root
/
jiachenlong
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

修正冠字号矫正逻辑

This commit is contained in:
甲辰生产 2026-04-12 00:56:34 +08:00
parent dd1790b0be
commit 7db66d5a7b
1 changed files with 30 additions and 32 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

62
backend/app/routers/information.py
View File

@ -455,18 +455,18 @@ def create_information(
# 验证并矫正冠字号：J0 + 8位数字 = 共10位 # 验证并矫正冠字号：J0 + 8位数字 = 共10位
if data.title: if data.title:
import re import re
# 提取冠字号（J开头后面跟数字） # 提取冠字号（J0开头后面跟数字）
match = re.search(r'J(\d+)', data.title) match = re.search(r'J0(\d+)', data.title)
if match: if match:
num = match.group(1) num = match.group(1)
# 必须是10位：J + 9位数字 # 必须是10位：J0 + 8位数字
if len(num) > 9: if len(num) > 8:
num = num[:9] num = num[:8]
elif len(num) < 9: elif len(num) < 8:
num = num.zfill(9) num = num.zfill(8)
# 重新构建title # 重新构建title
original = match.group(0) original = match.group(0)
data.title = data.title.replace(original, 'J' + num, 1) data.title = data.title.replace(original, 'J0' + num, 1)
# 生成行情编号：日期 + 5位自然数（从00001开始） # 生成行情编号：日期 + 5位自然数（从00001开始）
deal_no = None deal_no = None
if data.info_type == 'deal': if data.info_type == 'deal':
@ -1294,31 +1294,32 @@ async def batch_parse_deals(text: str = Body(..., embed=True)):
data = json.loads(content.strip()) data = json.loads(content.strip())
# 对AI返回的数据进行冠字号矫正 # 对AI返回的数据进行冠字号矫正
def normalize_serial_ai(num_str): def normalize_serial_ai(num_str):
"""矫正冠字号：J+9位数字（共10位）""" """矫正冠字号：J0开头，9位数字"""
if not num_str.startswith('J'): if not num_str.startswith('J0'):
return None return None
num = num_str[1:] # 去掉J num = num_str[2:] # 去掉J0
diff = 9 - len(num) diff = 9 - len(num)
# 位数正好9位，不需要矫正 # 位数正好9位，不需要矫正
if diff == 0: if diff == 0:
return num_str return num_str
elif diff == -1: elif diff == -1:
# 多1位：取前5位+最后4位
if len(num) >= 4: if len(num) >= 4:
result = num[:5] + num[-4:] result = num[:5] + num[-4:]
if len(result) == 9: if len(result) == 9:
return 'J' + result return 'J0' + result
result2 = num[:4] + num[-5:]
if len(result2) == 9:
return 'J' + result2
elif diff == 1: elif diff == 1:
return 'J' + '0' + num # 少1位：J0 + 0 + 数字
return 'J0' + '0' + num
elif diff == -2: elif diff == -2:
# 多2位：取前5位+最后4位
if len(num) >= 5: if len(num) >= 5:
result = num[:5] + num[-4:] result = num[:5] + num[-4:]
if len(result) == 9: if len(result) == 9:
return 'J' + result return 'J0' + result
elif diff == 2: elif diff == 2:
return 'J' + '00' + num # 少2位：J0 + 00 + 数字
return 'J0' + '00' + num
return None return None
# 矫正每条记录的冠字号 # 矫正每条记录的冠字号
@ -1394,37 +1395,34 @@ def parse_deals_locally(text: str, default_packaging: str = '', default_date: st
if not serial_match: if not serial_match:
continue continue
# 冠字号矫正函数 - 统一为J+9位数字（10位） # 冠字号矫正函数 - J0开头的情况
def normalize_serial(num_str): def normalize_serial(num_str):
"""矫正冠字号：J+9位数字（共10位）""" """矫正冠字号：J0开头，9位数字"""
num = num_str num = num_str
diff = 9 - len(num) diff = 9 - len(num)
# 位数正好9位，不需要矫正 # 位数正好9位，不需要矫正
if diff == 0: if diff == 0:
return 'J' + num return 'J0' + num
# 位数不对才需要矫正 # 位数不对才需要矫正
elif diff == -1: elif diff == -1:
# 多1位（10位数字）→ 去掉倒数第4位 # 多1位（10位数字）→ 取前5位+最后4位
if len(num) >= 4: if len(num) >= 4:
result = num[:5] + num[-4:] result = num[:5] + num[-4:]
if len(result) == 9: if len(result) == 9:
return 'J' + result return 'J0' + result
result2 = num[:4] + num[-5:]
if len(result2) == 9:
return 'J' + result2
elif diff == 1: elif diff == 1:
# 少1位（8位数字）→ 在J后加0 # 少1位（8位数字）→ J0 + 0 + 数字
return 'J' + '0' + num return 'J0' + '0' + num
elif diff == -2: elif diff == -2:
# 多2位（11位数字）→ 去掉倒数第4、5位 # 多2位（11位数字）→ 取前5位+最后4位
if len(num) >= 5: if len(num) >= 5:
result = num[:5] + num[-4:] result = num[:5] + num[-4:]
if len(result) == 9: if len(result) == 9:
return 'J' + result return 'J0' + result
elif diff == 2: elif diff == 2:
# 少2位（7位数字）→ 在J后加00 # 少2位（7位数字）→ J0 + 00 + 数字
return 'J' + '00' + num return 'J0' + '00' + num
return None # 无法矫正 return None # 无法矫正
serial_num = serial_match.group(1) serial_num = serial_match.group(1)