add

search_tips/settings.py

@@ -28,7 +28,7 @@ SECRET_KEY = 'e%$v6snev0807=t0@gk_n2#r5m6r1h(eil6cp^y3ub@ja@gk_t'
 # SECURITY WARNING: don't run with debug turned on in production!
 DEBUG = True
 
-ALLOWED_HOSTS = []
+ALLOWED_HOSTS = ["192.168.78.2"]
 
 SENTRY_CELERY_ENDPOINT="http://60b0004c8884420f8067fb32fc3ed244:20f97fc73ffa4aad9735d0e6542a6d78@sentry.igengmei.com/140"
 BROKER_URL = "redis://127.0.0.1:6379/8"

trans2es/management/commands/trans2es_tipsdata2es.py

+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+import django.db.models
+from django.conf import settings
+from django.core.management.base import BaseCommand, CommandError
+import traceback
+import logging
+import six
+import sys
+
+from libs.es import ESPerform
+import trans2es.models as md
+from libs.table_scan import TableSlicer,TableSlicerChunk
+from trans2es.type_info import get_type_info_map,TypeInfo
+
+from libs.cache import redis_client
+import json
+
+
+
+
+class Command(BaseCommand):
+
+    args = ''
+    help = 'dump data to elasticsearch, parallel'
+
+    from optparse import make_option
+
+    option_list = BaseCommand.option_list + (
+        make_option('-t', '--type', dest='type', help='type name to dump data to elasticsearch', metavar='TYPE',default=''),
+        make_option('-i', '--index-prefix', dest='index_prefix', help='index name to dump data to elasticsearch', metavar='INDEX_PREFIX'),
+        make_option('-p', '--parallel', dest='parallel', help='parallel process count', metavar='PARALLEL'),
+        make_option('-s', '--pks', dest='pks', help='specify sync pks, comma separated', metavar='PKS', default=''),
+        make_option('--streaming-slicing', dest='streaming_slicing', action='store_true', default=True),
+        make_option('--no-streaming-slicing', dest='streaming_slicing', action='store_false', default=True),
+        make_option('-S', '--sync_type',dest='sync_type', help='sync data to es',metavar='TYPE',default='')
+    )
+
+    def handle(self, *args, **options):
+        try:
+            logging.info("test")
+        except:
+            logging.error("catch exception,err_msg:%s" % traceback.format_exc())

trans2es/mapping/suggest.json

@@ -6,7 +6,7 @@
       "type":"completion"
     },
     "suggest_type":{"type":"long"},//0-汉字,1-汉字全拼,2-拼音,3-拼音全拼,4-拼音简写,5-拼音简写全拼
-    "data_type":{"type":"long"},//0-tag,1-hospital,2-doctor,3-wiki
+    "tips_name_type":{"type":"long"},//tips数据所属类型,0-tag,1-hospital,2-doctor,3-wiki
     "ori_name":{"type":"keyword"},//原名称
     "order_weight":{"type":"double"},//订单权重
     "offline_score":{"type":"double"},//离线分

trans2es/utils/doctor_transfer.py

@@ -19,6 +19,35 @@ class DoctorTransfer(object):
     ch_prefix_weight = 1.5
     py_prefix_weight = 1.0
 
+    # 命中开始部分加权
+    begin_prefix_weight = 1.2
+
+    @classmethod
+    def get_doctor_data_name_mapping_results_to_redis(cls, instance):
+        try:
+            doctor_name = instance.name
+            if doctor_name.strip():
+                multi_fields = {
+                    'hospital.city_province_name': 1,
+                    'hospital.name': 1,
+                    'name': 1,
+                    'hospital.city_name':1,
+                    'hospital.officer_name':1,
+                    'service_closure_tags':1
+                }
+                query_fields = ['^'.join((k, str(v))) for (k, v) in multi_fields.items()]
+                multi_match = {
+                    'query': doctor_name,
+                    'type': 'cross_fields',
+                    'operator': 'and',
+                    'fields': query_fields,
+                }
+
+        except:
+            logging.error("catch exception,err_msg:%s" % traceback.format_exc())
+
+
+
     @classmethod
     def get_doctor_suggest_data_list(cls,instance):
         try:
@@ -33,7 +62,7 @@ class DoctorTransfer(object):
             item_dict["results_num"] = 0
             item_dict["type_flag"] = "unknown"
             item_dict["offline_score"] = 0.0
-            item_dict["data_type"] = 2
+            item_dict["tips_name_type"] = 2
 
             ret_list.append(item_dict)
             suggest_list = list()
@@ -50,9 +79,10 @@ class DoctorTransfer(object):
                     if ch_name_term:
                         prefix_weight = cls.ch_prefix_weight if len(ch_name_term)!=len(ch_full_word) else cls.ch_full_weight
                         suggest_type = 0 if len(ch_name_term)!=len(ch_full_word) else 1
+                        begin_prefix_weight = cls.begin_prefix_weight if i==0 else 1.0
                         suggest_item = {
                             "input":[ch_name_term],
-                            "word_weight":(1.0*len(ch_name_term)/len((ch_full_word))) * prefix_weight,
+                            "word_weight":(1.0*len(ch_name_term)/len((ch_full_word))) * prefix_weight * begin_prefix_weight,
                             "suggest_type": suggest_type,
                             "cur_index": cur_index
                         }
@@ -66,9 +96,10 @@ class DoctorTransfer(object):
                     if py_name_term:
                         prefix_weight = cls.py_prefix_weight if len(py_name_term)!=len(py_full_word) else cls.py_full_weight
                         suggest_type = 2 if len(py_name_term)!=len(py_full_word) else 3
+                        begin_prefix_weight = cls.begin_prefix_weight if i==0 else 1.0
                         suggest_item = {
                             "input":[py_name_term],
-                            "word_weight":(1.0*len(py_name_term)/len(py_full_word)) * prefix_weight,
+                            "word_weight":(1.0*len(py_name_term)/len(py_full_word)) * prefix_weight * begin_prefix_weight,
                             "suggest_type": suggest_type,
                             "cur_index": cur_index
                         }
@@ -82,9 +113,10 @@ class DoctorTransfer(object):
                     if py_acronym_term:
                         prefix_weight = cls.py_acronym_prefix_weight if len(py_acronym_term)!=len(py_acronym_full_word) else cls.py_acronym_full_weight
                         suggest_type = 4 if len(py_acronym_term)!=len(py_acronym_full_word) else 5
+                        begin_prefix_weight = cls.begin_prefix_weight if i==0 else 1.0
                         suggest_item = {
                             "input":[py_acronym_term],
-                            "word_weight":(1.0*len(py_acronym_term)/len(py_acronym_full_word)) * prefix_weight,
+                            "word_weight":(1.0*len(py_acronym_term)/len(py_acronym_full_word)) * prefix_weight * begin_prefix_weight,
                             "suggest_type": suggest_type,
                             "cur_index": cur_index
                         }