+        """刷新当前选中的标签页"""
+        current_tab_idx = self.notebook.index(self.notebook.select())
+        table_ids = list(self.table_configs.keys())
+        if current_tab_idx < len(table_ids):
+            self.load_table_data(table_ids[current_tab_idx])
+
+    def clear_current_table(self):
+        """清空当前表的记录"""
+        current_tab_idx = self.notebook.index(self.notebook.select())
+        table_ids = list(self.table_configs.keys())
+        if current_tab_idx >= len(table_ids):
+            return
+            
+        table_id = table_ids[current_tab_idx]
+        title = self.table_configs[table_id]["title"]
+        
+        if not messagebox.askyesno("警告", f"确定要永久清空表 '{title}' 的所有记录吗？此操作不可恢复！"):
+            return
+            
+        try:
+            conn = sqlite3.connect(self.db_path)
+            conn.execute(f"DELETE FROM {table_id}")
+            conn.commit()
+            conn.close()
+            self.load_table_data(table_id)
+            messagebox.showinfo("成功", f"表 '{title}' 已清空")
+        except Exception as e:
+            messagebox.showerror("错误", f"清空表失败: {e}")
+
+def show_db_viewer(root, config: Optional[ConfigManager] = None):
+    """显示数据库查看器"""
+    DatabaseViewer(root, config=config)

app/ui/file_operations.py

@@ -16,7 +16,17 @@ def select_file(log_widget, file_types=None, title="选择文件"):
     """通用文件选择对话框"""
     if file_types is None:
         file_types = [("所有文件", "*.*")]
-    file_path = filedialog.askopenfilename(title=title, filetypes=file_types)
+    
+    # 获取默认输入目录
+    try:
+        config = ConfigManager()
+        initial_dir = config.get_path('Paths', 'input_folder', fallback='data/input')
+        if not os.path.exists(initial_dir):
+            initial_dir = os.getcwd()
+    except Exception:
+        initial_dir = os.getcwd()
+
+    file_path = filedialog.askopenfilename(title=title, filetypes=file_types, initialdir=initial_dir)
     if file_path:
         add_to_log(log_widget, f"已选择文件: {file_path}\n", "info")
     return file_path

app/ui/main_window.py

@@ -13,7 +13,7 @@ from app.core.utils.log_utils import set_log_level
 
 from .theme import THEMES, get_theme_mode, set_theme_mode, create_modern_button, create_card_frame
 from .logging_ui import add_to_log, poll_log_queue
-from .ui_widgets import StatusBar
+from .ui_widgets import StatusBar, ToolTip
 from .user_settings import (
     load_user_settings, save_user_settings, refresh_recent_list_widget,
     _extract_path_from_recent_item, clear_recent_files, RECENT_LIST_WIDGET,
@@ -33,6 +33,7 @@ from .config_dialog import show_config_dialog
 from .barcode_editor import edit_barcode_mappings
 from .shortcuts import bind_keyboard_shortcuts
 from app.core.utils.dialog_utils import show_cloud_sync_dialog
+from .db_viewer import show_db_viewer
 
 
 def _init_window():
@@ -100,8 +101,14 @@ def _create_left_panel(content_frame, theme, log_text, status_bar):
     pipeline_section.pack(fill=tk.X, pady=(0, 8))
     pipeline_frame = tk.Frame(pipeline_section, bg=theme["card_bg"])
     pipeline_frame.pack(fill=tk.X, padx=8, pady=6)
-    create_modern_button(pipeline_frame, "一键处理", lambda: run_pipeline_directly(log_text, status_bar), "primary", px_width=150, px_height=32).pack(anchor='w', pady=3)
-    create_modern_button(pipeline_frame, "一键处理全部图片", lambda: batch_process_all_inputs(log_text, status_bar), "primary", px_width=180, px_height=32).pack(anchor='w', pady=3)
+    
+    btn_onekey = create_modern_button(pipeline_frame, "一键处理", lambda: run_pipeline_directly(log_text, status_bar), "primary", px_width=150, px_height=32)
+    btn_onekey.pack(anchor='w', pady=3)
+    ToolTip(btn_onekey.winfo_children()[0], "选择图片或Excel，自动完成OCR识别和采购单生成全流程")
+    
+    btn_batch = create_modern_button(pipeline_frame, "批量处理", lambda: batch_process_all_inputs(log_text, status_bar), "primary", px_width=150, px_height=32)
+    btn_batch.pack(anchor='w', pady=3)
+    ToolTip(btn_batch.winfo_children()[0], "扫描 data/input 文件夹，处理所有未识别的图片")
 
     # OCR处理区
     core_section = tk.LabelFrame(
@@ -220,7 +227,7 @@ def _create_recent_files_section(parent, theme, log_text):
     create_modern_button(rf_btns, "清理无效", purge_invalid, "primary", px_width=72, px_height=32).pack(side=tk.LEFT, padx=(3, 0))
 
 
-def _create_right_panel(content_frame, theme, log_text, root):
+def _create_right_panel(content_frame, theme, log_text, root, status_bar, config):
     """创建右侧面板：快捷操作、系统设置"""
     right_panel = create_card_frame(content_frame)
     right_panel.pack(side=tk.RIGHT, fill=tk.BOTH, expand=False, padx=(5, 0), pady=5)
@@ -240,12 +247,13 @@ def _create_right_panel(content_frame, theme, log_text, root):
     tk.Frame(tools_buttons_frame, bg=theme["card_bg"]).pack(fill=tk.X, pady=3)
 
     create_modern_button(tools_buttons_frame, "打开结果目录", lambda: open_result_directory(), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
-    create_modern_button(tools_buttons_frame, "打开输出目录", lambda: os.startfile(ConfigManager().get_path('Paths', 'output_folder', fallback='data/output', create=True)), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
-    create_modern_button(tools_buttons_frame, "打开输入目录", lambda: os.startfile(ConfigManager().get_path('Paths', 'input_folder', fallback='data/input', create=True)), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
-    create_modern_button(tools_buttons_frame, "合并订单", lambda: merge_orders_with_status(log_text, StatusBar(root)), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
+    create_modern_button(tools_buttons_frame, "打开输出目录", lambda: os.startfile(config.get_path('Paths', 'output_folder', fallback='data/output', create=True)), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
+    create_modern_button(tools_buttons_frame, "打开输入目录", lambda: os.startfile(config.get_path('Paths', 'input_folder', fallback='data/input', create=True)), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
+    create_modern_button(tools_buttons_frame, "合并订单", lambda: merge_orders_with_status(log_text, status_bar), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
+    create_modern_button(tools_buttons_frame, "数据库内容", lambda: show_db_viewer(root, config=config), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
     create_modern_button(tools_buttons_frame, "清除缓存", lambda: clean_cache(log_text), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
     create_modern_button(tools_buttons_frame, "清理input/out文件", lambda: clean_data_files(log_text), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
-    create_modern_button(tools_buttons_frame, "清理result文件", lambda: clean_result_files(log_text), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
+    create_modern_button(tools_buttons_frame, "清理结果文件", lambda: clean_result_files(log_text), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
 
     # 系统设置区
     settings_section = tk.LabelFrame(
@@ -255,7 +263,7 @@ def _create_right_panel(content_frame, theme, log_text, root):
     settings_section.pack(fill=tk.X, pady=(0, 8))
     settings_buttons_frame = tk.Frame(settings_section, bg=theme["card_bg"])
     settings_buttons_frame.pack(fill=tk.X, padx=8, pady=6)
-    create_modern_button(settings_buttons_frame, "系统设置", lambda: show_config_dialog(root, ConfigManager()), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
+    create_modern_button(settings_buttons_frame, "系统设置", lambda: show_config_dialog(root, config), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
     create_modern_button(settings_buttons_frame, "条码映射", lambda: edit_barcode_mappings(log_text), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
     create_modern_button(settings_buttons_frame, "云端同步", lambda: show_cloud_sync_dialog(root), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
     create_modern_button(settings_buttons_frame, "商品记忆库", lambda: show_memory_editor(root), "primary", px_width=132, px_height=32).pack(anchor='w', pady=3)
@@ -448,6 +456,7 @@ def main():
     """主函数"""
     try:
         root, theme, settings, dnd_supported = _init_window()
+        config = ConfigManager()
 
         # 主容器
         main_container = tk.Frame(root, bg=theme["bg"])
@@ -469,7 +478,7 @@ def main():
         _create_left_panel(content_frame, theme, log_text, status_bar)
 
         # 右侧面板
-        _create_right_panel(content_frame, theme, log_text, root)
+        _create_right_panel(content_frame, theme, log_text, root, status_bar, config)
 
         # 拖拽区域
         _setup_drag_area(mid_container, theme, dnd_supported, log_text, status_bar)

app/ui/memory_editor.py

@@ -6,7 +6,7 @@ from tkinter import ttk, messagebox, simpledialog
 
 from app.config.settings import ConfigManager
 from app.core.db.product_db import ProductDatabase
-from .ui_widgets import center_window
+from app.ui.ui_widgets import center_window
 
 
 def _get_product_db():

app/ui/result_previews.py

@@ -8,8 +8,8 @@ import datetime
 import tkinter as tk
 from tkinter import messagebox, scrolledtext
 
-from .theme import THEMES, get_theme_mode, apply_theme
-from .ui_widgets import center_window
+from app.ui.theme import THEMES, get_theme_mode, apply_theme
+from app.ui.ui_widgets import center_window
 from app.core.utils.file_utils import format_file_size
 from app.config.settings import ConfigManager
 

app/ui/shortcuts.py

@@ -5,15 +5,15 @@
 import tkinter as tk
 from tkinter import messagebox
 
-from .ui_widgets import center_window
-from .action_handlers import (
+from app.ui.ui_widgets import center_window
+from app.ui.action_handlers import (
     process_single_image_with_status,
     process_excel_file_with_status,
     batch_ocr_with_status,
     run_pipeline_directly,
     merge_orders_with_status,
 )
-from .file_operations import clean_cache
+from app.ui.file_operations import clean_cache
 
 
 def bind_keyboard_shortcuts(root, log_widget, status_bar):

app/ui/ui_widgets.py

@@ -74,6 +74,36 @@ class ProgressReporter:
             pass
 
 
+class ToolTip:
+    """给组件添加悬停提示"""
+    def __init__(self, widget, text):
+        self.widget = widget
+        self.text = text
+        self.tip_window = None
+        self.widget.bind("<Enter>", self.show_tip)
+        self.widget.bind("<Leave>", self.hide_tip)
+
+    def show_tip(self, event=None):
+        if self.tip_window or not self.text:
+            return
+        x, y, _cx, cy = self.widget.bbox("insert")
+        x = x + self.widget.winfo_rootx() + 27
+        y = y + cy + self.widget.winfo_rooty() + 27
+        self.tip_window = tw = tk.Toplevel(self.widget)
+        tw.wm_overrideredirect(1)
+        tw.wm_geometry("+%d+%d" % (x, y))
+        label = tk.Label(tw, text=self.text, justify=tk.LEFT,
+                      background="#ffffe0", relief=tk.SOLID, borderwidth=1,
+                      font=("tahoma", "8", "normal"), padx=4, pady=2)
+        label.pack(ipadx=1)
+
+    def hide_tip(self, event=None):
+        tw = self.tip_window
+        self.tip_window = None
+        if tw:
+            tw.destroy()
+
+
 def create_collapsible_frame(parent, title, initial_state=True):
     """创建可折叠的面板"""
     frame = tk.Frame(parent)