7.1 客户端全文搜索API调用(Verison2.0)

更新日志

• V2.0

• 支持Description 参数使用 拼音搜索，比如搜索：防晒霜 时 可以传入 “fangshai”。

• 优化中文分词。 支持 口红的 搜索 包含 口红 的索引文档，并且过滤 stop 词库中的 document。

7.1.1全文搜索 Url

// 示例：搜索资讯、产品、评论、所有文档
// 基本搜索参数：第一页 、 10条 、 排序按照 _score 评分值
// 分词搜索参数：搜索与'粉嫩'相关的文档

// 具体参数参数使用参考，7.1.2

7.1.2传递 Json 规范(搜索参数)

调用方式

http-get

例如：http://crawler.imacco.com/getElkSearch/?IndexList=info,comment,product&IndexType=full&CurrentPage=1&PageSize=10&Sort=_score&Description=粉嫩

Python-Code

#!/usr/bin/python2.7
#-*- coding:utf8 -*-
import sys
reload(sys)
sys.setdefaultencoding('utf-8')

import json
import time
import datetime
import os
import redis
from es_connect_host import ES
from esbash.info_bash import Model



class CJsonEncoder(json.JSONEncoder):
    def default(self, obj):
        if isinstance(obj, datetime.datetime):
            return obj.strftime('%Y-%m-%d %H:%M:%S')
        elif isinstance(obj, date):
            return obj.strftime("%Y-%m-%d")
        else:
            return json.JSONEncoder.default(self, obj)

class full_es_search:
    def __init__(self):

        pool = redis.ConnectionPool(host='xxx', port=6379, db=0 ,password='xxx')
        self.r = redis.Redis(connection_pool=pool)

        self.es = ES()

    # 全文搜索的Api
    ## 至少匹配一种should-match
    ## info :Title + Description
    ## comment :Comment + Description
    ## product :ProductCName.pinyin（支持拼音+分词） + ProductEName + Description
    def get_full_query(self,index,type,fromSize,size,sort,Description):
        index = index
        type = type
        query_body = {
            "query": {
                "bool": {
                  "should": [
                              { "match": { "Description": Description   }},
                              {"match": {"Title": Description}},
                              {"match": {"Comment": Description}},
                              { "match": { "ProductCName.pinyin": Description   }}
                  ]
                }
              },
            # "query": {"match": {"Description": Description}},
            # "_source": ["Description", "Title","ID"],
            # "highlight": {
            #     "pre_tags": ["<tag1>", "<tag2>"],
            #     "post_tags": ["</tag1>", "</tag2>"],
            #     "fields": {
            #         "Description": {}
            #     }
            # },

            "from": fromSize,
            "size": size,
            "sort": {
                sort: {"order": "desc"}
            }
        }

        info = self.es.get_index_query(index,type,query_body)
        infolen = len(info['hits']['hits'])
        # print (json.dumps(info['hits']['hits'][0]))
        # print (json.dumps(info['hits']['hits'][0]['_source']['KeyNO']))
        for i in range(0,infolen):
            # 添加当前资讯的评论总数
            if(info['hits']['hits'][i]['_index']=='info'):
                KeyNO = info['hits']['hits'][i]['_source']['KeyNo']
                # print KeyNO
                # print info['hits']['hits'][i]['_source']
                if  info['hits']['hits'][i]['_source']['Tag']:
                    model = Model('ali', 'InfoTag')
                    all_info_list = model.getTagListByTag(info['hits']['hits'][i]['_source']['Tag'])

                    info['hits']['hits'][i]['_source']['Tag'] = all_info_list


                info['hits']['hits'][i]['_source']['CommentCount'] = self.r.hget('comment2', KeyNO)
                model = Model('ali', 'InfoStatic')
                Static = model.getAllStaticByKeyNO(KeyNO)

                # # print info['hits']['hits'][i]['_source']['CommentCount']
                info['hits']['hits'][i]['_source']['CommentCount'] = (info['hits']['hits'][i]['_source']['CommentCount'] if info['hits']['hits'][i]['_source']['CommentCount']>0  else  0)
                # # print info['hits']['hits'][i]['_source']['CommentCount']
                # # 添加资讯的点赞数
                info['hits']['hits'][i]['_source']['LikeCount'] = self.r.bitcount('operation'+ KeyNO+'4')
                # # print info['hits']['hits'][i]['_source']['LikeCount']
                info['hits']['hits'][i]['_source']['LikeCount'] = (info['hits']['hits'][i]['_source']['LikeCount'] if info['hits']['hits'][i]['_source']['LikeCount']>0  else  0)
                info['hits']['hits'][i]['_source']['LikeCount'] = Static[0]['FakeLikeCount'] + Static[0]['LikeCount']
                # #
                info['hits']['hits'][i]['_source']['ViewCount'] = Static[0]['FakeViewCount']+ Static[0]['ViewCount']
        print (json.dumps(info,cls=CJsonEncoder))



if __name__ == "__main__":
    index = str(sys.argv[1])
    type = str(sys.argv[2])
    fromSize = str(sys.argv[3])
    size = str(sys.argv[4])
    sort = str(sys.argv[5])
    Description = str(sys.argv[6])


    full_es_search().get_full_query(index,type,fromSize,size,sort,Description)