一.keyword 字段和keyword数据类型
1、测试准备数据
POST /forum/article/_bulk{ "index": { "_id": 1 }}{ "articleID" : "XHDK-A-1293-#fJ3", "userID" : 1, "hidden": false, "postDate": "2017-01-01" }{ "index": { "_id": 2 }}{ "articleID" : "KDKE-B-9947-#kL5", "userID" : 1, "hidden": false, "postDate": "2017-01-02" }{ "index": { "_id": 3 }}{ "articleID" : "JODL-X-1937-#pV7", "userID" : 2, "hidden": false, "postDate": "2017-01-01" }{ "index": { "_id": 4 }}{ "articleID" : "QQPX-R-3956-#aD8", "userID" : 2, "hidden": true, "postDate": "2017-01-02" }
2、查询mapping
GET /forum/_mapping/article{"forum": {"mappings": {"article": {"properties": {"articleID": {"type": "text","fields": {"keyword": {"type": "keyword","ignore_above": 256}}},"hidden": {"type": "boolean"},"postDate": {"type": "date"},"userID": {"type": "long"}}}}}}
es 5.2版本,字段数据类型为text的字段(type=text) ,es默认会设置两个field,一个是field本身,比如articleID,就是分词的;还有一个的话,就是field.keyword,articleID.keyword,默认不分词,会最多保留256个字符articleID.keyword,是es最新版本内置建立的field,就是不分词的。所以一个articleID过来的时候,会建立两次索引,一次是自己本身,是要分词的,分词后放入倒排索引;另外一次是基于articleID.keyword,不分词,保留256个字符最多,直接一个字符串放入倒排索引中。所以term filter,对text过滤,可以考虑使用内置的field.keyword来进行匹配。但是有个问题,默认就保留256个字符。所以尽可能还是自己去手动建立索引,指定not_analyzed吧。在最新版本的es中,不需要指定not_analyzed也可以,将type=keyword即可。3、测试测试1:使用articleID搜索GET /forum/article/_search{"query" : {"constant_score" : {"filter" : {"term" : {"articleID" : "XHDK-A-1293-#fJ3"}}}}}结果:查询不到指定的document{
"took": 1,"timed_out": false,"_shards": {"total": 5,"successful": 5,"failed": 0},"hits": {"total": 0,"max_score": null,"hits": []}
}测试2:使用articleID.keyword搜索GET /forum/article/_search{"query" : {"constant_score" : {"filter" : {"term" : {"articleID.keyword" : "XHDK-A-1293-#fJ3"}}}}}结果:
{"took": 2,"timed_out": false,"_shards": {"total": 5,"successful": 5,"failed": 0},"hits": {"total": 1,"max_score": 1,"hits": [{"_index": "forum","_type": "article","_id": "1","_score": 1,"_source": {"articleID": "XHDK-A-1293-#fJ3","userID": 1,"hidden": false,"postDate": "2017-01-01"}}]}}
测试3:term查询GET /forum/article/_search{"query" : {"constant_score" : {"filter" : {"term" : {"userID" : 1}}}}}term filter/query:对搜索文本不分词,直接拿去倒排索引中匹配,你输入的是什么,就去匹配什么比如说,如果对搜索文本进行分词的话,“helle world” --> “hello”和“world”,两个词分别去倒排索引中匹配term,“hello world” --> “hello world”,直接去倒排索引中匹配“hello world”
4、查看分词GET /forum/_analyze{"field": "articleID","text": "XHDK-A-1293-#fJ3"}GET /forum/_analyze{"field": "articleID.keyword","text": "XHDK-A-1293-#fJ3"}默认是analyzed的text类型的field,建立倒排索引的时候,就会对所有的articleID分词,分词以后,原本的articleID就没有了,只有分词后的各个word存在于倒排索引中。term,是不对搜索文本分词的,XHDK-A-1293-#fJ3 --> XHDK-A-1293-#fJ3;但是articleID建立索引的时候,XHDK-A-1293-#fJ3 --> xhdk,a,1293,fj3
5、定义keyword数据类型的字段
(1)删除索引 DELETE /forum
(2)重建索引
PUT /forum{"mappings": {"article": {"properties": {"articleID": {"type": "keyword"}}}}}
(3)准备数据
POST /forum/article/_bulk{ "index": { "_id": 1 }}{ "articleID" : "XHDK-A-1293-#fJ3", "userID" : 1, "hidden": false, "postDate": "2017-01-01" }{ "index": { "_id": 2 }}{ "articleID" : "KDKE-B-9947-#kL5", "userID" : 1, "hidden": false, "postDate": "2017-01-02" }{ "index": { "_id": 3 }}{ "articleID" : "JODL-X-1937-#pV7", "userID" : 2, "hidden": false, "postDate": "2017-01-01" }{ "index": { "_id": 4 }}{ "articleID" : "QQPX-R-3956-#aD8", "userID" : 2, "hidden": true, "postDate": "2017-01-02" }
(4)测试articleID查询
GET /forum/article/_search{"query" : {"constant_score" : {"filter" : {"term" : {"articleID" : "XHDK-A-1293-#fJ3"}}}}}6、小结(1)term filter:根据exact value进行搜索,数字、boolean、date天然支持
(2)text需要建索引时指定为not_analyzed,才能用term
query
(3)相当于SQL中的单个where条件
二、filter执行原理深度剖析
1、bitset机制
每个filter根据在倒排索引中搜索的结果构建一个bitset(位集),用以存储搜索的结果。简单的数据结构去实现复杂的功能,可以节省内存空间,提升性能。bitset,就是一个二进制的数组,数组每个元素都是0或1,用来标识一个doc对一个filter条件是否匹配,如果匹配就是1,不匹配就是0。比如:[0, 1, 1]。
遍历每个过滤条件对应的bitset,优先从最稀疏的开始搜索,查找满足所有条件的document(先遍历比较稀疏的bitset,就可以先过滤掉尽可能多的数据发)
2、caching
bitset机制
跟踪query,在最近256个query中超过一定次数的过滤条件,缓存其bitset。对于小segment(<1000,或<3%),不缓存bitset。在最近的256个filter中,有某个filter超过了一定的次数,次数不固定,就会自动缓存这个filter对应的bitset。filter针对小segment获取到的结果,可以不缓存,segment记录数<1000,或者segment大小<index总大小的3% segment数据量很小,此时哪怕是扫描也很快;segment会在后台自动合并,小segment很快就会跟其他小segment合并成大segment,此时就缓存也没有什么意义,segment很快就消失了。
cache biset的自动更新:如果document有新增或修改,那么cached
bitset会被自动更新
3、filter与query的对比
filter比query的好处就在于会caching。filter大部分情况下来说,在query之前执行,先尽量过滤掉尽可能多的数据query:是会计算doc对搜索条件的relevance score(相关评分),还会根据这个score去排序filter:只是简单过滤出想要的数据,不计算relevance score,也不排序
三、基于bool组合多个filter条件来搜索数据
1、搜索发帖日期为2017-01-01,或者帖子ID为XHDK-A-1293-#fJ3的帖子,同时要求帖子的发帖日期绝对不为2017-01-02
GET /forum/article/_search{"query": {"constant_score": {"filter": {"bool": {"should":[{"term":{"postDate":"2017-01-01"}},{"term":{"articleID":"HDK-A-1293-#fJ3"}}],"must_not":{"term":{"postDate":"2017-01-02"}}}}}}}
2、搜索帖子ID为XHDK-A-1293-#fJ3,或者是帖子ID为JODL-X-1937-#pV7而且发帖日期为2017-01-01的帖子
GET /forum/article/_search{"query": {"constant_score": {"filter": {"bool": {"should":[{"term":{"articleID":"XHDK-A-1293-#fJ3"}},{"bool":{"must":[{"term":{"articleID":"JODL-X-1937-#pV7"}},{"term":{"postDate":"2017-01-01"}}]}}]}}}}}
四、term和terms
五、filter
range
测试数据:
为帖子数据增加浏览量的字段
POST /forum/article/_bulk{ "update": { "_id": "1"} }{ "doc" : {"view_cnt" : 30} }{ "update": { "_id": "2"} }{ "doc" : {"view_cnt" : 50} }{ "update": { "_id": "3"} }{ "doc" : {"view_cnt" : 100} }{ "update": { "_id": "4"} }{ "doc" : {"view_cnt" : 80} }
1、搜索浏览量在30~60之间的帖子GET /forum/article/_search{"query": {"constant_score": {"filter": {"range": {"view_cnt": {"gt": 30, //gt大于 gte大于或 等于"lt": 60 //lt大于 lte大于或等于}}}}}}2、搜索发帖日期在最近1个月的帖子
GET /forum/article/_search{"query": {"constant_score": {"filter": {"range": {"postDate": {"gt": "2017-03-10||-30d"}}}}}}GET /forum/article/_search{"query": {"constant_score": {"filter": {"range": {"postDate": {"gt": "now-30d"}}}}}}
六、match
query 精准查询
测试数据:为帖子数据增加标题字段POST /forum/article/_bulk{ "update": { "_id": "1"} }{ "doc" : {"title" : "this is java and elasticsearch blog"} }{ "update": { "_id": "2"} }{ "doc" : {"title" : "this is java blog"} }{ "update": { "_id": "3"} }{ "doc" : {"title" : "this is elasticsearch blog"} }{ "update": { "_id": "4"} }{ "doc" : {"title" : "this is java, elasticsearch, hadoop blog"} }{ "update": { "_id": "5"} }{ "doc" : {"title" : "this is spark blog"} }
1、match queryGET /forum/article/_search{"query": {"match": {"title": "java elasticsearch"}}}相当于:{"bool": {"should": [{ "term": { "title": "java" }},{ "term": { "title": "elasticsearch" }}]}}如果title字段是analyzed则进行full text全文搜索,则返回title字段包含java 或者elasticsearch 或者两个都包含的document如果是not_analyzed则进行exact value(相当于temr query),则只返回包含java elasticsearch的documentGET /forum/article/_search{"query": {"match": {"title": {
"query": "java elasticsearch","operator": "and" //full text 中 返回都包含“java”和"elasticsearch“的document
}}}}相当于:{"bool": {"must": [{ "term": { "title": "java" }},{ "term": { "title": "elasticsearch" }}]}}GET /forum/article/_search{"query": {"match": {"title": {"query": "java elasticsearch spark hadoop","minimum_should_match": "75%" // full text中返回,包含指定条件的75%的document}}}}相当于:{"bool": {"should": [{ "term": { "title": "java" }},{ "term": { "title": "elasticsearch" }},{ "term": { "title": "hadoop" }},{ "term": { "title": "spark" }}],"minimum_should_match": 3}}
2、用bool组合多个搜索条件,来搜索title
GET /forum/article/_search{"query": {"bool": {"must": { "match": { "title": "java" }},"must_not": { "match": { "title": "spark" }},"should": [{ "match": { "title": "hadoop" }},{ "match": { "title": "elasticsearch" }}]}}}
bool组合多个搜索条件,如何计算relevance
score
must和should搜索对应的分数,加起来,除以must和should的总数排名第一:java,同时包含should中所有的关键字,hadoop,elasticsearch排名第二:java,同时包含should中的elasticsearch排名第三:java,不包含should中的任何关键字should是可以影响相关度分数的must是确保说,谁必须有这个关键字,同时会根据这个must的条件去计算出document对这个搜索条件的relevance score在满足must的基础之上,should中的条件,不匹配也可以,但是如果匹配的更多,那么document的relevance score就会更高
默认情况下,should是可以不匹配任何一个的,比如上面的搜索中,this is java blog,就不匹配任何一个should条件但是有个例外的情况,如果没有must的话,那么should中必须至少匹配一个才可以比如下面的搜索,should中有4个条件,默认情况下,只要满足其中一个条件,就可以匹配作为结果返回但是可以精准控制,should的4个条件中,至少匹配几个才能作为结果返回GET /forum/article/_search{"query": {"bool": {"should": [{ "match": { "title": "java" }},{ "match": { "title": "elasticsearch" }},{ "match": { "title": "hadoop" }},{ "match": { "title": "spark" }}],"minimum_should_match": 3}}}