Merge branch 'del_weight' into test

libs/tools.py

@@ -58,26 +58,26 @@ def point_distance(lon1, lat1, lon2, lat2):
     # haversine公式
     dlon = lon2 - lon1
     dlat = lat2 - lat1
-    a = sin(dlat / 2)**2 + cos(lat1) * cos(lat2) * sin(dlon / 2)**2
+    a = sin(dlat / 2) ** 2 + cos(lat1) * cos(lat2) * sin(dlon / 2) ** 2
     c = 2 * asin(sqrt(a))
     r = 6371  # 地球平均半径，单位为km
     return c * r * 1000
 
 
 class HostpitalPosInfo(object):
-
     hospital_pos_dict = dict()
+
     @classmethod
     def get_hostpital_info(cls):
         try:
             if len(cls.hospital_pos_dict) == 0:
-                query_results = doctor.Hospital.objects.filter(is_online=True)
-                #query_results = django.db.models.QuerySet(model=doctor.Hospital, query=query)
+                query_results = doctor.Hospital.objects.filter()
+                # query_results = django.db.models.QuerySet(model=doctor.Hospital, query=query)
                 for item in query_results:
                     name = item.name
                     lng = float(item.baidu_loc_lng)
                     lat = float(item.baidu_loc_lat)
-                    cls.hospital_pos_dict[name] = [lng,lat]
+                    cls.hospital_pos_dict[name] = [lng, lat]
 
             return cls.hospital_pos_dict
         except:
@@ -85,4 +85,4 @@ class HostpitalPosInfo(object):
             return cls.hospital_pos_dict
 
 
-g_hospital_pos_dict = HostpitalPosInfo.get_hostpital_info()
\ No newline at end of file
+g_hospital_pos_dict = HostpitalPosInfo.get_hostpital_info()

search/utils/auto_tips.py

@@ -31,14 +31,13 @@ def get_suggest_tips(query, lat, lng, offset=0, size=50, device_id=None):
     """
 
     try:
+        gray_number = 1
         ###在原来的逻辑上加两层灰度
-        gray_number = recommed_service_category_device_id(device_id)
-        if gray_number == 1:
-            sub_index_name = "suggest"
-            size = 50
-        else:
+        if device_id in ["057F28DF-20B8-488F-A285-931367FCC110"]:
             sub_index_name = "suggest-v1"
-            size = 30
+
+        else:
+            sub_index_name = "suggest"
 
         have_read_tips_set, ret_list, result_dict = get_query_by_es(query=str(query).lower(), lat=lat, lng=lng,
                                                                     offset=offset,
@@ -210,7 +209,8 @@ def get_query_by_es(query='', lat=0, lng=0, size=0, offset=0, highlight_query=No
 
         for tips_item in result_dict["suggest"]["tips-suggest"]:
             for hit_item in tips_item["options"]:
-
+                hit_item["_source"]["ori_name"] = hit_item["_source"]["ori_name"].replace("超声刀","超声提升")
+                hit_item["_source"]["ori_name"] = hit_item["_source"]["ori_name"].replace("私密超声刀", "私密超声紧致")
                 if hit_item["_source"]["ori_name"] not in have_read_tips_set:
                     have_read_tips_set.add(hit_item["_source"]["ori_name"])
                     highlight_marks = u'<ems>%s</ems>' % query

search/view/auto_tips.py

@@ -13,7 +13,7 @@ from search.utils.auto_tips import get_suggest_tips
 
 
 @bind("search_tips/search/auto_complete_query")
-def auto_complete_query(q, lat, lng, device_id=None):
+def auto_complete_query(q="", lat=0, lng=0, device_id=None):
     try:
         """auto complate words/tags/doctors etc.
 

trans2es/commons/words_utils.py

 # coding=utf8
 from __future__ import unicode_literals, absolute_import, print_function
-
 import logging
 import traceback
 import json
 from libs.cache import redis_client
 from trans2es.models.query_word_conversion import TagConversion, DoctorConversion, HospitalConversion
 from trans2es.models.strategy_search_word import StrategyHistoryQueryWords
+from libs.cache import redis_client
+import base64
+from gm_types.doris import MIND_TYPE
+from gm_rpcd.all import bind
+
+QUERY_KEY = "query:{}:set"
+
+LABEL_VALUE = {
+    MIND_TYPE.PROJECT: 8,
+    MIND_TYPE.HOSPITAL: 7,
+    MIND_TYPE.DOCTOR: 6,
+    MIND_TYPE.FREE_FACE: 4,
+    MIND_TYPE.FACE: 3,
+    MIND_TYPE.AREA: 2,
+    MIND_TYPE.USER: 1,
+    MIND_TYPE.UNKNOWN: 0
+}
+
+QUERY_WORD_LABEL_NEED_MODIFIED = {
+    u"玻尿酸": MIND_TYPE.PROJECT,
+    u"鼻": MIND_TYPE.PROJECT,
+    u"眼": MIND_TYPE.PROJECT,
+    u"嘴": MIND_TYPE.PROJECT,
+    u"脱毛": MIND_TYPE.PROJECT
+}
 
 
 class TagTab(object):
@@ -99,6 +123,18 @@ class QueryWordAttr(object):
 
     @classmethod
     def get_hot_search_query_word_weight(cls, name):
+        try:
+            search_info = StrategyHistoryQueryWords.objects.filter(keyword=name).first()
+            weight = int(0.0001 * (
+                    search_info.search_num * 0.8 + 0.2 * (search_info.answer_num + search_info.tractate_num + search_info.diary_num)))
+
+            return weight
+        except:
+            logging.error("catch exception,err_msg:%s" % traceback.format_exc())
+            return 0.0
+
+    @classmethod
+    def get_hot_search_query_word_weight_v1(cls, name):
         try:
             search_num = StrategyHistoryQueryWords.objects.filter(keyword=name).values_list(
                 "search_num", flat=True).first()
@@ -129,12 +165,13 @@ class QueryWordAttr(object):
             return 0.0
 
     @classmethod
-    def get_doctor_query_word_weight(cls, name):
+    def get_doctor_query_word_weight(cls, name=None, doctor_id=None):
         try:
             if not cls.doctor_latest_date:
                 cls.doctor_latest_date = DoctorConversion.objects.latest('update_date').update_date
 
-            doc_query_results = DoctorConversion.objects.filter(query=name, update_date=cls.doctor_latest_date)
+            doc_query_results = DoctorConversion.objects.filter(business_id=doctor_id,
+                                                                update_date=cls.doctor_latest_date)
 
             total_score = 0.0
             for query_item in doc_query_results:
@@ -147,12 +184,13 @@ class QueryWordAttr(object):
             return 0.0
 
     @classmethod
-    def get_hospital_query_word_weight(cls, name):
+    def get_hospital_query_word_weight(cls, name=None, hospital_id=None):
         try:
             if not cls.hospital_latest_date:
                 cls.hospital_latest_date = HospitalConversion.objects.latest('update_date').update_date
 
-            hospital_query_results = HospitalConversion.objects.filter(query=name, update_date=cls.hospital_latest_date)
+            hospital_query_results = HospitalConversion.objects.filter(business_id=hospital_id,
+                                                                       update_date=cls.hospital_latest_date)
 
             total_score = 0.0
             for query_item in hospital_query_results:
@@ -186,6 +224,7 @@ class QueryWordAttr(object):
             if data:
                 for value in data.values():
                     total_num += value
+
             return total_num
         except:
             logging.error("catch exception,err_msg:%s" % traceback.format_exc())
@@ -252,33 +291,6 @@ class QueryWordAttr(object):
             return False
 
 
-from gm_rpcd.all import bind
-from libs.cache import redis_client
-import base64
-from gm_types.doris import MIND_TYPE
-
-QUERY_KEY = "query:{}:set"
-
-LABEL_VALUE = {
-    MIND_TYPE.PROJECT: 8,
-    MIND_TYPE.HOSPITAL: 7,
-    MIND_TYPE.DOCTOR: 6,
-    MIND_TYPE.FREE_FACE: 4,
-    MIND_TYPE.FACE: 3,
-    MIND_TYPE.AREA: 2,
-    MIND_TYPE.USER: 1,
-    MIND_TYPE.UNKNOWN: 0
-}
-
-QUERY_WORD_LABEL_NEED_MODIFIED = {
-    u"玻尿酸": MIND_TYPE.PROJECT,
-    u"鼻": MIND_TYPE.PROJECT,
-    u"眼": MIND_TYPE.PROJECT,
-    u"嘴": MIND_TYPE.PROJECT,
-    u"脱毛": MIND_TYPE.PROJECT
-}
-
-
 def label_key(label):
     return LABEL_VALUE.get(label)
 

trans2es/utils/doctor_transfer.py

@@ -94,7 +94,8 @@ class DoctorTransfer(object):
                 item_dict["id"] = getMd5Digest(str(instance.name))
                 item_dict["ori_name"] = instance.name
                 item_dict["is_online"] = instance.is_online
-                item_dict["order_weight"] = QueryWordAttr.get_doctor_query_word_weight(instance.name)
+                item_dict["order_weight"] = QueryWordAttr.get_doctor_query_word_weight(name=instance.name,
+                                                                                       doctor_id=instance.id)
                 item_dict["results_num"] = QueryWordAttr.get_query_results_num(instance.name)
                 item_dict["type_flag"] = get_tips_word_type(str(instance.name).lower())
                 item_dict["offline_score"] = 0.0
@@ -190,7 +191,8 @@ class DoctorTransferV1(object):
                 item_dict["id"] = getMd5Digest(str(instance.name))
                 item_dict["ori_name"] = instance.name
                 item_dict["is_online"] = instance.is_online
-                item_dict["order_weight"] = QueryWordAttr.get_doctor_query_word_weight(instance.name)
+                item_dict["order_weight"] = QueryWordAttr.get_doctor_query_word_weight(name=instance.name,
+                                                                                       doctor_id=instance.id)
                 item_dict["results_num"] = QueryWordAttr.get_query_results_num(instance.name)
                 item_dict["type_flag"] = get_tips_word_type(instance.name)
                 item_dict["offline_score"] = 0.0
@@ -201,5 +203,3 @@ class DoctorTransferV1(object):
         except:
             logging.error("catch exception,err_msg:%s" % traceback.format_exc())
             return (dict(), list())
-
-

trans2es/utils/hospital_transfer.py

@@ -20,11 +20,8 @@ class HospitalTransfer(object):
         try:
             ret_list = list()
             name = instance.name.strip()
-            sub_query = re.sub('\W+', '', name)
-            name = sub_query
             cut_bool = False
             cut_word = ["下线", "停用", "已经下线", "账号停用"]
-
             for i in cut_word:
                 if name.find(i) >= 0:
                     cut_bool = True
@@ -36,9 +33,9 @@ class HospitalTransfer(object):
                 item_dict["id"] = getMd5Digest(str(name))
                 item_dict["ori_name"] = name
                 item_dict["is_online"] = by_hospital_id_judge_office_online(is_online=instance.is_online,
-                                                                            hospital_id=instance.id)
-
-                item_dict["order_weight"] = QueryWordAttr.get_hospital_query_word_weight(name)
+                                                                            hospital_id=instance.id, name=name)
+                item_dict["order_weight"] = QueryWordAttr.get_hospital_query_word_weight(name=name,
+                                                                                         hospital_id=instance.id)
                 item_dict["results_num"] = QueryWordAttr.get_query_results_num(name)
                 item_dict["type_flag"] = get_tips_word_type(name)
                 item_dict["offline_score"] = 0.0
@@ -73,7 +70,6 @@ class HospitalTransferV1(object):
     def get_hospital_suggest_data_list(cls, instance):
         try:
             ret_list = list()
-
             name = instance.name.strip()
             cut_bool = False
             cut_word = ["下线", "停用", "已经下线", "账号停用"]
@@ -88,9 +84,10 @@ class HospitalTransferV1(object):
                 item_dict["id"] = getMd5Digest(str(name))
                 item_dict["ori_name"] = name
                 item_dict["is_online"] = by_hospital_id_judge_office_online(is_online=instance.is_online,
-                                                                            hospital_id=instance.id)
+                                                                            hospital_id=instance.id, name=name)
 
-                item_dict["order_weight"] = QueryWordAttr.get_hospital_query_word_weight(name)
+                item_dict["order_weight"] = QueryWordAttr.get_hospital_query_word_weight(name=name,
+                                                                                         hospital_id=instance.id)
                 item_dict["results_num"] = QueryWordAttr.get_query_results_num(name)
                 item_dict["type_flag"] = get_tips_word_type(name)
                 item_dict["offline_score"] = 0.0
@@ -119,10 +116,10 @@ class HospitalTransferV1(object):
             logging.error("catch exception,err_msg:%s" % traceback.format_exc())
 
 
-def by_hospital_id_judge_office_online(is_online=False, hospital_id=None):
+def by_hospital_id_judge_office_online(is_online=False, hospital_id=None, name=None):
     try:
-        office = Doctor.objects.filter(hospital_id=hospital_id, is_online=True, doctor_type=1)
-        if office and is_online:
+        office = Doctor.objects.filter(is_online=True, doctor_type=1, hospital_id=hospital_id)
+        if office:
             return True
         return False
     except:

trans2es/utils/tag_transfer.py

@@ -70,23 +70,91 @@ class TagTransfer(object):
                                                            sub_index_name="service", doc_type="service", query_body=q,
                                                            offset=0, size=0)
 
-                doctor_results = result_dict["total_count"]
+                service_results = result_dict["total_count"]
+
+                if service_results == 0:
+                    service_results = cls.get_content_results_to_redis(instance=instance)
 
                 redis_data = redis_client.hget(cls.tips_num_redis_key_prefix, tag_name)
 
                 redis_val_dict = json.loads(str(redis_data, encoding="utf-8")) if redis_data else {}
 
-                total_count = doctor_results
+                total_count = service_results
                 if 't' in redis_val_dict:
                     total_count += int(redis_val_dict['t'])
 
                 redis_val_dict['t'] = total_count
-                redis_val_dict['s'] = doctor_results
+                redis_val_dict['s'] = service_results
 
                 redis_client.hset(cls.tips_num_redis_key_prefix, tag_name, json.dumps(redis_val_dict))
         except:
             logging.error("catch exception,err_msg:%s" % traceback.format_exc())
 
+    @classmethod
+    def get_content_results_to_redis(cls, instance):
+        try:
+            word = instance.name.strip()
+            answer_query = {
+                "query": {
+                    "bool": {
+                        "minimum_should_match": 1,
+                        "should": [{"match_phrase": {"title": {"query": word, "analyzer": "gm_default_index"}}},
+                                   {"match_phrase": {"desc": {"query": word, "analyzer": "gm_default_index"}}},
+                                   {"match_phrase": {"answer": {"query": word, "analyzer": "gm_default_index"}}}],
+                        "must": [{"term": {"is_online": True}}]
+                    }
+                }
+            }
+            answer_result_dict = ESPerform.get_search_results(ESPerform.get_cli(settings.GM_ORI_ES_INFO_LIST),
+                                                              sub_index_name="answer", doc_type="answer",
+                                                              query_body=answer_query,
+                                                              offset=0, size=0)
+            answer_results = answer_result_dict["total_count"]
+
+            tractate_query = {
+                "query": {
+                    "bool": {
+                        "minimum_should_match": 1,
+                        "should": [{"match_phrase": {"content": {"query": word, "analyzer": "gm_default_index"}}}, {
+                            "match_phrase": {"tractate_tag_name": {"query": word, "analyzer": "gm_default_index"}}}, {
+                                       "match_phrase": {"tractate_tag_name_content": {"query": word,
+                                                                                      "analyzer": "gm_default_index"}}}],
+                        "must": [{"term": {"is_online": True}}]
+                    }
+                }
+            }
+            tractate_result_dict = ESPerform.get_search_results(ESPerform.get_cli(settings.GM_ORI_ES_INFO_LIST),
+                                                                sub_index_name="tractate", doc_type="tractate",
+                                                                query_body=tractate_query,
+                                                                offset=0, size=0)
+            tractate_results = tractate_result_dict["total_count"]
+
+            diary_query = {
+                "query": {
+                    "bool": {
+                        "minimum_should_match": 1,
+                        "should": [{"match_phrase": {"tags": {"query": word, "analyzer": "gm_default_index"}}},
+                                   {"match_phrase": {"answer": {"query": word, "analyzer": "gm_default_index"}}},
+                                   {"match_phrase": {"service.name": {"query": word, "analyzer": "gm_default_index"}}}],
+                        "must": [{"term": {"is_online": True}}, {"range": {"content_level": {"gte": "3"}}}]
+                    }
+                },
+            }
+            diary_result_dict = ESPerform.get_search_results(ESPerform.get_cli(settings.GM_ORI_ES_INFO_LIST),
+                                                             sub_index_name="diary", doc_type="diary",
+                                                             query_body=diary_query,
+                                                             offset=0, size=0)
+            diary_results = diary_result_dict["total_count"]
+
+            total_count = int(answer_results) + int(tractate_results) + int(diary_results)
+
+            logging.info("get content_info:%s" % total_count)
+
+            return total_count
+        except:
+            logging.error("catch exception,err_msg:%s" % traceback.format_exc())
+            return 0
+
     @classmethod
     def get_tag_suggest_data_list(cls, instance):
         try: