add jira-4562 content, add income_keywords

chenyao
Showing 2 changed files with 41 additions and 5 deletions
src/apps/doc/consts.py
src/apps/doc/management/commands/ocr_process.py
--- a/src/apps/doc/consts.py
View file @cbe57bd
+++ b/src/apps/doc/consts.py
View file @cbe57bd
@@ -2532,4 +2532,17 @@ FSM_ACTIVITED_STATUS = {
    "APIPN": "Activated-Invoice Passed-Non PT",
    "APIPP": "Activated-Invoice Passed-PT Doc Required",
    "APARD": "Activated-Review done",
+}
+# Jira-4562 - 银行流水首页提取关键词
+INCOME_KEYWORDS_LIST = ["养老金", "社保", "代发工资", "工资入账", "奖金", "养老保险", "代发", "工资"]
+INCOME_KEYWORDS_DICT = {
+    "养老金": "yanglaojin",
+    "社保": "shebao",
+    "代发工资": "daifagongzi",
+    "工资入账": "gongziruzhang",
+    "奖金": "jiangjin",
+    "养老保险": "yanglaobaoxian",
+    "代发": "daifa",
+    "工资": "gongzi"
 }
\ No newline at end of file
--- a/src/apps/doc/management/commands/ocr_process.py
View file @cbe57bd
+++ b/src/apps/doc/management/commands/ocr_process.py
View file @cbe57bd
@@ -177,7 +177,7 @@ class Command(BaseCommand, LoggerMixin):
    #             raise EDMSException(edms_exc)
    #     self.online_log.info('{0} [edms download success] [pdf_path={1}]'.format(self.log_base, pdf_path))
-    def bs_process(self, wb, ocr_data, bs_summary, unknown_summary, classify, res_list, pno, ino, part_idx):
+    def bs_process(self, wb, ocr_data, bs_summary, unknown_summary, classify, res_list, pno, ino, part_idx, income_keywords_dictionary):
        sheets = ocr_data.get('data', [])
        if not sheets:
            res_list.append((pno, ino, part_idx, consts.RES_SUCCESS_EMPTY))
@@ -196,6 +196,10 @@ class Command(BaseCommand, LoggerMixin):
                c1 = cell.get('start_column')
                r1 = cell.get('start_row')
                words = cell.get('words')
+                if words is not None:
+                    if words in consts.INCOME_KEYWORDS_LIST:
+                        if consts.INCOME_KEYWORDS_DICT.get(words) not in income_keywords_dictionary["income_keywords"]:
+                            income_keywords_dictionary["income_keywords"].append(consts.INCOME_KEYWORDS_DICT.setdefault(words, ""))
                ws.cell(row=r1 + 1, column=c1 + 1, value=words)
            # 真伪
@@ -921,7 +925,7 @@ class Command(BaseCommand, LoggerMixin):
            summary['role'] = self.get_most(summary['role'])
        return bs_summary
-    def rebuild_bs(self, bs_summary):
+    def rebuild_bs(self, bs_summary, income_keywords_dictionary):
        # bs_summary = {
        #     '卡号': {
        #         'classify': 0,
@@ -935,7 +939,24 @@ class Command(BaseCommand, LoggerMixin):
        #         'sheet': ['sheet_name']
        #     }
        # }
+        # income_keywords_dictionary = {
+        #     'income_keywords': [ # 其中 0-8 个
+        #         'yanglaojin',
+        #         "shebao",
+        #         "daifagongzi",
+        #         "gongziruzhang",
+        #         "jiangjin",
+        #         "yanglaobaoxian",
+        #         "daifa",
+        #         "gongzi"
+        #     ]
+        # }
        res = []
+        income_keywords_list = income_keywords_dictionary.get('income_keywords', [])
+        income_filtered_keywords = [keyword_str for keyword_str in income_keywords_list if keyword_str]
+        income_keywords_str = ",".join(income_filtered_keywords)
        for bs_info in bs_summary.values():
            try:
                print_date = bs_info.get('print_time', '').strftime("%Y-%m-%d")
@@ -950,7 +971,8 @@ class Command(BaseCommand, LoggerMixin):
                    'print_time': print_date,
                    'timedelta': bs_info.get('timedelta', ''),
                    'verify': bs_info.get('verify_res_ebank', True),
-                    'e_bank': bs_info.get('e_bank', False)
+                    'e_bank': bs_info.get('e_bank', False),
+                    'income_keywords': income_keywords_str
                }
            )
        return res
@@ -1724,6 +1746,7 @@ class Command(BaseCommand, LoggerMixin):
                        license_summary = {}
                        contract_result = {}
                        contract_result_compare = {}
+                        income_keywords_dictionary = {"income_keywords": []}
                        res_list = []
                        interest_keyword = Keywords.objects.filter(
                            type=KeywordsType.INTEREST.value, on_off=True).values_list('keyword', flat=True)
@@ -1879,7 +1902,7 @@ class Command(BaseCommand, LoggerMixin):
                                                                  ino, part_idx, img_path, contract_result_compare)
                                        else:  # 流水处理
                                            bs_classify_set.add(classify)
-                                            self.bs_process(wb, ocr_data, bs_summary, unknown_summary, classify, res_list, pno, ino, part_idx)
+                                            self.bs_process(wb, ocr_data, bs_summary, unknown_summary, classify, res_list, pno, ino, part_idx, income_keywords_dictionary)
                            else:
                                res_list.append((pno, ino, part_idx, consts.RES_FAILED_1))
                                self.online_log.info('{0} [ocr_1 res error] [img={1}]'.format(self.log_base, img_path))
@@ -2035,7 +2058,7 @@ class Command(BaseCommand, LoggerMixin):
                                                                     license_summary, contract_result_compare))
                            self.rebuild_contract(license_summary, contract_result_compare)
-                            bs_rebuild = self.rebuild_bs(merged_bs_summary)
+                            bs_rebuild = self.rebuild_bs(merged_bs_summary, income_keywords_dictionary)
                            if len(bs_rebuild) > 0:
                                license_summary[consts.BS_CLASSIFY] = bs_rebuild