Skip to content

S
search_tips
Commit d8dba5ae authored by 李小芳's avatar 李小芳
Browse files
Options

Merge branch 'new_change' into 'master' 

New change

See merge request !47
parents 2d063776 dfff527d
Hide whitespace changes
Inline Side-by-side
Showing with 910 additions and 123 deletions
libs/es.py
View file @ d8dba5ae
......@@ -58,7 +58,7 @@ class ESPerform(object):
os.path.dirname(__file__),
'..', 'trans2es', 'mapping', '%s.json' % (doc_type,))
if doc_type == "associate_tag" or doc_type == "associate_tag_brand" or doc_type == "associate_tag_doctor" \
or doc_type == "associate_tag_instrument" or doc_type == "associate_tag_hospital" or doc_type == "associate_tag_project"\
or doc_type == "associate_tag_instrument" or doc_type == "associate_tag_hospital" or doc_type == "associate_tag_project" \
or doc_type == "associate_tag_position":
mapping_file_path = os.path.join(
os.path.dirname(__file__),
......
search/utils/auto_tips.py
View file @ d8dba5ae
......@@ -5,7 +5,7 @@ from __future__ import unicode_literals, absolute_import, print_function
import logging
import traceback
import json
import base64
import base64, hashlib
from libs.es import ESPerform
from libs.cache import redis_client
from libs.tools import g_hospital_pos_dict
......@@ -15,12 +15,54 @@ from trans2es.commons.words_utils import QueryWordAttr, get_tips_word_type
from pypinyin import pinyin, lazy_pinyin
def get_suggest_tips(query, lat, lng, offset=0, size=50):
def get_suggest_tips(query, lat, lng, offset=0, size=50, device_id=None):
"""
这个地方目前是有三个逻辑 分两个灰度验证
50%前段强加权后根据tab转化率和词频进行排序 索引 suggest gray_number=1 尾号["0", "1", "2", "3", "c", "d", "e", "f"]
25%去掉前段强加权后 直接根据tag转化率和词频进行排序 suggest-v1 gray_number=2 尾号["4", "5", "6", "a"]
25%去掉前段强加权后 根据tag转化率和词频进行排序 再根据前段加权展示排序 suggest-v1 gray_number=3 尾号["9", "8", "7", "b"]
:param query:
:param lat:
:param lng:
:param offset:
:param size:
:param device_id:
:return:
"""
try:
###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"
else:
sub_index_name = "suggest-v1"
have_read_tips_set, ret_list, result_dict = get_query_by_es(query=str(query).lower(), lat=lat, lng=lng,
offset=offset,
size=size, highlight_query=query,
have_read_tips_set=set())
have_read_tips_set=set(),
sub_index_name=sub_index_name)
if gray_number == 3:
# 在去掉强加权的逻辑上根据词频和tag转化率排序后再前段强加权
front_data = []
end_data = []
equal_data = []
need_change_sort = ret_list[:30]
for item in need_change_sort:
ori_name = item.get("ori_name", None)
if query == ori_name:
equal_data.append(item)
elif query == ori_name[:len(query)]:
front_data.append(item)
else:
end_data.append(item)
equal_data.extend(front_data)
equal_data.extend(end_data)
equal_data.extend(ret_list[30:])
ret_list = equal_data
if len(ret_list) >= 50:
return ret_list
......@@ -92,7 +134,8 @@ def get_suggest_tips(query, lat, lng, offset=0, size=50):
offset=offset,
size=size - len(ret_list),
highlight_query=query,
have_read_tips_set=have_read_tips_set)
have_read_tips_set=have_read_tips_set,
sub_index_name=sub_index_name)
ret_list.extend(pinyin_ret_list)
logging.info("get ret_list:%s" % ret_list)
......@@ -123,7 +166,7 @@ def set_highlihgt(query=None, ori_name=None):
def get_query_by_es(query='', lat=0, lng=0, size=0, offset=0, highlight_query=None,
have_read_tips_set=None):
have_read_tips_set=None, sub_index_name="suggest"):
try:
query = query.replace("\u2006", '')
......@@ -152,7 +195,7 @@ def get_query_by_es(query='', lat=0, lng=0, size=0, offset=0, highlight_query=No
ret_list = list()
doctor_hospital_equal_query = list()
tag_equal_query = list()
result_dict = ESPerform.get_search_results(ESPerform.get_cli(), sub_index_name="suggest", query_body=q,
result_dict = ESPerform.get_search_results(ESPerform.get_cli(), sub_index_name=sub_index_name, query_body=q,
offset=offset, size=size, is_suggest_request=True)
for tips_item in result_dict["suggest"]["tips-suggest"]:
......@@ -218,3 +261,36 @@ def get_query_by_es(query='', lat=0, lng=0, size=0, offset=0, highlight_query=No
return have_read_tips_set, ret_list, result_dict
except:
return set(), list(), list()
def recommed_service_category_device_id(device_id):
try:
if device_id == "868771031984211":
return 2
elif device_id == "867961030707277":
return 3
else:
return 1
'''
设备品类显示, 是否命中灰度
'''
# categroy_select_cary1 = ["0", "1", "2", "3", "c", "d", "e", "f"]
# categroy_select_cary2 = ["4", "5", "6", "a"]
# categroy_select_cary3 = ["9", "8", "7", "b"]
#
# if not device_id:
# return 1
#
# hd_id = hashlib.md5(str(device_id).encode()).hexdigest()
# is_gray = hd_id[-1]
#
# if is_gray in categroy_select_cary2:
# return 2
# elif is_gray in categroy_select_cary3:
# return 3
# else:
# return 1
except:
return 1
search/view/auto_tips.py
View file @ d8dba5ae
......@@ -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):
def auto_complete_query(q, lat, lng, device_id=None):
try:
"""auto complate words/tags/doctors etc.
......@@ -28,7 +28,7 @@ def auto_complete_query(q, lat, lng):
# if not q:
# return json_http_response({'error': 0, 'data': []})
data = get_suggest_tips(q, float(lat), float(lng))
data = get_suggest_tips(q, float(lat), float(lng), device_id=device_id)
result = {
'error': 0,
......
search_tips/__init__.pyc
View file @ d8dba5ae
No preview for this file type
trans2es/commons/commons.py
View file @ d8dba5ae
......@@ -115,3 +115,101 @@ def get_tips_suggest_list(instance_cn_name):
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return list()
def get_tips_suggest_list_v1(instance_cn_name):
try:
# ch_full_weight = 6.0 * 1000
# py_full_weight = 3.0 * 1000
full_weight = 1
py_acronym_full_weight = 1
py_acronym_prefix_weight = 1
ch_prefix_weight = 1
py_prefix_weight = 1.0
# 命中开始部分加权
begin_prefix_weight = 1
ch_full_word = instance_cn_name.strip()
py_full_word = ''.join(lazy_pinyin(ch_full_word))
py_acronym_full_word = ''.join(lazy_pinyin(ch_full_word, style=pypinyin.FIRST_LETTER))
suggest_dict = dict()
cur_index = 0
# 中文
for i in range(len(ch_full_word)):
ch_name_term = ch_full_word[i:].strip()
if ch_name_term and ch_full_word[i] != "（" and ch_full_word[i] != "）":
prefix_weight = ch_prefix_weight if len(ch_name_term) != len(ch_full_word) else full_weight
suggest_type = 0 if len(ch_name_term) != len(ch_full_word) else 1
term_begin_prefix_weight = 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 * term_begin_prefix_weight,
"suggest_type": suggest_type
}
if ch_name_term[0] not in suggest_dict:
cur_index += 1
suggest_item["cur_index"] = cur_index
suggest_dict[ch_name_term[0]] = suggest_item
else:
suggest_dict[ch_name_term[0]]["input"].append(ch_name_term)
if suggest_item["word_weight"] > suggest_dict[ch_name_term[0]]["word_weight"]:
suggest_dict[ch_name_term[0]]["word_weight"] = suggest_item["word_weight"]
suggest_dict[ch_name_term[0]]["suggest_type"] = suggest_item["suggest_type"]
# 拼音
if py_full_word != ch_full_word:
for i in range(len(py_full_word)):
py_name_term = py_full_word[i:].strip()
if py_name_term and py_full_word[i] != "（" and py_full_word[i] != "）":
prefix_weight = py_prefix_weight if len(py_name_term) != len(py_full_word) else full_weight
suggest_type = 2 if len(py_name_term) != len(py_full_word) else 3
term_begin_prefix_weight = 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 * term_begin_prefix_weight,
"suggest_type": suggest_type
}
if py_name_term[0] not in suggest_dict:
cur_index += 1
suggest_item["cur_index"] = cur_index
suggest_dict[py_name_term[0]] = suggest_item
else:
suggest_dict[py_name_term[0]]["input"].append(py_name_term)
if suggest_item["word_weight"] > suggest_dict[py_name_term[0]]["word_weight"]:
suggest_dict[py_name_term[0]]["word_weight"] = suggest_item["word_weight"]
suggest_dict[py_name_term[0]]["suggest_type"] = suggest_item["suggest_type"]
# 简写
if py_acronym_full_word != py_full_word:
for i in range(len(py_acronym_full_word)):
py_acronym_term = py_acronym_full_word[i:].strip()
if py_acronym_term and py_acronym_full_word[i] != "（" and py_acronym_full_word[i] != "）":
prefix_weight = py_acronym_prefix_weight if len(py_acronym_term) != len(
py_acronym_full_word) else py_acronym_full_weight
suggest_type = 4 if len(py_acronym_term) != len(py_acronym_full_word) else 5
term_begin_prefix_weight = 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 * term_begin_prefix_weight,
"suggest_type": suggest_type
}
if py_acronym_term[0] not in suggest_dict:
cur_index += 1
suggest_item["cur_index"] = cur_index
suggest_dict[py_acronym_term[0]] = suggest_item
else:
suggest_dict[py_acronym_term[0]]["input"].append(py_acronym_term)
if suggest_item["word_weight"] > suggest_dict[py_acronym_term[0]]["word_weight"]:
suggest_dict[py_acronym_term[0]]["word_weight"] = suggest_item["word_weight"]
suggest_dict[py_acronym_term[0]]["suggest_type"] = suggest_item["suggest_type"]
return suggest_dict.values()
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return list()
trans2es/models/wordresemble.py
View file @ d8dba5ae
......@@ -29,9 +29,10 @@ class WordRel(models.Model):
class WordRelResemble(models.Model):
"""近义词数据表
"""
class Meta:
app_label = 'api'
db_table = 'api_wordrelresemble'
wordrel = models.ForeignKey(WordRel, related_name='all_resembles')
word = models.CharField(u'近义词', max_length=50, db_index=True)
\ No newline at end of file
word = models.CharField(u'近义词', max_length=50, db_index=True)
trans2es/type_info.py
View file @ d8dba5ae
......@@ -14,14 +14,14 @@ import sys
import copy
from trans2es.models import doctor, itemwiki, collectwiki, brandwiki, productwiki, tag, wordresemble
from trans2es.utils.doctor_transfer import DoctorTransfer
from trans2es.utils.hospital_transfer import HospitalTransfer
from trans2es.utils.itemwiki_transfer import ItemWikiTransfer
from trans2es.utils.collectwiki_transfer import CollectWikiTransfer
from trans2es.utils.brandwiki_transfer import BrandWikiTransfer
from trans2es.utils.productwiki_transfer import ProduceWikiTransfer
from trans2es.utils.tag_transfer import TagTransfer
from trans2es.utils.wordresemble import WordResemble
from trans2es.utils.doctor_transfer import DoctorTransfer, DoctorTransferV1
from trans2es.utils.hospital_transfer import HospitalTransfer, HospitalTransferV1
from trans2es.utils.itemwiki_transfer import ItemWikiTransfer, ItemWikiTransferV1
from trans2es.utils.collectwiki_transfer import CollectWikiTransfer, CollectWikiTransferV1
from trans2es.utils.brandwiki_transfer import BrandWikiTransfer, BrandWikiTransferV1
from trans2es.utils.productwiki_transfer import ProduceWikiTransfer, ProduceWikiTransferV1
from trans2es.utils.tag_transfer import TagTransfer, TagTransferV1
from trans2es.utils.wordresemble import WordResemble, WordResembleV1
from libs.es import ESPerform
from libs.tools import tzlc, getMd5Digest
from trans2es.commons.words_utils import QueryWordAttr
......@@ -286,7 +286,8 @@ def get_type_info_map():
name='suggest',
type='doctor_tips', # doctor
model=doctor.Doctor,
query_deferred=lambda: doctor.Doctor.objects.all().filter(doctor_type__in=[DOCTOR_TYPE.DOCTOR,DOCTOR_TYPE.OFFICER]).query,
query_deferred=lambda: doctor.Doctor.objects.all().filter(
doctor_type__in=[DOCTOR_TYPE.DOCTOR, DOCTOR_TYPE.OFFICER]).query,
get_data_func=DoctorTransfer.get_doctor_suggest_data_list,
bulk_insert_chunk_size=100,
round_insert_chunk_size=5,
......@@ -365,89 +366,90 @@ def get_type_info_map():
round_insert_chunk_size=5,
round_insert_period=2,
),
# TypeInfo(
# name='suggest-v1',
# type='doctor_tips', # doctor
# model=doctor.Doctor,
# query_deferred=lambda: doctor.Doctor.objects.all().filter(doctor_type=DOCTOR_TYPE.DOCTOR).query,
# get_data_func=DoctorTransfer.get_doctor_suggest_data_list,
# bulk_insert_chunk_size=100,
# round_insert_chunk_size=5,
# round_insert_period=2,
# ),
# TypeInfo(
# name='suggest-v1',
# type='hospital_tips', # hospital
# model=doctor.Hospital,
# # query_deferred=lambda: doctor.Doctor.objects.all().filter(doctor_type=DOCTOR_TYPE.OFFICER).query,
# query_deferred=lambda: doctor.Hospital.objects.all().query,
# get_data_func=HospitalTransfer.get_hospital_suggest_data_list,
# bulk_insert_chunk_size=100,
# round_insert_chunk_size=5,
# round_insert_period=2,
# ),
# TypeInfo(
# name='suggest-v1',
# type='itemwiki_tips', # itemwiki
# model=itemwiki.ItemWiki,
# query_deferred=lambda: itemwiki.ItemWiki.objects.all().query,
# get_data_func=ItemWikiTransfer.get_itemwiki_suggest_data_list,
# bulk_insert_chunk_size=100,
# round_insert_chunk_size=5,
# round_insert_period=2,
# ),
# TypeInfo(
# name='suggest-v1',
# type='collectwiki_tips', # collectwiki
# model=collectwiki.CollectWiki,
# query_deferred=lambda: collectwiki.CollectWiki.objects.all().query,
# get_data_func=CollectWikiTransfer.get_collectwiki_suggest_data_list,
# bulk_insert_chunk_size=100,
# round_insert_chunk_size=5,
# round_insert_period=2,
# ),
# TypeInfo(
# name='suggest-v1',
# type='brandwiki_tips', # brandwiki
# model=brandwiki.BrandWiki,
# query_deferred=lambda: brandwiki.BrandWiki.objects.all().query,
# get_data_func=BrandWikiTransfer.get_brandwiki_suggest_data_list,
# bulk_insert_chunk_size=100,
# round_insert_chunk_size=5,
# round_insert_period=2,
# ),
# TypeInfo(
# name='suggest-v1',
# type='productwiki_tips', # productwiki
# model=productwiki.ProductWiki,
# query_deferred=lambda: productwiki.ProductWiki.objects.all().query,
# get_data_func=ProduceWikiTransfer.get_productwiki_suggest_data_list,
# bulk_insert_chunk_size=100,
# round_insert_chunk_size=5,
# round_insert_period=2,
# ),
# TypeInfo(
# name='suggest-v1',
# type='tag_tips', # tag
# model=tag.Tag,
# query_deferred=lambda: tag.Tag.objects.all().filter(
# tag_type__in=[TAG_TYPE.BODY_PART, TAG_TYPE.BODY_PART_SUB_ITEM, TAG_TYPE.ITEM_WIKI]).query,
# get_data_func=TagTransfer.get_tag_suggest_data_list,
# bulk_insert_chunk_size=100,
# round_insert_chunk_size=5,
# round_insert_period=2,
# ),
# TypeInfo(
# name='suggest-v1',
# type='wordrel_tips', # tag
# model=wordresemble.WordRel,
# query_deferred=lambda: wordresemble.WordRel.objects.filter(
# category__in=[13, 12, 11, 9, 1]).query,
# get_data_func=WordResemble.get_resemble_list,
# bulk_insert_chunk_size=100,
# round_insert_chunk_size=5,
# round_insert_period=2,
# )
TypeInfo(
name='suggest-v1',
type='doctor_tips-v1', # doctor
model=doctor.Doctor,
query_deferred=lambda: doctor.Doctor.objects.all().filter(
doctor_type__in=[DOCTOR_TYPE.DOCTOR, DOCTOR_TYPE.OFFICER]).query,
get_data_func=DoctorTransferV1.get_doctor_suggest_data_list,
bulk_insert_chunk_size=100,
round_insert_chunk_size=5,
round_insert_period=2,
),
TypeInfo(
name='suggest-v1',
type='hospital_tips-v1', # hospital
model=doctor.Hospital,
# query_deferred=lambda: doctor.Doctor.objects.all().filter(doctor_type=DOCTOR_TYPE.OFFICER).query,
query_deferred=lambda: doctor.Hospital.objects.all().query,
get_data_func=HospitalTransferV1.get_hospital_suggest_data_list,
bulk_insert_chunk_size=100,
round_insert_chunk_size=5,
round_insert_period=2,
),
TypeInfo(
name='suggest-v1',
type='itemwiki_tips-v1', # itemwiki
model=itemwiki.ItemWiki,
query_deferred=lambda: itemwiki.ItemWiki.objects.all().query,
get_data_func=ItemWikiTransferV1.get_itemwiki_suggest_data_list,
bulk_insert_chunk_size=100,
round_insert_chunk_size=5,
round_insert_period=2,
),
TypeInfo(
name='suggest-v1',
type='collectwiki_tips-v1', # collectwiki
model=collectwiki.CollectWiki,
query_deferred=lambda: collectwiki.CollectWiki.objects.all().query,
get_data_func=CollectWikiTransferV1.get_collectwiki_suggest_data_list,
bulk_insert_chunk_size=100,
round_insert_chunk_size=5,
round_insert_period=2,
),
TypeInfo(
name='suggest-v1',
type='brandwiki_tips-v1', # brandwiki
model=brandwiki.BrandWiki,
query_deferred=lambda: brandwiki.BrandWiki.objects.all().query,
get_data_func=BrandWikiTransferV1.get_brandwiki_suggest_data_list,
bulk_insert_chunk_size=100,
round_insert_chunk_size=5,
round_insert_period=2,
),
TypeInfo(
name='suggest-v1',
type='productwiki_tips-v1', # productwiki
model=productwiki.ProductWiki,
query_deferred=lambda: productwiki.ProductWiki.objects.all().query,
get_data_func=ProduceWikiTransferV1.get_productwiki_suggest_data_list,
bulk_insert_chunk_size=100,
round_insert_chunk_size=5,
round_insert_period=2,
),
TypeInfo(
name='suggest-v1',
type='tag_tips-v1', # tag
model=tag.Tag,
query_deferred=lambda: tag.Tag.objects.all().filter(
tag_type__in=[TAG_TYPE.BODY_PART, TAG_TYPE.BODY_PART_SUB_ITEM, TAG_TYPE.ITEM_WIKI]).query,
get_data_func=TagTransferV1.get_tag_suggest_data_list,
bulk_insert_chunk_size=100,
round_insert_chunk_size=5,
round_insert_period=2,
),
TypeInfo(
name='suggest-v1',
type='wordrel_tips-v1', # tag
model=wordresemble.WordRel,
query_deferred=lambda: wordresemble.WordRel.objects.filter(
category__in=[13, 12, 11, 9, 1]).query,
get_data_func=WordResembleV1.get_resemble_list,
bulk_insert_chunk_size=100,
round_insert_chunk_size=5,
round_insert_period=2,
)
]
type_info_map = {
......@@ -550,7 +552,7 @@ def tips_attr_sync_to_redis_type_info_map():
TypeInfo(
name='suggest',
type='wordresemble_results_num', # api_wordrelresemble
model= wordresemble.WordRel,
model=wordresemble.WordRel,
query_deferred=lambda: wordresemble.WordRel.objects.filter(
category__in=[13, 12, 11, 9, 1]).query,
get_data_func=WordResemble.set_data_to_redis,
......
trans2es/utils/brandwiki_transfer.py
View file @ d8dba5ae
......@@ -4,13 +4,13 @@ import os
import sys
import logging
import traceback
from libs.tools import tzlc,getMd5Digest
from libs.tools import tzlc, getMd5Digest
from libs.es import ESPerform
from libs.cache import redis_client
import json
from trans2es.commons.commons import get_tips_suggest_list
from trans2es.commons.words_utils import QueryWordAttr,get_tips_word_type
from trans2es.commons.commons import get_tips_suggest_list, get_tips_suggest_list_v1
from trans2es.commons.words_utils import QueryWordAttr, get_tips_word_type
class BrandWikiTransfer(object):
......@@ -35,7 +35,35 @@ class BrandWikiTransfer(object):
suggest_list = get_tips_suggest_list(str(instance.name).lower())
return (item_dict,suggest_list)
return (item_dict, suggest_list)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return ([],[])
\ No newline at end of file
return ([], [])
class BrandWikiTransferV1(object):
@classmethod
def get_brandwiki_suggest_data_list(cls, instance):
try:
ret_list = list()
item_dict = dict()
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_project_query_word_weight(instance.name)
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
item_dict["tips_name_type"] = 3
ret_list.append(item_dict)
suggest_list = get_tips_suggest_list_v1(str(instance.name).lower())
return (item_dict, suggest_list)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return ([], [])
trans2es/utils/collectwiki_transfer.py
View file @ d8dba5ae
......@@ -4,13 +4,13 @@ import os
import sys
import logging
import traceback
from libs.tools import tzlc,getMd5Digest
from libs.tools import tzlc, getMd5Digest
from libs.es import ESPerform
from libs.cache import redis_client
import json
from trans2es.commons.commons import get_tips_suggest_list
from trans2es.commons.words_utils import QueryWordAttr,get_tips_word_type
from trans2es.commons.commons import get_tips_suggest_list, get_tips_suggest_list_v1
from trans2es.commons.words_utils import QueryWordAttr, get_tips_word_type
class CollectWikiTransfer(object):
......@@ -35,7 +35,35 @@ class CollectWikiTransfer(object):
suggest_list = get_tips_suggest_list(str(instance.name).lower())
return (item_dict,suggest_list)
return (item_dict, suggest_list)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return ([],[])
\ No newline at end of file
return ([], [])
class CollectWikiTransferV1(object):
@classmethod
def get_collectwiki_suggest_data_list(cls, instance):
try:
ret_list = list()
item_dict = dict()
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_project_query_word_weight(instance.name)
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
item_dict["tips_name_type"] = 3
ret_list.append(item_dict)
suggest_list = get_tips_suggest_list_v1(str(instance.name).lower())
return (item_dict, suggest_list)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return ([], [])
trans2es/utils/doctor_transfer.py
View file @ d8dba5ae
......@@ -10,7 +10,7 @@ from libs.cache import redis_client
import json
from django.conf import settings
from trans2es.commons.commons import get_tips_suggest_list
from trans2es.commons.commons import get_tips_suggest_list, get_tips_suggest_list_v1
from trans2es.commons.words_utils import QueryWordAttr, get_tips_word_type
......@@ -109,3 +109,100 @@ class DoctorTransfer(object):
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return (dict(), list())
class DoctorTransferV1(object):
@classmethod
def get_doctor_data_name_mapping_results_to_redis(cls, instance):
try:
doctor_name = instance.name.strip()
if doctor_name:
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,
}
q = {
"size": 0,
"query": {
"bool": {
"should": [
{"multi_match": multi_match}
],
"must": [
{"term": {"is_online": True}}
],
"minimum_should_match": 1
}
}
}
result_dict = ESPerform.get_search_results(ESPerform.get_cli(settings.GM_ORI_ES_INFO_LIST),
sub_index_name="doctor", doc_type="doctor", query_body=q,
offset=0, size=0)
doctor_results = result_dict["total_count"]
redis_key_prefix = "search_tips:tips_mapping_num"
redis_data = redis_client.hget(redis_key_prefix, doctor_name)
redis_val_dict = json.loads(str(redis_data, encoding="utf-8")) if redis_data else {}
total_count = doctor_results
if 't' in redis_val_dict:
total_count += int(redis_val_dict['t'])
redis_val_dict['t'] = total_count
redis_val_dict['d'] = doctor_results
logging.info(
"duan add,redis_key_prefix:%s,redis_val_dict:%s" % (str(redis_key_prefix), str(redis_val_dict)))
redis_client.hset(redis_key_prefix, doctor_name, json.dumps(redis_val_dict))
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
@classmethod
def get_doctor_suggest_data_list(cls, instance):
try:
ret_list = list()
keyword = instance.name
cut_bool = False
cut_word = ["下线", "停用", "已经下线", "账号停用"]
for i in cut_word:
if keyword.find(i) >= 0:
cut_bool = True
item_dict = dict()
suggest_list = list()
if cut_bool == False:
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["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
item_dict["tips_name_type"] = 2
ret_list.append(item_dict)
suggest_list = get_tips_suggest_list_v1(instance.name)
return (item_dict, suggest_list)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return (dict(), list())
trans2es/utils/hospital_transfer.py
View file @ d8dba5ae
......@@ -9,7 +9,7 @@ from libs.es import ESPerform
from libs.cache import redis_client
import json
from trans2es.commons.commons import get_tips_suggest_list
from trans2es.commons.commons import get_tips_suggest_list, get_tips_suggest_list_v1
from trans2es.commons.words_utils import QueryWordAttr, get_tips_word_type
......@@ -62,3 +62,54 @@ class HospitalTransfer(object):
redis_client.hset(redis_key_prefix, name, pos_list)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
class HospitalTransferV1(object):
@classmethod
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"]
for i in cut_word:
if name.find(i) >= 0:
cut_bool = True
item_dict = dict()
suggest_list = list()
if cut_bool == False:
item_dict = dict()
item_dict["id"] = getMd5Digest(str(name))
item_dict["ori_name"] = name
item_dict["is_online"] = instance.is_online
item_dict["order_weight"] = QueryWordAttr.get_hospital_query_word_weight(name)
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
item_dict["tips_name_type"] = 1
ret_list.append(item_dict)
suggest_list = get_tips_suggest_list_v1(str(name).lower())
return (item_dict, suggest_list)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return (dict(), list())
@classmethod
def get_hospital_lat_lng_info_to_redis(cls, instance):
try:
redis_key_prefix = "search_tips:hospital_latlng_info"
name = instance.name.strip()
lng = instance.baidu_loc_lng
lat = instance.baidu_loc_lat
pos_list = [lng, lat]
redis_client.hset(redis_key_prefix, name, pos_list)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
trans2es/utils/itemwiki_transfer.py
View file @ d8dba5ae
......@@ -10,7 +10,7 @@ from libs.cache import redis_client
import json
from django.conf import settings
from trans2es.commons.commons import get_tips_suggest_list
from trans2es.commons.commons import get_tips_suggest_list,get_tips_suggest_list_v1
from trans2es.commons.words_utils import QueryWordAttr, get_tips_word_type
......@@ -104,3 +104,94 @@ class ItemWikiTransfer(object):
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return ([], [])
class ItemWikiTransferV1(object):
@classmethod
def get_wiki_data_name_mapping_results_to_redis(cls, instance):
try:
wiki_name = instance.name.strip()
if wiki_name:
multi_fields = {
'name': 1,
'association_tags': 1,
"description": 1,
"use_result": 1,
"treatment_method": 1,
"effect": 1,
"ingredient": 1,
"nature_type": 1,
"shape": 1,
}
query_fields = ['^'.join((k, str(v))) for (k, v) in multi_fields.items()]
multi_match = {
'query': wiki_name,
'type': 'cross_fields',
'operator': 'or',
'fields': query_fields,
}
q = {
"size": 0,
"query": {
"bool": {
"should": [
{"multi_match": multi_match}
],
"must": [
{"term": {"is_online": True}}
],
"minimum_should_match": 1
}
}
}
result_dict = ESPerform.get_search_results(ESPerform.get_cli(settings.GM_ORI_ES_INFO_LIST),
sub_index_name="newwiki", doc_type="newwiki", query_body=q,
offset=0, size=0)
doctor_results = result_dict["total_count"]
redis_key_prefix = "search_tips:tips_mapping_num"
redis_data = redis_client.hget(redis_key_prefix, wiki_name)
redis_val_dict = json.loads(str(redis_data, encoding="utf-8")) if redis_data else {}
total_count = doctor_results
if 't' in redis_val_dict:
total_count += int(redis_val_dict['t'])
redis_val_dict['t'] = total_count
redis_val_dict['w'] = doctor_results
redis_client.hset(redis_key_prefix, wiki_name, json.dumps(redis_val_dict))
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
@classmethod
def get_itemwiki_suggest_data_list(cls, instance):
try:
ret_list = list()
item_dict = dict()
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_project_query_word_weight(instance.name)
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
item_dict["tips_name_type"] = 3
ret_list.append(item_dict)
suggest_list = get_tips_suggest_list_v1(str(instance.name).lower())
return (item_dict, suggest_list)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return ([], [])
trans2es/utils/productwiki_transfer.py
View file @ d8dba5ae
......@@ -9,7 +9,7 @@ from libs.es import ESPerform
from libs.cache import redis_client
import json
from trans2es.commons.commons import get_tips_suggest_list
from trans2es.commons.commons import get_tips_suggest_list, get_tips_suggest_list_v1
from trans2es.commons.words_utils import QueryWordAttr, get_tips_word_type
......@@ -39,3 +39,31 @@ class ProduceWikiTransfer(object):
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return ([], [])
class ProduceWikiTransferV1(object):
@classmethod
def get_productwiki_suggest_data_list(cls, instance):
try:
ret_list = list()
item_dict = dict()
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_project_query_word_weight(instance.name)
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
item_dict["tips_name_type"] = 3
ret_list.append(item_dict)
suggest_list = get_tips_suggest_list_v1(str(instance.name).lower())
return (item_dict, suggest_list)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return ([], [])
trans2es/utils/tag_transfer.py
View file @ d8dba5ae
......@@ -10,7 +10,7 @@ from libs.cache import redis_client
import json
from django.conf import settings
from trans2es.commons.commons import get_tips_suggest_list
from trans2es.commons.commons import get_tips_suggest_list, get_tips_suggest_list_v1
from trans2es.commons.words_utils import QueryWordAttr, get_tips_word_type
......@@ -97,8 +97,8 @@ class TagTransfer(object):
for i in cut_word:
if keyword.find(i) >= 0:
cut_bool = True
item_dict = dict()
if cut_bool == False:
item_dict = dict()
item_dict["id"] = getMd5Digest(str(instance.name))
item_dict["ori_name"] = instance.name
item_dict["is_online"] = instance.is_online
......@@ -117,3 +117,108 @@ class TagTransfer(object):
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return ([], [])
class TagTransferV1(object):
tips_num_redis_key_prefix = "search_tips:tips_mapping_num"
@classmethod
def get_tag_data_name_mapping_results_to_redis(cls, instance):
try:
tag_name = instance.name.strip()
q = dict()
if tag_name:
multi_fields = {
'short_description': 1,
'doctor.name': 1,
'doctor.hospital.name': 1,
'doctor.hospital.city_name': 1,
'doctor.hospital.city_province_name': 1,
'closure_tags': 1, # 5.9版 搜索所有tag
'doctor.hospital.officer_name': 1 # 搜索机构管理者
}
query_fields = ['^'.join((k, str(v))) for (k, v) in multi_fields.items()]
multi_match = {
'query': tag_name,
'type': 'cross_fields',
'operator': 'and',
'fields': query_fields,
}
sku_query = {
"nested": {
"path": "sku_list",
"query": {
"multi_match": {
"query": tag_name,
"fields": ["sku_list.name^2"],
'operator': 'and',
'type': 'cross_fields'
}
}
}
}
q['query'] = {
'bool': {
"should": [
{'multi_match': multi_match},
sku_query
],
"must": [
{"term": {"is_online": True}}
],
"minimum_should_match": 1
}
}
result_dict = ESPerform.get_search_results(ESPerform.get_cli(settings.GM_ORI_ES_INFO_LIST),
sub_index_name="service", doc_type="service", query_body=q,
offset=0, size=0)
doctor_results = result_dict["total_count"]
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
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_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_tag_suggest_data_list(cls, instance):
try:
ret_list = list()
keyword = instance.name
cut_bool = False
cut_word = ["下线", "停用", "已经下线", "账号停用"]
for i in cut_word:
if keyword.find(i) >= 0:
cut_bool = True
item_dict = dict()
if cut_bool == False:
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_project_query_word_weight(instance.name)
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
item_dict["tips_name_type"] = 0
ret_list.append(item_dict)
suggest_list = get_tips_suggest_list_v1(str(instance.name).lower())
return (item_dict, suggest_list)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return ([], [])
trans2es/utils/wordresemble.py
View file @ d8dba5ae
......@@ -12,7 +12,7 @@ import json
from django.conf import settings
from trans2es.models import wordresemble
from trans2es.commons.words_utils import QueryWordAttr, get_tips_word_type
from trans2es.commons.commons import get_tips_suggest_list
from trans2es.commons.commons import get_tips_suggest_list, get_tips_suggest_list_v1
from trans2es.utils.doctor_transfer import DoctorTransfer
from trans2es.utils.itemwiki_transfer import ItemWikiTransfer
from trans2es.utils.tag_transfer import TagTransfer
......@@ -201,3 +201,188 @@ class WordResemble(object):
redis_client.hset(QUERY_KEY, query_base64, json.dumps(keyword_value))
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
class WordResembleV1(object):
@classmethod
def get_word_resemble_list(cls, keyword):
try:
query_sql_item = wordresemble.WordRel.objects.filter(keyword=keyword)
temp_list = list()
for sql_obj in query_sql_item:
temp_list.extend(list(sql_obj.all_resembles.all().values_list('word', flat=True)))
resemble_list = list()
for item in temp_list:
resemble_list.extend(item.split("、"))
return resemble_list
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return list()
@classmethod
def get_resemble_list(cls, instance):
try:
try:
ret_list = list()
item_dict = dict()
keyword = instance.keyword
cut_bool = False
cut_word = ["下线", "停用", "已经下线", "账号停用"]
for i in cut_word:
if keyword.find(i) >= 0:
cut_bool = True
if cut_bool == False:
item_dict["id"] = getMd5Digest(str(instance.keyword))
item_dict["ori_name"] = instance.keyword
item_dict["is_online"] = True
item_dict["order_weight"] = QueryWordAttr.get_project_query_word_weight(instance.keyword)
item_dict["results_num"] = QueryWordAttr.get_query_results_num(instance.keyword)
item_dict["type_flag"] = get_tips_word_type(instance.keyword)
item_dict["offline_score"] = 0.0
item_dict["tips_name_type"] = 4
ret_list.append(item_dict)
suggest_list = get_tips_suggest_list_v1(str(instance.keyword).lower())
return (item_dict, suggest_list)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return ([], [])
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
return list()
@classmethod
def get_all_data_name_mapping_results_to_redis(cls, instance):
try:
total_count = 0
instance.name = instance.keyword
# 获取百科的
ItemWikiTransfer.get_wiki_data_name_mapping_results_to_redis(instance)
# 获取美购的
TagTransfer.get_tag_data_name_mapping_results_to_redis(instance)
# 获取医生医院的
DoctorTransfer.get_doctor_data_name_mapping_results_to_redis(instance)
# 获取日记的
WordResemble.get_diary_data_name_mapping_results_to_redis(instance)
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
@classmethod
def get_diary_data_name_mapping_results_to_redis(cls, instance):
try:
tips_num_redis_key_prefix = "search_tips:tips_mapping_num"
tag_name = instance.name.strip()
q = dict()
if tag_name:
multi_fields = {
'tags': 8,
'doctor.name': 4,
'doctor.hospital.name': 3,
'doctor.hospital.officer_name': 3,
'user.last_name': 2,
'service.name': 1,
"title": 2}
query_fields = ['^'.join((k, str(v))) for (k, v) in multi_fields.items()]
multi_match = {
'query': tag_name,
'type': 'cross_fields',
'operator': 'and',
'fields': query_fields,
}
q['query'] = {
'bool': {
"should": [
{'multi_match': multi_match}
],
"must": [
{"term": {"is_online": True}}
],
"minimum_should_match": 1
}
}
result_dict = ESPerform.get_search_results(ESPerform.get_cli(settings.GM_ORI_ES_INFO_LIST),
sub_index_name="diary", doc_type="diary", query_body=q,
offset=0, size=0)
doctor_results = result_dict["total_count"]
redis_data = redis_client.hget(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
if 't' in redis_val_dict:
total_count += int(redis_val_dict['t'])
redis_val_dict['t'] = total_count
redis_val_dict['r'] = doctor_results
redis_client.hset(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 set_data_to_redis(cls, instance):
try:
keyword_value = []
QUERY_KEY = "query:search_tip"
tips_num_redis_key_prefix = "search_tips:tips_mapping_num"
query_sql_item = wordresemble.WordRel.objects.filter(keyword=instance.keyword)
for sql_obj in query_sql_item:
count = 0
words = list(sql_obj.all_resembles.all().values_list('word', flat=True))
query_base64 = base64.b64encode(instance.keyword.encode('utf8')).decode('utf8')
for items in words:
count += 1
wordresemble_value = []
# 先存储本体词
item_name = base64.b64encode(items.encode('utf8')).decode('utf8')
# 获取该同义词的结果
redis_data = redis_client.hget(tips_num_redis_key_prefix, items)
logging.info("get redis_data:%s" % redis_data)
redis_val_dict = json.loads(str(redis_data, encoding="utf-8")) if redis_data else {}
total_count = 0
if 't' in redis_val_dict:
total_count = int(redis_val_dict['t'])
value = {item_name: total_count}
keyword_value.append(value)
redis_data = redis_client.hget(tips_num_redis_key_prefix, instance.keyword)
redis_val_dict = json.loads(str(redis_data, encoding="utf-8")) if redis_data else {}
total_count = 0
if 't' in redis_val_dict:
total_count = int(redis_val_dict['t'])
value = {query_base64: total_count}
wordresemble_value.append(value)
if count == 1:
keyword_value.append(value)
if words:
for w in words:
value_name_w = base64.b64encode(w.encode('utf8')).decode('utf8')
redis_data = redis_client.hget(tips_num_redis_key_prefix, w)
redis_val_dict = json.loads(str(redis_data, encoding="utf-8")) if redis_data else {}
total_count = 0
if 't' in redis_val_dict:
total_count = int(redis_val_dict['t'])
value = {value_name_w: total_count}
wordresemble_value.append(value)
if len(wordresemble_value) > 0:
logging.info("get type wordresemble_value:%s" % type(json.dumps(wordresemble_value)))
redis_client.hset(QUERY_KEY, item_name, json.dumps(wordresemble_value))
if len(keyword_value) > 0:
logging.info("get type keyword_value:%s" % type(json.dumps(keyword_value)))
redis_client.hset(QUERY_KEY, query_base64, json.dumps(keyword_value))
except:
logging.error("catch exception,err_msg:%s" % traceback.format_exc())
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment