测试匹配未能识别的型号

config.py

@@ -16,6 +16,8 @@ else:
     redis_config = {'host': '127.0.0.1', 'port': 6379, 'decode_responses': True, 'password': 'icDb29mLy2s'}
     # redis_config = {'host': '192.168.1.235', 'port': 6379, 'decode_responses': True, 'password': 'icDb29mLy2s'} #部署在252时
 
+# brand_redis_config = {'host': '39.108.51.147', 'port': 6379, 'decode_responses': True, 'password': 'icDb29mLy2s'}
+
 MG_HOST_SET = {
     'test': '192.168.1.237',
     'produce': '172.18.137.23'

utils/functions.py

@@ -14,6 +14,7 @@ num_pattern = re.compile('(\s|[\u4e00-\u9fa5])(\d+\.\d+|\d+)(\s|个|片)')
 k_num_pattern = re.compile('(\s|[\u4e00-\u9fa5])(\d+\.\d+|\d+)(K)', re.I)
 pcs_num_pattern = re.compile('(\s|[\u4e00-\u9fa5])(\d+\.\d+|\d+)(pcs)', re.I)
 price_num_pattern = re.compile('(\d+\.\d+|\d+)')
+price2_num_pattern = re.compile('([.a-zA-Z\\-]?)(\d+\.\d+|\d+)([.a-zA-Z\\-]?)')
 zh_pattern = re.compile('[\u4e00-\u9fa5]+')
 usd_num_pattern = re.compile('(\d+\.\d+|\d+)(usd)', re.I)
 time_pattern = re.compile(
@@ -425,9 +426,9 @@ def check_price(kw_info):
         if res:
             return res[0][0].strip()
         else:
-             res = price_num_pattern.findall(kw_info)
-             if res and "+" not in kw_info:
-                 return res[0].strip()
+             res = price2_num_pattern.findall(kw_info)
+             if res and "+" not in kw_info and not res[0] and not res[2]:
+                 return res[1].strip()
              else:
                  return False
 

utils/redis_cli.py

@@ -5,3 +5,4 @@ import redis
 from config import redis_config
 
 redis_cli = redis.Redis(**redis_config)
+# brand_redis_cli = redis.Redis(**brand_redis_config)

valid_server.py

@@ -138,6 +138,7 @@ class ImHandler(tornado.web.RequestHandler):
         super().__init__(*args, **kwargs)
         self.predic = KwPredict('validSingle')
         self.redis_cli = redis_cli
+        # self.brand_redis_cli = brand_redis_cli #暂未启用
 
         self.num_pattern = re.compile('')  # 单位匹配
 
@@ -150,7 +151,7 @@ class ImHandler(tornado.web.RequestHandler):
             code = '100001'
             res = code2msg(code)
         else:
-            brand_set = redis_cli.smembers('brand_set')
+            brand_set = self.redis_cli.smembers('brand_set')
             data = []
             for data_info in data_list:
                 data_obj = {}
@@ -163,7 +164,7 @@ class ImHandler(tornado.web.RequestHandler):
 
                 new_data_info = []
                 # print(data_info.decode("gbk").encode("utf-8"))
-                data_info = replace_symbol(data_info).replace("（", "(").replace("）", ")").replace("\xa0","").replace("\t"," ")  # 统一分隔符,替换中文括号
+                data_info = replace_symbol(data_info).replace("（", "(").replace("）", ")").replace("\xa0", "").replace("\t", " ")  # 统一分隔符,替换中文括号
                 data_cut_list = data_info.split(" ")
                 new_cut_info = ""
                 for cut_info in data_cut_list:  # 第一遍过滤
@@ -260,7 +261,7 @@ class ImHandler(tornado.web.RequestHandler):
                     for end_cut_info in end_cut_list:  # 将剩下的未知单词剔除包含中文的单词
                         if end_cut_info.strip() and check_zh(end_cut_info.strip()) and len(end_cut_info.strip()) > 4:
                             len_lsit.append(end_cut_info)
-                    if len_lsit and len(len_lsit) == 1 and not data_obj["gn"]:  # 若剔除完中文后，此时型号还未空，则剩下的词很大概率是型号，暂时按时型号处理
+                    if len_lsit and len(len_lsit) == 1 and not data_obj["gn"]:  # 若剔除完中文后，此时型号还为空，则剩下的词很大概率是型号，暂时按时型号处理
                         data_obj["gn"].append(len_lsit[0].strip())
                 if data_obj:
                     data.append(data_obj)