fix format

周伟奇
Showing 1 changed file with 46 additions and 14 deletions
src/common/electronic_afc_contract/get_char.py
--- a/src/common/electronic_afc_contract/get_char.py
View file @fdb7ca9
+++ b/src/common/electronic_afc_contract/get_char.py
View file @fdb7ca9
@@ -18,6 +18,7 @@ class Finder:
        self.item = {"words": None,
                     "position": None,
                     }
    def gen_init_result(self, is_asp):
        # 格式化算法输出
        self.init_result = {"page_1": {"合同编号": self.item,
@@ -108,10 +109,12 @@ class Finder:
                                                    "日期": self.item,
                                                    },
                                          }
    def poly_to_rectangle(self, poly):
        xmin, ymin, xmax, ymin, xmax, ymax, xmin, ymax = poly
        bbox = [xmin, ymin, xmax, ymax]
        return bbox
    def get_contract_no(self, page_num):
        """传入页码,查看该页码右上角的编号
@@ -133,6 +136,7 @@ class Finder:
                contract_no['words'] = words
                contract_no['position'] = location
        return contract_no
    def get_vehicle_price(self, page_num='0'):
        vehicle_price = self.item.copy()
        # vehicle_price['words'] = ''
@@ -145,6 +149,7 @@ class Finder:
                vehicle_price['words'] = words
                vehicle_price['position'] = location
        return vehicle_price
    def get_vin(self, page_num='0'):
        vin = self.item.copy()
        # vin['words'] = ''
@@ -157,6 +162,7 @@ class Finder:
                vin['words'] = words
                vin['position'] = location
        return vin
    def get_loan_principal(self, page_num='0'):
        chinese_keywords = ['壹', '贰', '叁', '肆', '伍', '陆', '柒', '捌', '玖', '拾',
                            '佰', '仟', '万', '亿', '元', '角', '分', '零', '整']
@@ -197,6 +203,7 @@ class Finder:
                            asp_2['position'] = bbox
                            asp_2['words'] = words
        return upper, lower, asp_1, asp_2
    def get_loan_term(self, page_num='0'):
        loan_term = self.item.copy()
        all_text = ''
@@ -220,6 +227,7 @@ class Finder:
                            loan_term['position'] = bbox
                            loan_term['words'] = words
        return loan_term
    def mergelist(self, text_list):
        pattern = re.compile("[^\u4e00-\u9fa5]")        # 匹配不是中文的其他字符
        mergeindex = -1
@@ -230,8 +238,10 @@ class Finder:
        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):
        asp_details_table_term = self.item.copy()
        asp_details_table = []
@@ -262,6 +272,7 @@ class Finder:
        if len(asp_details_table) > 0:
            asp_details_table_term['words'] = asp_details_table
        return asp_details_table_term
    def get_signature(self):
        signature = self.item.copy()
        for block in self.pdf_info['0']['blocks']:
@@ -275,6 +286,7 @@ class Finder:
                        signature['words'] = words
                        signature['position'] = bbox
        return signature
    def get_somebody(self, top, bottom):
        # 指定上下边界后,返回上下边界内的客户信息
        _name = self.item.copy()
@@ -309,6 +321,7 @@ class Finder:
                            _id['position'] = bbox
                            _id['words'] = words
        return _name, _id
    def get_seller(self):
        seller = self.item.copy()
        # 先找到 key
@@ -330,11 +343,12 @@ class Finder:
                for line in block['lines']:
                    for span in line['spans']:
                        bbox, text = span['bbox'], span['text']
-                        if anchor_bbox[2]<np.mean(bbox[::2])<half_width and \
+                        if anchor_bbox[2] < np.mean(bbox[::2]) < half_width and \
-                            anchor_bbox[1]<np.mean(bbox[1::2])<anchor_bbox[3]:
+                                anchor_bbox[1] < np.mean(bbox[1::2]) < anchor_bbox[3]:
                            seller['position'] = bbox
                            seller['words'] = text
        return seller
    def get_payback_account(self):
        account = self.item.copy()
        account_name = self.item.copy()
@@ -387,6 +401,7 @@ class Finder:
                                account_bank['position'] = bbox
                                account_bank['words'] = words
        return account, account_name, account_bank
    def get_repayment_schedule(self):
        repayment_schedule = self.item.copy()
        # 只看第二页
@@ -416,6 +431,7 @@ class Finder:
        if len(repayment_schedule_table) > 0:
            repayment_schedule['words'] = repayment_schedule_table
        return repayment_schedule
    def get_signature_role_1(self):
        signature_role_1 = self.init_item.copy()
        # 先定位签字区域
@@ -445,11 +461,13 @@ 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
        return signature_role_1
    def get_signature_role_2(self):
        signature_role_2 = self.init_item.copy()
        # 先定位签字区域
@@ -479,11 +497,13 @@ 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
        return signature_role_2
    def get_signature_role_3(self):
        signature_role_3 = self.init_item.copy()
        # 先定位签字区域
@@ -513,11 +533,13 @@ 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
        return signature_role_3
    def get_signature_role_4(self):
        signature_role_4 = self.init_item.copy()
        # 先定位签字区域
@@ -547,11 +569,13 @@ 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
        return signature_role_4
    def get_signature_role_5(self):
        signature_role_5 = self.init_item.copy()
        # 先定位签字区域
@@ -582,11 +606,13 @@ 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
        return signature_role_5
    def get_last_page_signature(self, page_num, top, bottom):
        signature_name = self.item.copy()
        signature_date = self.item.copy()
@@ -610,7 +636,7 @@ class Finder:
                for line in block['lines']:
                    for span in line['spans']:
                        bbox, text = span['bbox'], span['text']
-                        if '签署日期' in text and int(anchor_top)<np.mean(bbox[1::2])<int(anchor_bottom):
+                        if '签署日期' in text and int(anchor_top) < np.mean(bbox[1::2]) < int(anchor_bottom):
                            name = text.split(' ')[0]
                            date = text.split(':')[-1]
                            signature_name['words'] = name
@@ -618,6 +644,7 @@ class Finder:
                            signature_date['words'] = date
                            signature_date['position'] = bbox
        return signature_name, signature_date
    def get_info(self):
        """
            block['type'] == 0 : 表示该元素为图片
@@ -672,22 +699,27 @@ 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