Diving Into ElasticSearch

elasticsearch里面的search_type共有如下几种：
The type of the search operation to perform. Can be
dfs_query_then_fetch,
dfs_query_and_fetch,
query_then_fetch,
query_and_fetch. 【removed，since：http://groups.google.com/group/elasticsearch/browse_thread/thread/7aa5ea823afb499/d9e3cf3a1e1f6964】
Defaults to query_then_fetch.

form google group：“
You get proper sorted results when you use query_then_fetch (across all top
"size" results), if you use query_and_fetch, then each shard return the size
requested hits, and then they are sorted between them.
”

直接看代码里面吧，里面都有注释。

public enum SearchType {
    /**
     * Same as {@link #QUERY_THEN_FETCH}, except for an initial scatter phase which goes and computes the distributed
     * term frequencies for more accurate scoring.
     */
    DFS_QUERY_THEN_FETCH((byte) 0),
    /**
     * The query is executed against all shards, but only enough information is returned (not the document content).
     * The results are then sorted and ranked, and based on it, only the relevant shards are asked for the actual
     * document content. The return number of hits is exactly as specified in size, since they are the only ones that
     * are fetched. This is very handy when the index has a lot of shards (not replicas, shard id groups).
     */
    QUERY_THEN_FETCH((byte) 1),
    /**
     * Same as {@link #QUERY_AND_FETCH}, except for an initial scatter phase which goes and computes the distributed
     * term frequencies for more accurate scoring.
     */
    DFS_QUERY_AND_FETCH((byte) 2),
    /**
     * The most naive (and possibly fastest) implementation is to simply execute the query on all relevant shards
     * and return the results. Each shard returns size results. Since each shard already returns size hits, this
     * type actually returns size times number of shards results back to the caller.
     */
    QUERY_AND_FETCH((byte) 3),
    /**
     * Performs scanning of the results which executes the search without any sorting.
     * It will automatically start scrolling the result set.
     */
    SCAN((byte) 4),
    /**
     * Only counts the results, will still execute facets and the like.
     */
    COUNT((byte) 5);
 
    /**
     * The default search type ({@link #QUERY_THEN_FETCH}.
     */
    public static final SearchType DEFAULT = QUERY_THEN_FETCH;
 
    }

本文来自: elasticsearch里的search_type

DIE, elasticsearch, es, search_type

Diving Into ElasticSearch（10）精确控制之Routing使用

<Date: 2011-10-22> <Author: medcl> <Category: Diving Into ElasticSearch> 发表评论

前面一篇介绍parent-child的使用，我们来回顾一下：

1.先建好mapping和索引几条数据

curl -XPUT 'http://localhost:9200/news/comment/_mapping' -d '{    "comment" : {        "_parent" : {            "type" : "hot"        }    }}'
curl -XPUT 'http://localhost:9200/news/comment/1?parent=1'    -d '{    "uname" : "凤凰网安徽省六安市网友",    "content" : "河南警方的话没人信"}'
curl -XPUT 'http://localhost:9200/news/comment/2?parent=1'    -d '{    "uname" : "凤凰网湖北省武汉市网友",    "content" : "没有监督权\n没有调查\n一切当然只能是谣言"}'
curl -XPUT 'http://localhost:9200/news/comment/3?parent=1'    -d '{    "uname" : "ladygaga",    "content" : "双下肢不活动，存在废用性肌肉萎缩。请问，这个是怎么做到的？"}'
curl -XPUT 'http://localhost:9200/news/comment/4?parent=1'    -d '{    "uname" : "medcl",    "content" : "额"}'

2.获取一下这几条数据看看

http://localhost:9200/news/comment/1

结果：

{"_index":"news","_type":"comment","_id":"1","_version":1, "_source" : {    "uname" : "凤凰网安徽省六安市网友",    "content" : "河南警方的话没人信"}}

没有问题，我们再试试后面的

http://localhost:9200/news/comment/2

结果：

HTTP/1.1 404 Not Found
Access-Control-Allow-Origin: *
Content-Type: application/json; charset=UTF-8
Content-Length: 45
 
{"_index":"news","_type":"comment","_id":"2"}

嘿嘿，发现了么，居然是404，你可以继续试试后面的id为3的也是404，id为4的可以出来

试试：http://localhost:9200/news/comment/2?parent=1（索引时的path）

HTTP/1.1 404 Not Found

哈哈，貌似不行o.(ps:其实kimchy可以实现这个url pattern，但是目前没有)

那正确的方式是怎样的呢？

http://localhost:9200/news/comment/1?routing=1
http://localhost:9200/news/comment/1?routing=2
http://localhost:9200/news/comment/1?routing=3
http://localhost:9200/news/comment/1?routing=4

答案就在routing,ES帮助：http://www.elasticsearch.org/guide/reference/mapping/routing-field.html

使用我写的partial_update插件也是支持routing的，如下：

curl -XPUT 'http://localhost:9200/news/comment/4/_update?parent=1'    -d '{ "content" : "连老卡都不斗争了，难道真的登船去了吗？"}'

结果：

curl -XGET http://localhost:9200/news/comment/4?routing=1
{"_index":"news","_type":"comment","_id":"4","_version":2, "_source" : {"content":"连老卡都不斗争了，难道真的登船去了吗？","uname":"medcl"}}

发散一下,parent=2试试：

curl -XPUT 'http://localhost:9200/news/comment/4/_update?parent=2'    -d '{ "content" : "周公使管叔监殷"}'

结果：

curl -XGET http://localhost:9200/news/comment/4?routing=2
{"_index":"news","_type":"comment","_id":"4","_version":1, "_source" : {"content":"周公使管叔监殷","uname":"medcl"}}
 
curl -XGET http://localhost:9200/news/comment/4?routing=1
{"_index":"news","_type":"comment","_id":"4","_version":2, "_source" : {"content":"连老卡都不斗争了，难道真的登船去了吗？","uname":"medcl"}}

很明细，/news/comment/4存在两条记录，routing的出现，使ES的id的唯一性丢失了，并且删除索引记录的时候也必须带上routing才行，此外，查询的结果中可能会出现重复的_id。

curl -XDELETE http://localhost:9200/news/comment/4?routing=1
curl -XDELETE http://localhost:9200/news/comment/4?routing=2

再看看查询的操作,查询的时候可以指定routing，默认不区分routing，即全部扫描：

curl -XGET http://localhost:9200/news/comment/_search?q=*
curl -XGET http://localhost:9200/news/comment/_search?q=*&routing=3

总之，一旦你决定使用routing，你必须保证对这些routing做到心中有数。

本文来自: Diving Into ElasticSearch（10）精确控制之Routing使用

DIE, elasticsearch

Diving Into ElasticSearch（9）Parent-Child特性使用

<Date: 2011-10-19> <Author: medcl> <Category: Diving Into ElasticSearch> 发表评论

介绍下ElasticSearch里Parent-Child特性的使用。

//首先创建一系列新闻的索引，这里我们将hot类型作为parent-chid关系里面的parent。

curl -XPUT 'http://localhost:9200/news/hot/1'  -d '{    "uname" : "medcl",    "content" : "河南警方：“南阳老板遭逼供致残与狗同笼”纯属谎言"}'
curl -XPUT 'http://localhost:9200/news/hot/2'  -d '{    "uname" : "medcl",    "content" : "马英九打两岸牌反制绿营"}'
curl -XPUT 'http://localhost:9200/news/hot/3'  -d '{    "uname" : "medcl",    "content" : "专题：中共十七届六中全会公报"}'

阅读这篇文章的其余部分 »

本文来自: Diving Into ElasticSearch（9）Parent-Child特性使用

DIE, elasticsearch

Diving Into ElasticSearch（8）Mapping&Schema

<Date: 2011-09-27> <Author: medcl> <Category: Diving Into ElasticSearch> 4 条评论

前面应该介绍过ES是Schema Free，但是Schema Free不是说没有Schema，和Solr一样，ElasticSearch也可以设置document的schema，ES里的名字叫Mapping，其实无非就是设置document包含哪些Field，然后对每一个Field个性化的设置索引类型，是否存储，以及设置索引分析器和查询使用的分析器，Es和Solr相比有一个我认为最好的特性：就是支持Object类型，你可以像操作对象一样对对象的某个属性进行索引和查询，简单演示如下：
阅读这篇文章的其余部分 »

本文来自: Diving Into ElasticSearch（8）Mapping&Schema

DIE, elasticsearch, tutorial

ElasticSearch插件发布：PartialUpdate

<Date: 2011-09-17> <Author: medcl> <Category: Diving Into ElasticSearch, 搜索> 发表评论

地址：https://github.com/medcl/ElasticSearch.PartialUpdate

是否碰到过因为需要修改索引中某个字段，而需要将整个索引文档进行重建，是不是很麻烦啊，昨天弄了个插件，就是方便索引文档局部更新的，使用说明如下：

1.先下载插件，解压到ES的plugin目录：

elasticsearch/plguin/es-partial-update/

2.试一把

先索引一个文档吧

curl -XDELETE http://localhost:9200/index/type1/1/
curl -XPOST http://localhost:9200/index/type1/1/ -d'{"name":"medcl","blog":"http://log.medcl.net"}'
 
{"ok":true,"_index":"index","_type":"type1","_id":"1","_version":1}

执行修改操作，添加一个字段，修改一个字段

curl -XPOST http://localhost:9200/index/type1/1/_update -d'{"name":"medcl?","time":"2011-1-1"}'
 
{"ok":true,"_index":"index","_type":"type1","_id":"1","_version":2}

看看修改之后的结果吧

curl -XGET http://localhost:9200/index/type1/1/
 
{"_index":"index","_type":"type1","_id":"1","_version":2, "_source" : {"time":"2011-1-1","name":"medcl?","blog":"http://log.medcl.net"}}

本文来自: ElasticSearch插件发布：PartialUpdate

elsaticsearch, github, partialupdate, plugin

[翻译]Diving Into ElasticSearch（7）模块配置介绍：cluster

<Date: 2011-09-14> <Author: medcl> <Category: Diving Into ElasticSearch> 发表评论

纯翻译：http://www.elasticsearch.org/guide/reference/modules/cluster.html
貌似国内最近关注elasticsearch的人多了起来。
阅读这篇文章的其余部分 »

本文来自: [翻译]Diving Into ElasticSearch（7）模块配置介绍：cluster

DIE, elasticsearch

Diving Into ElasticSearch (6) 配置文件elasticsearch.yml

<Date: 2011-09-14> <Author: medcl> <Category: Diving Into ElasticSearch> 发表评论

接前面那篇吧[5]，ElasticSearch分布式架构要说清楚真不是那么容易，我那就从细节一点点的切入吧。

先看看配置文件吧：elasticsearch.yml

#gateway类型，表示持久化数据存放位置，默认local，推荐的方式，此外还有NFS、HDFS、S3
gateway.type : local
#集群名称，区分集群的唯一名称
cluster.name : 'TEST'
 
#索引文件存放目录
#path.data : '/var/elasticsearch/data'
#日志文件存放目录
#path.logs : '/var/elasticsearch/logs'
 
#网络配置
#network.tcp.keep_alive : true
#network.tcp.send_buffer_size : 8192
#network.tcp.receive_buffer_size : 8192
#gateway.recover_after_nodes : 1
#gateway.recover_after_time : 10s
#gateway.expected_nodes : 2
 
#自动发现相关配置
#discovery.zen.fd.connect_on_network_disconnect : true
#discovery.zen.initial_ping_timeout : 10s
#discovery.zen.fd.ping_interval : 2s
#discovery.zen.fd.ping_retries  : 10
 
#索引snapshot时间只对当gateway设置为NFS时有效
#index.gateway.snapshot_interval : 1s
#刷新时间间隔
#index.engine.robin.refresh_interval : -1
 
#默认索引碎片数
index.number_of_shards : 3
#默认索引副本数
index.number_of_replicas : 1
 
#默认索引合并因子
#index.merge.policy.merge_factor : 100
#index.merge.policy.min_merge_docs : 1000
#index.merge.policy.use_compound_file : true
#indices.memory.index_buffer_size : 5%
 
#Gateway相关配置
# Gateway Settings
#gateway:
#  recover_after_nodes: 1
#  recover_after_time: 5m
#  expected_nodes: 2
#提示：当集群期望节点达不到的时候，集群就会处于block，无法正常索引和查询，说明集群中某个节点未能正常启动，这正是我们期望的效果，block住，避免照成数据的不一致
 
#强制所有内存锁定，不要没事搞个swap什么的来影响性能
# Force all memory to be locked, forcing JVM to never swap
#  (make sure to set MIN and MAX mem to the same value)
#bootstrap:
#  mlockall: true
 
 
#当禁用multcast广播的时候，可以手动设置集群的节点ip
# Unicast Discovery (disable multicast)
#discovery:
#  zen:
#    multicast.enabled: false
#    unicast.hosts: ["host1", "host2"]

配置文件elasticsearch.yml的可配项比较多，并且修改之后需要重启服务才能生效。

本文来自: Diving Into ElasticSearch (6) 配置文件elasticsearch.yml

DIE, elasticsearch

ElasticSearch里DSL使用事故一则：_cache参数

doc.elasticsearch.cn解析成功了

基于私有云的elasticsearch批量部署

elasticsearch里的search_type

Diving Into ElasticSearch（10）精确控制之Routing使用

Diving Into ElasticSearch（9）Parent-Child特性使用

Diving Into ElasticSearch（8）Mapping&Schema

ElasticSearch插件发布：PartialUpdate

[翻译]Diving Into ElasticSearch（7）模块配置介绍：cluster

Diving Into ElasticSearch (6) 配置文件elasticsearch.yml

最新文章

标签

广而告之