add ic true/false

周伟奇
Showing 3 changed files with 66 additions and 4 deletions
src/apps/doc/consts.py
src/apps/doc/exceptions.py
src/apps/doc/management/commands/ocr_process.py
--- a/src/apps/doc/consts.py
View file @591d1ac
+++ b/src/apps/doc/consts.py
View file @591d1ac
@@ -624,7 +624,9 @@ OTHER_CLASSIFY = 2
 # 身份证
 IC_CN_NAME = '身份证'
 IC_CLASSIFY = 33
-IC_FIELD_ORDER_0 = (('姓名', '姓名'),
+IC_TURE_OR_FALSE = '真伪'
+IC_FIELD_ORDER_0 = ((IC_TURE_OR_FALSE, '身份证'),
+                    ('姓名', '姓名'),
                    ('公民身份号码', '公民身份号码'),
                    ('出生年月', '出生年月'),
                    ('住址', '住址'),
@@ -934,3 +936,10 @@ PATTERN_LIST = ['交易名称', '收入/支出金额', '收入', '存入', '支
                '短摘要', '本次余额', '交易后余额', '交易说明', '帐户余额', '交易日期 记账日期']
 CN_RE = re.compile(u'[\u4e00-\u9fa5]')
+IC_RES_MAPPING = {
+    -2: '不是有效证件',
+    -1: '无法判断',
+    0: '伪造证件',
+    1: '真实证件',
+}
--- a/src/apps/doc/exceptions.py
View file @591d1ac
+++ b/src/apps/doc/exceptions.py
View file @591d1ac
@@ -8,3 +8,7 @@ class OCR1Exception(Exception):
 class OCR2Exception(Exception):
    pass
+class OCR4Exception(Exception):
+    pass
--- a/src/apps/doc/management/commands/ocr_process.py
View file @591d1ac
+++ b/src/apps/doc/management/commands/ocr_process.py
View file @591d1ac
@@ -23,7 +23,7 @@ from common.tools.pdf_to_img import PDFHandler
 from apps.doc import consts
 from apps.doc.ocr.edms import EDMS, rh
 from apps.doc.named_enum import KeywordsType
-from apps.doc.exceptions import EDMSException, OCR1Exception, OCR2Exception
+from apps.doc.exceptions import EDMSException, OCR1Exception, OCR2Exception, OCR4Exception
 from apps.doc.ocr.wb import BSWorkbook, Workbook
 from apps.doc.models import DocStatus, HILDoc, AFCDoc, Keywords
@@ -48,6 +48,7 @@ class Command(BaseCommand, LoggerMixin):
        self.ocr_1_urls = conf.get_namespace('OCR_URL_1_')
        self.ocr_url_2 = conf.OCR_URL_2
        self.ocr_url_3 = conf.BC_URL
+        self.ocr_url_4 = conf.IC_URL
        # EDMS web_service_api
        self.edms = EDMS()
        # 优雅退出信号：15
@@ -189,13 +190,60 @@ class Command(BaseCommand, LoggerMixin):
        else:
            res_list.append((pno, ino, part_idx, consts.RES_SUCCESS_EMPTY))
-    def license1_process(self, ocr_data, license_summary, classify, res_list, pno, ino, part_idx):
+    def license1_process(self, ocr_data, license_summary, classify, res_list, pno, ino, part_idx, img_path):
        # 类别：'0'身份证， '1'居住证
        license_data = ocr_data.get('data', [])
        if not license_data:
            res_list.append((pno, ino, part_idx, consts.RES_SUCCESS_EMPTY))
            return
        res_list.append((pno, ino, part_idx, consts.RES_SUCCESS))
+        if classify == consts.IC_CLASSIFY:
+            for id_card_dict in license_data:
+                try:
+                    base64_img = id_card_dict.pop('base64_img')
+                except Exception as e:
+                    continue
+                else:
+                    card_type = -1
+                    json_data_4 = {
+                        'mode': 1,
+                        'user_info': {
+                            'image_content': base64_img,
+                        },
+                        'options': {
+                            'distinguish_type': 1,
+                            'auto_rotate': True,
+                        },
+                    }
+                    for times in range(consts.RETRY_TIMES):
+                        try:
+                            start_time = time.time()
+                            ocr_4_response = requests.post(self.ocr_url_4, json=json_data_4)
+                            if ocr_4_response.status_code != 200:
+                                raise OCR4Exception('ocr_4 status code: {0}'.format(ocr_4_response.status_code))
+                        except Exception as e:
+                            self.cronjob_log.warn(
+                                '{0} [ocr_4 failed] [times={1}] [img_path={2}] [error={3}]'.format(
+                                    self.log_base, times, img_path, traceback.format_exc()))
+                        else:
+                            ocr_4_res = json.loads(ocr_4_response.json())
+                            end_time = time.time()
+                            speed_time = int(end_time - start_time)
+                            if ocr_4_res.get('code') == 0 and ocr_4_res.get('result', {}).get('rtn') == 0:
+                                card_type = ocr_4_res.get('result', {}).get(
+                                    'idcard_distinguish_result', {}).get('result', -1)
+                            self.cronjob_log.info(
+                                '{0} [ocr_4 success] [img_path={1}] [speed_time={2}]'.format(
+                                    self.log_base, img_path, speed_time))
+                            break
+                    else:
+                        self.cronjob_log.warn(
+                            '{0} [ocr_4 failed] [img_path={1}]'.format(self.log_base, img_path))
+                    id_card_dict[consts.IC_TURE_OR_FALSE] = consts.IC_RES_MAPPING.get(card_type)
        license_summary.setdefault(classify, []).extend(license_data)
    def license2_process(self, ocr_res_2, license_summary, pid, classify, res_list, pno, ino, part_idx):
@@ -559,7 +607,8 @@ class Command(BaseCommand, LoggerMixin):
                                        res_list.append((pno, ino, part_idx, consts.RES_SUCCESS_OTHER))
                                        continue
                                    elif classify in consts.LICENSE_CLASSIFY_SET_1:  # 证件1
-                                        self.license1_process(ocr_data, license_summary, classify, res_list, pno, ino, part_idx)
+                                        self.license1_process(ocr_data, license_summary, classify, res_list,
+                                                              pno, ino, part_idx, img_path)
                                    elif classify in consts.LICENSE_CLASSIFY_SET_2:  # 证件2
                                        pid, _, _, _, _, _ = consts.LICENSE_CLASSIFY_MAPPING.get(classify)
                                        file_data = ocr_data.get('section_img')