Skip to content

周伟奇 / bmw-ocr

Go to a project
Toggle navigation
Toggle navigation pinning
c113f2a4 by 周伟奇
Options

add ltgt bzll

1 parent 12319785
Inline Side-by-side
Showing 2 changed files with 38 additions and 28 deletions
src/apps/doc/consts.py
......@@ -1434,6 +1434,7 @@ ASP_KEY = 'is_asp'
AFC_CON_MAP_LTGT = {
'合同编号': (2, 2, '合同编号', None),
'标准利率': (2, 2, '标准利率', None),
'借款人姓名': (2, 2, '借款人及抵押人', 'name'),
'共借人姓名': (2, 2, '共同借款人及共同抵押人', 'name'),
'保证人姓名1': (2, 2, '保证人1', 'name'),
......@@ -1446,6 +1447,7 @@ AFC_CON_FIELD_ORDER_LTGT = (
('共借人姓名', '共借人姓名'),
('保证人姓名1', '保证人姓名1'),
('保证人姓名2', '保证人姓名2'),
('标准利率', '标准利率'),
)
SE_AFC_CON_MAP = {
......
src/common/electronic_afc_contract/get_char.py
......@@ -55,6 +55,7 @@ class Finder:
"附加产品融资贷款本金总金额": self.item,
},
"贷款期限": self.item,
"标准利率": self.item,
"还款账户": {"账号": self.item,
"户名": self.item,
"开户行": self.item,
......@@ -228,18 +229,32 @@ class Finder:
loan_term['words'] = words
return loan_term
def get_standard_rate(self, page_num='0'):
standard_rate = self.item.copy()
for block in self.pdf_info[page_num]['blocks']:
if block['type'] != 0:
continue
for line in block['lines']:
for span in line['spans']:
bbox, text = span['bbox'], span['text']
matchs = re.search(r'本合同当期的标准利率为(\S+)%/年', text)
if matchs:
standard_rate['position'] = bbox
standard_rate['words'] = matchs.group(1)
return standard_rate
def mergelist(self, text_list):
pattern = re.compile("[^\u4e00-\u9fa5]") # 匹配不是中文的其他字符
mergeindex = -1
for index, i in enumerate(text_list):
if '所购' in i and len(pattern.sub('', pattern.sub('', text_list[index+1]))) != 0:
if '所购' in i and len(pattern.sub('', pattern.sub('', text_list[index + 1]))) != 0:
# if '所购' in i and '.00' not in text_list[index+1]:
mergeindex = index
if mergeindex == -1:
return text_list
else:
new_text_list = text_list[:mergeindex] + [
text_list[mergeindex] + text_list[mergeindex+1]] + text_list[mergeindex+2:]
new_text_list = text_list[:mergeindex] + [text_list[mergeindex] + text_list[mergeindex + 1]] + \
text_list[mergeindex + 2:]
return self.mergelist(new_text_list)
def get_asp_details(self, page_num):
......@@ -261,13 +276,13 @@ class Finder:
# print(text)
asp_details_text_list.append(text)
asp_details_text_list = self.mergelist(asp_details_text_list)
for i in range((len(asp_details_text_list)+2)//3):
for i in range((len(asp_details_text_list) + 2) // 3):
line = []
if i == 0:
line = [asp_details_text_list[0]]
else:
for j in range(3):
line.append(asp_details_text_list[i*3-2+j])
line.append(asp_details_text_list[i * 3 - 2 + j])
asp_details_table.append(line)
if len(asp_details_table) > 0:
asp_details_table_term['words'] = asp_details_table
......@@ -420,12 +435,12 @@ class Finder:
table = False
if table == True:
repayment_schedule_text_list.append(text)
for i in range(len(repayment_schedule_text_list)//5):
for i in range(len(repayment_schedule_text_list) // 5):
line = []
# 5表示5列的意思
for j in range(5):
line.append(repayment_schedule_text_list[i*5+j])
if str(i+1) == line[1]:
line.append(repayment_schedule_text_list[i * 5 + j])
if str(i + 1) == line[1]:
break
repayment_schedule_table.append(line)
if len(repayment_schedule_table) > 0:
......@@ -461,8 +476,7 @@ class Finder:
else:
words = '无'
boxes = np.array(boxes).reshape((-1, 2))
position = [min(boxes[:, 0]), min(boxes[:, 1]),
max(boxes[:, 0]), max(boxes[:, 1])]
position = [min(boxes[:, 0]), min(boxes[:, 1]), max(boxes[:, 0]), max(boxes[:, 1])]
signature_role_1['page_num'] = page_num
signature_role_1['position'] = position
signature_role_1['words'] = words
......@@ -497,8 +511,7 @@ class Finder:
else:
words = '无'
boxes = np.array(boxes).reshape((-1, 2))
position = [min(boxes[:, 0]), min(boxes[:, 1]),
max(boxes[:, 0]), max(boxes[:, 1])]
position = [min(boxes[:, 0]), min(boxes[:, 1]), max(boxes[:, 0]), max(boxes[:, 1])]
signature_role_2['page_num'] = page_num
signature_role_2['position'] = position
signature_role_2['words'] = words
......@@ -533,8 +546,7 @@ class Finder:
else:
words = '无'
boxes = np.array(boxes).reshape((-1, 2))
position = [min(boxes[:, 0]), min(boxes[:, 1]),
max(boxes[:, 0]), max(boxes[:, 1])]
position = [min(boxes[:, 0]), min(boxes[:, 1]), max(boxes[:, 0]), max(boxes[:, 1])]
signature_role_3['page_num'] = page_num
signature_role_3['position'] = position
signature_role_3['words'] = words
......@@ -569,8 +581,7 @@ class Finder:
else:
words = '无'
boxes = np.array(boxes).reshape((-1, 2))
position = [min(boxes[:, 0]), min(boxes[:, 1]),
max(boxes[:, 0]), max(boxes[:, 1])]
position = [min(boxes[:, 0]), min(boxes[:, 1]), max(boxes[:, 0]), max(boxes[:, 1])]
signature_role_4['page_num'] = page_num
signature_role_4['position'] = position
signature_role_4['words'] = words
......@@ -606,8 +617,7 @@ class Finder:
else:
words = '无'
boxes = np.array(boxes).reshape((-1, 2))
position = [min(boxes[:, 0]), min(boxes[:, 1]),
max(boxes[:, 0]), max(boxes[:, 1])]
position = [min(boxes[:, 0]), min(boxes[:, 1]), max(boxes[:, 0]), max(boxes[:, 1])]
signature_role_5['page_num'] = page_num
signature_role_5['position'] = position
signature_role_5['words'] = words
......@@ -699,27 +709,22 @@ class Finder:
contract_no = self.get_contract_no(page_num='0')
self.init_result['page_2']['合同编号'] = contract_no
# 找借款人及抵押人(地址字段原本有空格)
borrower_name, borrower_id = self.get_somebody(
top='借款人及抵押人：', bottom='共同借款人：')
borrower_name, borrower_id = self.get_somebody(top='借款人及抵押人：', bottom='共同借款人：')
# 这是为了同时兼容 8.1 版本
if borrower_name['words'] == None:
borrower_name, borrower_id = self.get_somebody(
top='借款人及抵押人：', bottom='共同借款人及共同抵押人：')
borrower_name, borrower_id = self.get_somebody(top='借款人及抵押人：', bottom='共同借款人及共同抵押人：')
self.init_result['page_2']['借款人及抵押人']['name'] = borrower_name
self.init_result['page_2']['借款人及抵押人']['id'] = borrower_id
# 找共同借款人及共同抵押人
co_borrower_name, co_borrower_id = self.get_somebody(
top='共同借款人：', bottom='保证人1：')
co_borrower_name, co_borrower_id = self.get_somebody(top='共同借款人：', bottom='保证人1：')
self.init_result['page_2']['共同借款人及共同抵押人']['name'] = co_borrower_name
self.init_result['page_2']['共同借款人及共同抵押人']['id'] = co_borrower_id
# 保证人1
first_guarantor_name, first_guarantor_id = self.get_somebody(
top='保证人1：', bottom='保证人2：')
first_guarantor_name, first_guarantor_id = self.get_somebody(top='保证人1：', bottom='保证人2：')
self.init_result['page_2']['保证人1']['name'] = first_guarantor_name
self.init_result['page_2']['保证人1']['id'] = first_guarantor_id
# 保证人2
second_guarantor_name, second_guarantor_id = self.get_somebody(
top='保证人2：', bottom='第一章')
second_guarantor_name, second_guarantor_id = self.get_somebody(top='保证人2：', bottom='第一章')
self.init_result['page_2']['保证人2']['name'] = second_guarantor_name
self.init_result['page_2']['保证人2']['id'] = second_guarantor_id
# 所购车辆价格
......@@ -740,6 +745,9 @@ class Finder:
# 贷款期限
loan_term = self.get_loan_term(page_num='1')
self.init_result['page_2']['贷款期限'] = loan_term
# 本合同当期的标准利率
standard_rate = self.get_standard_rate(page_num='1')
self.init_result['page_2']['标准利率'] = standard_rate
# 还款账户
account, account_name, account_bank = self.get_payback_account()
self.init_result['page_2']['还款账户']['账号'] = account
......
Styling with Markdown is supported
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!