Merge branch 'hotfix/2025-02'

冯轩
Showing 3 changed files with 68 additions and 29 deletions
src/apps/doc/management/commands/ocr_process.py
src/celery_compare/tasks.py
src/settings/__init__.py
--- a/src/apps/doc/management/commands/ocr_process.py
View file @d6fc968
+++ b/src/apps/doc/management/commands/ocr_process.py
View file @d6fc968
@@ -2130,9 +2130,19 @@ class Command(BaseCommand, LoggerMixin):
                                        # 更新OCR累计识别结果表
                                        if business_type == consts.HIL_PREFIX:
                                            result_class = HILOCRResult if is_ca else HILSEOCRResult
+                                            try:
+                                                res_obj = atomicSaveDBHIL(self,result_class,doc,license_summary,ic_merge,rp_merge, task_str, financial_statement_dict, financial_explanation_dict)
+                                            except Exception as e:
+                                            # 遇到报错重试一次，希望解决两个文件首次入库都插入的问题
+                                                self.online_log.error('{0} [process error (ocr result save) retry] [task={1}] [error={2}]'.format(self.log_base, task_str, traceback.format_exc()))
                                                res_obj = atomicSaveDBHIL(self,result_class,doc,license_summary,ic_merge,rp_merge, task_str, financial_statement_dict, financial_explanation_dict)
                                        else:
                                            result_class = AFCOCRResult if is_ca else AFCSEOCRResult
+                                            try:
+                                                res_obj = atomicSaveDBAFC(self,result_class,doc,license_summary,ic_merge,rp_merge, task_str, financial_statement_dict, financial_explanation_dict)
+                                            except Exception as e:
+                                            # 遇到报错重试一次，希望解决两个文件首次入库都插入的问题
+                                                self.online_log.error('{0} [process error (ocr result save) retry] [task={1}] [error={2}]'.format(self.log_base, task_str, traceback.format_exc()))
                                                res_obj = atomicSaveDBAFC(self,result_class,doc,license_summary,ic_merge,rp_merge, task_str, financial_statement_dict, financial_explanation_dict)
                                    except Exception as e:
@@ -2172,9 +2182,19 @@ class Command(BaseCommand, LoggerMixin):
                                    # 更新OCR累计识别结果表
                                    if business_type == consts.HIL_PREFIX:
                                        result_class = HILOCRResult if is_ca else HILSEOCRResult
+                                        try:
+                                            res_obj = atomicSaveDBHIL(self,result_class,doc,license_summary,ic_merge,rp_merge, task_str, financial_statement_dict, financial_explanation_dict)
+                                        except Exception as e:
+                                            # 遇到报错重试一次，希望解决两个文件首次入库都插入的问题
+                                            self.online_log.error('{0} [process error (ocr result save) retry] [task={1}] [error={2}]'.format(self.log_base, task_str, traceback.format_exc()))
                                            res_obj = atomicSaveDBHIL(self,result_class,doc,license_summary,ic_merge,rp_merge, task_str, financial_statement_dict, financial_explanation_dict)
                                    else:
                                        result_class = AFCOCRResult if is_ca else AFCSEOCRResult
+                                        try:
+                                            res_obj = atomicSaveDBAFC(self,result_class,doc,license_summary,ic_merge,rp_merge, task_str, financial_statement_dict, financial_explanation_dict)
+                                        except Exception as e:
+                                            # 遇到报错重试一次，希望解决两个文件首次入库都插入的问题
+                                            self.online_log.error('{0} [process error (ocr result save) retry] [task={1}] [error={2}]'.format(self.log_base, task_str, traceback.format_exc()))
                                            res_obj = atomicSaveDBAFC(self,result_class,doc,license_summary,ic_merge,rp_merge, task_str, financial_statement_dict, financial_explanation_dict)
                                except Exception as e:
                                    self.online_log.error(
--- a/src/celery_compare/tasks.py
View file @d6fc968
+++ b/src/celery_compare/tasks.py
View file @d6fc968
@@ -3773,20 +3773,29 @@ def fsm_compare(application_id, application_entity, uniq_seq, ocr_res_id, is_ca=
    compare_log.info('[fsm thread]')
    #pool = ThreadPoolExecutor(max_workers=6, thread_name_prefix="fsm_thread_")
+    try:
+        # 这个try不生效
        pool.submit(fsm_compare_thread, application_id, application_entity, uniq_seq, ocr_res_id, is_ca, is_cms)
+    except Exception as e:
+        compare_log.info('[fsm thread fail] [error={0}]'.format(traceback.format_exc()))
    #pool.shutdown(wait=True)
 def fsm_compare_thread(application_id, application_entity, uniq_seq, ocr_res_id, is_ca=True, is_cms=False):
+    try:
        compare_log.info('{0} [receive fsm task] [entity={1}] [id={2}] [uniq_seq={3}] [ocr_res_id={4}] [is_ca={5}] '
                     '[is_cms={6}]'.format(log_base, application_entity, application_id, uniq_seq, ocr_res_id,
                                           is_ca, is_cms))
    # 查看此订单号下是否有未完成的文件，如果有，等1分钟
+        try:
            doc_wait_file_class = HILDoc if application_entity == consts.HIL_PREFIX else AFCDoc
            doc_wait_file_result = doc_wait_file_class.objects.filter(application_id=application_id, status=1).first()
            compare_log.info('doc_wait_file_result:{0}'.format(doc_wait_file_result))
            compare_log.info('{0} [comparison unfinished file check] [entity={1}] [id={2}] [doc_wait_file_result={3}]'.format(log_base, application_entity, application_id, doc_wait_file_result))
+        except Exception as e:
+            doc_wait_file_result = None
+            compare_log.info('[get doc_wait_file_result fail] [error={0}]'.format(traceback.format_exc()))
        if doc_wait_file_result is not None:
            # 实时查询延迟时间
            try:
@@ -3806,7 +3815,6 @@ def fsm_compare_thread(application_id, application_entity, uniq_seq, ocr_res_id,
                compare_log.info('[sleep error] [error={0}]'.format(traceback.format_exc()))
            compare_log.info('{0} [comparison unfinished file wait delay_time end] [entity={1}] [id={2}] [doc_id={3}]'.format(log_base, application_entity, application_id, doc_wait_file_result.id))
        # 调用java fsm 比对流程接口（http）
        # 调用Java fsm 比对流程接口, fsm 是se流程, ca可以暂时忽略
        auto_class = HILAutoSettlement if application_entity == consts.HIL_PREFIX else AFCAutoSettlement
@@ -3835,6 +3843,8 @@ def fsm_compare_thread(application_id, application_entity, uniq_seq, ocr_res_id,
        except Exception as e:
            compare_log.error("fsm full request to java error, url:{0}, param:{1}, errorMsg:{2}".format(
                url, json.dumps(body), traceback.format_exc()))
+    except Exception as e:
+        compare_log.info('[fsm_compare_thread error] [error={0}]'.format(traceback.format_exc()))
 @app.task
@@ -3851,42 +3861,21 @@ def compare(application_id, application_entity, uniq_seq, ocr_res_id, is_ca=True
    compare_log.info('[non fsm thread]')
    #pool = ThreadPoolExecutor(max_workers=6, thread_name_prefix="non_fsm_thread_")
+    try:
+        # 这个try不生效
        pool.submit(compare_thread, application_id, application_entity, uniq_seq, ocr_res_id, is_ca, is_cms)
+    except Exception as e:
+        compare_log.info('[non fsm thread fail] [error={0}]'.format(traceback.format_exc()))
    #pool.shutdown(wait=True)
 def compare_thread(application_id, application_entity, uniq_seq, ocr_res_id, is_ca=True, is_cms=False):
    # POS: application_id, application_entity, uniq_seq, None
    # OCR: application_id, business_type(application_entity), None, ocr_res_id
+    try:
        compare_log.info('{0} [receive task] [entity={1}] [id={2}] [uniq_seq={3}] [ocr_res_id={4}] [is_ca={5}] '
                     '[is_cms={6}]'.format(log_base, application_entity, application_id, uniq_seq, ocr_res_id,
                                           is_ca, is_cms))
-    # 查看此订单号下是否有未完成的文件，如果有，等?分钟
-    doc_wait_file_class = HILDoc if application_entity == consts.HIL_PREFIX else AFCDoc
-    doc_wait_file_result = doc_wait_file_class.objects.filter(application_id=application_id, status=1).first()
-    compare_log.info('doc_wait_file_result:{0}'.format(doc_wait_file_result))
-    compare_log.info('{0} [comparison unfinished file check] [entity={1}] [id={2}] [doc_wait_file_result={3}]'.format(log_base, application_entity, application_id, doc_wait_file_result))
-    if doc_wait_file_result is not None:
-        # 实时查询延迟时间
-        try:
-            delay_time_config = Configs.objects.filter(id=4).first()
-            if delay_time_config is not None and delay_time_config.value is not None and delay_time_config.value.isdigit():
-                delay_time = delay_time_config.value
-            else:
-                delay_time = 0
-        except Exception as e:
-            delay_time = 0
-            compare_log.info('[get delay_time_config fail] [error={0}]'.format(traceback.format_exc()))
-        compare_log.info('delay_time:{0}'.format(delay_time))
-        compare_log.info('{0} [comparison unfinished file wait delay_time start] [entity={1}] [id={2}] [doc_id={3}]'.format(log_base, application_entity, application_id, doc_wait_file_result.id))
-        try:
-            time.sleep(int(delay_time))
-        except Exception as e:
-            compare_log.info('[sleep error] [error={0}]'.format(traceback.format_exc()))
-        compare_log.info('{0} [comparison unfinished file wait delay_time end] [entity={1}] [id={2}] [doc_id={3}]'.format(log_base, application_entity, application_id, doc_wait_file_result.id))
        # 根据application_id查找最新的比对信息，如果没有，结束
        if is_ca:
            comparison_class = HILComparisonInfo if application_entity == consts.HIL_PREFIX else AFCComparisonInfo
@@ -3925,6 +3914,35 @@ def compare_thread(application_id, application_entity, uniq_seq, ocr_res_id, is_
                                                               uniq_seq, ocr_res_id, is_ca, is_cms))
            return
+        # 查看此订单号下是否有未完成的文件，如果有，等?分钟
+        try:
+            doc_wait_file_class = HILDoc if application_entity == consts.HIL_PREFIX else AFCDoc
+            doc_wait_file_result = doc_wait_file_class.objects.filter(application_id=application_id, status=1).first()
+            compare_log.info('doc_wait_file_result:{0}'.format(doc_wait_file_result))
+            compare_log.info('{0} [comparison unfinished file check] [entity={1}] [id={2}] [doc_wait_file_result={3}]'.format(log_base, application_entity, application_id, doc_wait_file_result))
+        except Exception as e:
+            doc_wait_file_result = None
+            compare_log.info('[get doc_wait_file_result fail] [error={0}]'.format(traceback.format_exc()))
+        if doc_wait_file_result is not None:
+            # 实时查询延迟时间
+            try:
+                delay_time_config = Configs.objects.filter(id=4).first()
+                if delay_time_config is not None and delay_time_config.value is not None and delay_time_config.value.isdigit():
+                    delay_time = delay_time_config.value
+                else:
+                    delay_time = 0
+            except Exception as e:
+                delay_time = 0
+                compare_log.info('[get delay_time_config fail] [error={0}]'.format(traceback.format_exc()))
+            compare_log.info('delay_time:{0}'.format(delay_time))
+            compare_log.info('{0} [comparison unfinished file wait delay_time start] [entity={1}] [id={2}] [doc_id={3}]'.format(log_base, application_entity, application_id, doc_wait_file_result.id))
+            try:
+                time.sleep(int(delay_time))
+            except Exception as e:
+                compare_log.info('[sleep error] [error={0}]'.format(traceback.format_exc()))
+            compare_log.info('{0} [comparison unfinished file wait delay_time end] [entity={1}] [id={2}] [doc_id={3}]'.format(log_base, application_entity, application_id, doc_wait_file_result.id))
        if is_ca:
            ca_compare(application_id, application_entity, ocr_res_id, last_obj, ocr_res_dict)
        else:
@@ -3969,4 +3987,5 @@ def compare_thread(application_id, application_entity, uniq_seq, ocr_res_id, is_
                    compare_log.error('{0} [Auto SE] [result save error] [entity={1}] [id={2}] [ocr_res_id={3}] '
                                      '[error={4}]'.format(log_base, application_entity, application_id, ocr_res_id,
                                                           traceback.format_exc()))
+    except Exception as e:
+        compare_log.info('[compare_thread error] [error={0}]'.format(traceback.format_exc()))
--- a/src/settings/__init__.py
View file @d6fc968
+++ b/src/settings/__init__.py
View file @d6fc968
@@ -102,7 +102,7 @@ for db_setting in DATABASES.values():
        'driver': 'ODBC Driver 17 for SQL Server',
        'extra_params': "odbc_cursortype=2"
    }
+    db_setting['CONN_MAX_AGE'] = 60
 # set this to False if you want to turn off pyodbc's connection pooling
 DATABASE_CONNECTION_POOLING = True