v1.2.91 行情录入优化版：批量录入冠字号矫正、价格提取、编码规则统一

backend/app/routers/information.py

@@ -452,19 +452,57 @@ def create_information(
     db: Session = Depends(get_db)
 ):
     """发布资讯"""
-    # 生成行情编号：日期 + 5位自然数（从00001开始）
+    # 验证并矫正冠字号：必须是J0开头 + 8位数字 = 共10位
+    if data.title:
+        import re
+        # 提取冠字号（J0开头后面跟数字）
+        match = re.search(r'J0(\d+)', data.title)
+        if match:
+            num = match.group(1)
+            # 必须是8位数字
+            if len(num) > 8:
+                # 多于8位：取前8位
+                num = num[:8]
+            elif len(num) < 8:
+                # 少于8位：前面补0
+                num = num.zfill(8)
+            # 重新构建title，确保是J0开头
+            original = match.group(0)
+            data.title = data.title.replace(original, 'J0' + num, 1)
+        else:
+            # 如果不是J0开头，尝试转换
+            other_match = re.search(r'J([1-9]\d{0,8})', data.title)
+            if other_match:
+                # 非J0开头的，尝试补0变成J0开头
+                num = other_match.group(1).zfill(8)[:8]
+                original = other_match.group(0)
+                data.title = data.title.replace(original, 'J0' + num, 1)
+    # 生成行情编号：年份后两位+月+日+当日序号（如260414001）
+    # 序号在当天最大序号基础上+1
     deal_no = None
     if data.info_type == 'deal':
-        today = datetime.now().strftime('%Y%m%d')
-        # 查询当天已有行情数量
         from app.models.models import Information
-        count_today = db.query(Information).filter(
+        today = datetime.now().strftime('%y%m%d')  # 如260414
+        # 查询当天最大的deal_no
+        max_deal = db.query(Information.deal_no).filter(
             Information.info_type == 'deal',
-            Information.deal_no.like(f'DJ{today}%')
+            Information.deal_no.isnot(None),
-        ).count()
+            Information.deal_no != '',
-        # 编号 = 日期 + 5位自然数（如 DJ2026041100001）
+            Information.deal_no.like(f'{today}%')
-        seq = count_today + 1
+        ).order_by(Information.deal_no.desc()).first()
-        deal_no = f'DJ{today}{seq:05d}'
+        
+        if max_deal and max_deal[0] and max_deal[0].startswith(today):
+            # 当天已有编号，提取序号并+1
+            try:
+                current_seq = int(max_deal[0][6:])  # 取最后3位序号
+                new_seq = current_seq + 1
+            except:
+                new_seq = 1
+        else:
+            # 新的一天，从001开始
+            new_seq = 1
+        
+        deal_no = f'{today}{new_seq:03d}'
     
     info = Information(
         user_id=current_user.f99_90_id,
@@ -511,6 +549,12 @@ def create_information(
         view_count=info.view_count,
         contact_count=info.contact_count,
         created_at=info.created_at,
+        packaging=info.packaging,
+        is_graded=info.is_graded,
+        grading_company=info.grading_company,
+        grading_score=info.grading_score,
+        category=info.category,
+        deal_no=info.deal_no,
         user_name=current_user.f01_01_name,
         user_avatar=current_user.avatar,
         collection_name=None,
@@ -1277,7 +1321,45 @@ async def batch_parse_deals(text: str = Body(..., embed=True)):
                 
                 # 尝试直接解析
                 data = json.loads(content.strip())
-                return {"success": True, "data": data}
+                # 对AI返回的数据进行冠字号矫正 - 确保J0开头+8位数字
+                def normalize_serial_ai(num_str):
+                    """矫正冠字号：J0开头，8位数字，共10位"""
+                    if not num_str.startswith('J0'):
+                        return None
+                    num = num_str[2:]  # 去掉J0
+                    diff = 8 - len(num)
+                    # 位数正好8位，不需要矫正
+                    if diff == 0:
+                        return num_str
+                    elif diff == -1:
+                        # 多1位：取前4位+最后4位
+                        if len(num) >= 4:
+                            result = num[:4] + num[-4:]
+                            if len(result) == 8:
+                                return 'J0' + result
+                    elif diff == 1:
+                        # 少1位：J0 + 0 + 数字
+                        return 'J0' + '0' + num
+                    elif diff == -2:
+                        # 多2位：取前4位+最后4位
+                        if len(num) >= 4:
+                            result = num[:4] + num[-4:]
+                            if len(result) == 8:
+                                return 'J0' + result
+                    elif diff == 2:
+                        # 少2位：J0 + 00 + 数字
+                        return 'J0' + '00' + num
+                    return None
+                
+                # 矫正每条记录的冠字号
+                corrected_data = []
+                for item in data:
+                    if 'serial' in item:
+                        normalized = normalize_serial_ai(item['serial'])
+                        if normalized:
+                            item['serial'] = normalized
+                            corrected_data.append(item)
+                return {"success": True, "data": corrected_data}
             except json.JSONDecodeError:
                 # 尝试用正则提取
                 match = re.search(r'\[.*\]', content, re.DOTALL)
@@ -1330,24 +1412,59 @@ def parse_deals_locally(text: str, default_packaging: str = '', default_date: st
     default_date = datetime.now().strftime('%Y-%m-%d')
     deal_date = extracted_date or default_date
     
+    # 冠字号矫正函数 - 确保是J0开头+8位数字
+    def normalize_serial(num_str, needs_j0=False):
+        """矫正冠字号：J0开头，8位数字，共10位"""
+        num = num_str
+        # 如果是单独的7-9位数字（无J0前缀），取后8位
+        if needs_j0 and len(num) > 8:
+            num = num[-8:]
+        
+        diff = 8 - len(num)
+        
+        # 位数正好8位，不需要矫正
+        if diff == 0:
+            return 'J0' + num
+        # 少1位（7位数字）→ J0 + 0 + 数字
+        elif diff == 1:
+            return 'J0' + '0' + num
+        # 多1位（9位数字）→ J0 + 取前4位+后4位 = 8位数字
+        elif diff == -1:
+            if len(num) >= 8:
+                result = num[:4] + num[-4:]
+                return 'J0' + result
+        return None  # 无法矫正
+    
     lines = text.strip().split('\n')
     
     for line in lines:
         line = line.strip()
-        if not line or 'J0' not in line:
+        if not line:
             continue
         
-        # 提取冠字号 J0 + 8-9位数字
+        # 提取冠字号：J0 + 至少7位数字（可能到10位），或者单独的7~9位数字
-        serial_match = re.search(r'J0(\d{8,9})', line)
+        # 优先匹配 J0开头，否则匹配单独的7-9位数字
+        serial_num = None
+        serial_match = re.search(r'J0(\d{7,11})', line)
         if not serial_match:
-            continue
+            # 尝试匹配单独的7-9位数字（前面是空格或行首）
+            serial_match = re.search(r'(?:^|\s)(\d{7,9})(?:\s|$|\n)', line)
+            if serial_match:
+                # 单独的7-9位数字需要加上J0前缀
+                num = serial_match.group(1)
+                serial_num = num
+                serial = normalize_serial(num, needs_j0=True)
+            else:
+                continue
+        else:
+            # J0开头的7-9位数字
+            serial_num = serial_match.group(1)
+            serial = normalize_serial(serial_num, needs_j0=False)
         
-        serial_num = serial_match.group(1)
+        if not serial:
-        if len(serial_num) == 9:
+            continue  # 跳过无法矫正的数据
-            serial_num = serial_num[:8]
-        serial = 'J0' + serial_num
         
-        # 提取价格 ¥xxx,xxx 或 xxx,xxx（必须在J0之后）
+        # 提取价格：在冠字号之后提取价格
         serial_pos = line.find(serial)
         after_serial = line[serial_pos + len(serial):]
         price_match = re.search(r'[¥￥]?\s*([1-9][\d,]{2,})', after_serial)
@@ -1356,13 +1473,13 @@ def parse_deals_locally(text: str, default_packaging: str = '', default_date: st
         price = int(price_match.group(1).replace(',', ''))
         
         # 提取卖家（价格后面的中文字符）
-        after_price_pos = after_serial.find(price_match.group(0)) + len(price_match.group(0))
+        after_price_pos = line.find(price_match.group(0)) + len(price_match.group(0))
-        after_price = after_serial[after_price_pos:]
+        after_price = line[after_price_pos:]
         seller_match = re.search(r'([^\d\s¥￥]+?)(?:\s*$|\s*\n)', after_price)
         seller = seller_match.group(1).strip() if seller_match else ''
         
-        # 分类判断
+        # 分类判断（使用矫正后的冠字号去掉J0前缀的8位数字）
-        digits = serial_num
+        digits = serial[2:]  # 去掉J0前缀
         d = digits
         category = '通货'
         if not any(c in d for c in ['1','2','3','4','5','7']): category = '圆圆号'