准备数据:
PUT /test_index/_create/1
{
"test_field": "C3D0-KD345"
}
PUT /test_index/_create/2
{
"test_field": "C3K5-DFG65"
}
PUT /test_index/_create/3
{
"test_field": "C4I8-UI365"
}
前缀搜索:
原理:前缀匹配不会计算相关度分数,与前缀过滤的唯一区别就是过滤会有cache bitset。它会扫描整个倒排索引。找到符合前缀条件的文档。所以说前缀越短,要处理的文档就越多,性能就越差,尽可能应该用长前缀搜索。
示例,搜索前缀为C3的文档:
GET /test_index/_search
{
"query": {
"match_phrase_prefix": {
"test_field": "C3"
}
}
}
结果:
{
"took" : 16,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 2,
"relation" : "eq"
},
"max_score" : 0.9808292,
"hits" : [
{
"_index" : "test_index",
"_type" : "_doc",
"_id" : "1",
"_score" : 0.9808292,
"_source" : {
"test_field" : "C3D0-KD345"
}
},
{
"_index" : "test_index",
"_type" : "_doc",
"_id" : "2",
"_score" : 0.9808292,
"_source" : {
"test_field" : "C3K5-DFG65"
}
}
]
}
}
通配符搜索:
通配符搜索跟前缀搜索类似,比前缀搜索要更加强大。也是需要扫描整个倒排索引,性能也是很差的。
?:表示匹配任意一个字符
:表示匹配任意多个字符
示例:通配符搜索条件为*4?的文档
GET /test_index/_search
{
"query": {
"wildcard": {
"test_field": {
"value": "*4?"
}
}
}
}
输出结果:
{
"took" : 1,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 1,
"relation" : "eq"
},
"max_score" : 1.0,
"hits" : [
{
"_index" : "test_index",
"_type" : "_doc",
"_id" : "1",
"_score" : 1.0,
"_source" : {
"test_field" : "C3D0-KD345"
}
}
]
}
}
正则搜索:
regexp可以说功能比之前的通配符搜索功能更加强大,但是都会扫描整个倒排索引,性能也是会非常的差。
[0-9]:指定范围内的数字
[a-z]:指定范围内的字母
.:一个字符
+:前面的正则表达式可以出现一次或多次
*:前面的正则表达式可以出现零次或多次
{n}: n是非负整数,表示匹配n次
示例,搜索条件为.*[a-z]{3}[0-9]{2}的文档
GET /test_index/_search
{
"query": {
"regexp": {
"test_field": {
"value": ".*[a-z]{3}[0-9]{2}"
}
}
}
}
输出结果:
{
"took" : 1,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 1,
"relation" : "eq"
},
"max_score" : 1.0,
"hits" : [
{
"_index" : "test_index",
"_type" : "_doc",
"_id" : "2",
"_score" : 1.0,
"_source" : {
"test_field" : "C3K5-DFG65"
}
}
]
}
}