Merge pull request #8 from AFEDQK/feat-add-extract-telephone

feat:增加识别固定电话

app/utils/get_parse_data.py

@@ -50,7 +50,12 @@ def format_return_result(res):
     working_place = to_list(res["期望工作地点"])
     # 处理单独省、市、区的情况，以及不正确的地址
     formated_res["期望工作地点"] = postprocess_working_place(working_place)
-    formated_res["招工单位"] = to_list(res["招工单位"])
+    # TODO:拍脑袋定的长度，后期修改
+    recruit_company = to_list(res["招工单位"])
+    if len(recruit_company) > 20:
+        formated_res["招工单位"] = []
+    else:
+        formated_res["招工单位"] = recruit_company
     recruit_infos = res["招工信息"]
     formated_res["联系人"] = to_str(res["联系人"])
     formated_res["联系电话"] = to_list(res["联系电话"])
@@ -121,10 +126,6 @@ def seg_punc(msg):
         if len(result[i]) != 0:
             qa_sentence += result[i]
             qa_sentence += ";"
-        # print(qa_sentence)
-        # if i == len(result) - 1:
-        #     qa_sentence += "。"
-    # print(qa_sentence)
     res['项目内容'] = qa_sentence
     formated_res = format_return_result(res)
     return formated_res
@@ -136,20 +137,14 @@ def find_job(msg):
         return {}
     result = re.split(pattern, msg)
     res, person, city, types, contact = handle_search(msg)
-    # types = "你好"
-    # print(len(types))
     print(result)
     for i in range(len(result)):
-        # print("测试", i, "：", result[i])
         for j in range(len(person)):
             # print(types[j])
             p1 = re.compile(person[j])
             m = p1.findall(result[i])
-            # print("工种匹配：", m)
             if len(m) != 0:
                 result[i] = ""
-                # print("工种匹配成功，以用空格替换")
-                # print("工种为：", types[j])
         for j in range(len(city)):
             p1 = re.compile(city[j])
             m = p1.findall(result[i])

app/utils/process_recruit_detail_info.py

@@ -66,17 +66,16 @@ def getLOCName(str_content):
     return res
 
 
-def getMNumber(str_content):
-    text = str_content
-    lac_result = lac.run(text)
-    resdict = dict(zip(lac_result[0], lac_result[1]))
-    # print(resdict)
-    res = []
-    for item in resdict.items():
-        mystr = str(item[0]).replace('\n', '').replace(' ', '')
-        if (str(item[1]) == 'm' or str(item[1]) == 'n') and len(mystr) == 11 and mystr.isdigit():
-            res.append(mystr)
-    return res
+def getMNumber(content):
+    tel_rules = ["1[3-9]\\d{9}", "0\\d{2,3}-[1-9]\\d{6,8}"]
+    phones = set()
+    for rule in tel_rules:
+        phone = re.findall(rule, content)
+        if not phone:
+            continue
+        for p in phone:
+            phones.add(p)
+    return list(phones)
 
 
 def getAllMsg(content, data_original):
@@ -132,7 +131,7 @@ def getAllMsg(content, data_original):
     money = ''
     acquire = ''
     contact = getMNumber(content)
-    if len(contact) == 0:
+    if not contact:
         contact = getMNumber(data_original.replace('   ', ','))
     # print(contact)
     work_time = ''
@@ -215,6 +214,7 @@ def process_return(types, money, acquire, number, contact, city, work_time, plac
         else:
             one_contact = contact
         one_contact = str(one_contact).replace(']', '').replace('[', '').replace(',', '').replace('\'', '')
+        print(f"place:{place}")
         if len(place) >= len(types) and len(types) > 1 and type(types) == list:
             one_place = place[i]
         else:
@@ -292,17 +292,40 @@ def get_special_content(forward_text, content):
     return re.findall(regex_concat, content)
 
 
+def extract_telephone(data):
+    # 处理 电话号码
+    tel_rules = ["1[3-9]\\d{9}", "0\\d{2,3} [1-9]\\d{3,4} [0-9]\\d{3,4}", "0\\d{2,3}-[1-9]\\d{6,8}",
+                 "1[0-9]{2} [0-9]{4} [0-9]{4}"]
+    all_phones = []
+    for each_rule in tel_rules:
+        phone = re.findall(each_rule, data)
+        if not phone:
+            continue
+        all_phones.extend(phone)
+    # 统一电话的格式
+    res = []
+    for ori_phone in all_phones:
+        if " " not in ori_phone:
+            res.append((ori_phone, " %s " % ori_phone))
+        else:
+            if ori_phone[0] == "0":
+                seg_phone = ori_phone.split(" ")
+                phone_str = " %s-%s " % (seg_phone[0], "".join(seg_phone[1:]))
+                res.append((ori_phone, phone_str))
+            else:
+                phone_str = ori_phone.replace(" ", "")
+                res.append((ori_phone, " %s " % phone_str))
+    return res
+
+
 def extract_info(data, wxid, raw, time):
     # globals().update(regex_config)
     data_original = data
-    # 处理 电话号码人
-    phones = re.findall('[0-9]{11}', data)
-    for p in phones:
-        data = data.replace(p, ' ' + p + ' ')
-    # 处理 189 0074 8140
-    phones = re.findall('[0-9]{3} [0-9]{4} [0-9]{4}', data)
-    for p in phones:
-        data = data.replace(p, ' ' + p.replace(' ', '') + ' ')
+
+    # 提取文本中的电话号码 or 座机号码
+    phones = extract_telephone(data)
+    for ori_phone, format_phone in phones:
+        data = data.replace(ori_phone, format_phone)
 
     # 十人   30到50人 四个 五六个 8九个人 shi ming shiwu ming 一个工管住不管吃 一个礼拜 一个班
     # 算0.5个工 不要暑期工 暑假工不要 物流仓库 28个班 热水空调 汽车玻璃 服务费20 威特电梯 8个通层