真题4
一、参考
二、自定义分词
2.1 题目
正确的排除 ' 对查询的影响, 确保对 waynes/wayne's, kings/king's 的查询有一样的结果和分值 (custom analyzer)
2.2 分析
自定义分词器，char filters 把 ' 过滤
PUT yztest-4-1
{
"settings": {
"analysis": {
"analyzer": {
"my_custom_analyzer": {
"type": "custom",
"tokenizer": "standard",
"char_filter": [
"my_char_filter"
]
}
},
"char_filter": {
"my_char_filter": {
"type": "mapping",
"mappings": [
"' => "
]
}
}
}
},
"mappings": {
"properties": {
"f1": {
"type": "text",
"analyzer": "my_custom_analyzer"
}
}
}
}
GET yztest-4-1/_analyze
{
"field": "f1",
"text": "tom's toms"
}
POST yztest-4-1/_bulk
{"index":{}}
{"f1": "tom's"}
{"index":{}}
{"f1": "toms"}
GET yztest-4-1/_search
{
"query": {
"match": {
"f1": "tom's"
}
}
}
三、动态模版
3.1 题目
要求将 string 类型映射为 keyword, 将 x_ 开头字段映射为 integer (dynamic template)
3.2 分析
动态模板
3.3 解答
PUT yztest-4-2
{
"mappings": {
"dynamic_templates": [
{
"long_to_int": {
"match": "long*",
"mapping": {
"type": "integer"
}
}
},
{
"string_to_keyword": {
"match_mapping_type": "string",
"mapping": {
"type": "keyword"
}
}
}
]
}
}
POST yztest-4-2/_bulk
{"index":{}}
{"long1": 100, "long2": 1.1, "f1": "v1"}
GET yztest-4-2/_mapping
四、update_by_query + pipeline
4.1 题目
为索引添加一个新字段 e 是已有四个字段 a b c d 的拼接 (pipeline + update_by_query)
4.2 分析
可以使用 pipeline script 或者 update with script
4.3 解答
PUT yztest-4-3
POST yztest-4-3/_bulk
{"index": {}}
{"f1": "v1", "f2": "v2", "f3": "v3"}
POST yztest-4-3/_update_by_query
{
"script": {
"source": "ctx._source.f4 = ctx._source.f1 + '-' + ctx._source.f2 + '-' + ctx._source.f3",
"lang": "painless"
},
"query": {
"match_all": {}
}
}
{
"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" : "yztest-4-3",
"_id" : "U84D5IYBHHgM2AP0bStj",
"_score" : 1.0,
"_source" : {
"f1" : "v1",
"f2" : "v2",
"f3" : "v3",
"f4" : "v1-v2-v3"
}
}
]
}
}
PUT yztest-4-3
POST yztest-4-3/_bulk
{"index": {}}
{"f1": "v1", "f2": "v2", "f3": "v3"}
POST yztest-4-3/_update_by_query?pipeline=yztest-4-3-pipeline
{
"query": {
"match_all": {}
}
}
GET yztest-4-3/_search
五、multi_match
5.1 题目
对三个字段 a/b/c 查询 xxx, 要求 c 字段 boost 2, 各字段查询算分加和 (bool query should 或者 multi_match)
5.2 分析
multi_match 的 most_fields 和 boost
5.3 解答
PUT yztest-4-4
POST yztest-4-4/_bulk
{"index": {}}
{"f1": "hello", "f2": "hello hello", "f3": "hello world"}
GET yztest-4-4/_search
{
"query": {
"multi_match": {
"query": "hello",
"fields": ["f1^2", "f2", "f3"],
"type": "most_fields"
}
}
}
六、聚合和子聚合
6.1 题目
earthquakes 索引按照月份进行聚合, 并统计出每月 magnitude 和 depth 最大值 (date_histogram aggregation + max aggregation)
6.2 分析
6.3 解答
PUT yztest-4-5
{
"mappings": {
"properties": {
"date": {
"type": "date"
},
"f1": {
"type": "long"
}
}
}
}
POST yztest-4-5/_bulk
{"index": {}}
{"date": "2023-01-01", "f1": 10}
{"index": {}}
{"date": "2023-01-02", "f1": 20}
{"index": {}}
{"date": "2023-02-01", "f1": 12}
{"index": {}}
{"date": "2023-03-02", "f1": 15}
GET yztest-4-5/_search
{
"size": 0,
"aggs": {
"a1": {
"date_histogram": {
"field": "date",
"calendar_interval": "month"
},
"aggs": {
"a2": {
"avg": {
"field": "f1"
}
}
}
},
"a3": {
"max_bucket": {
"buckets_path": "a1>a2"
}
}
}
}
{
"took" : 436,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 4,
"relation" : "eq"
},
"max_score" : null,
"hits" : [ ]
},
"aggregations" : {
"a1" : {
"buckets" : [
{
"key_as_string" : "2023-01-01T00:00:00.000Z",
"key" : 1672531200000,
"doc_count" : 2,
"a2" : {
"value" : 15.0
}
},
{
"key_as_string" : "2023-02-01T00:00:00.000Z",
"key" : 1675209600000,
"doc_count" : 1,
"a2" : {
"value" : 12.0
}
},
{
"key_as_string" : "2023-03-01T00:00:00.000Z",
"key" : 1677628800000,
"doc_count" : 1,
"a2" : {
"value" : 15.0
}
}
]
},
"a3" : {
"value" : 15.0,
"keys" : [
"2023-01-01T00:00:00.000Z",
"2023-03-01T00:00:00.000Z"
]
}
}
}
七、检索 + 高亮
7.1 题目
对 movie 的 title 使用 match phrase 查询 star trek, 按照 revenue 倒序排序, title 字段用 高亮 (order + highlight)
7.2 分析
短语查询，高亮和排序
7.3 解答
PUT yztest-4-6
POST yztest-4-6/_bulk
{"index": {}}
{"f1": "test doc"}
GET yztest-4-6/_search
{
"query": {
"match_phrase": {
"f1": "test"
}
},
"sort": [
{
"f1.keyword": {
"order": "desc"
}
}
],
"highlight": {
"fields": {
"f1": { "pre_tags" : ["<strong>"], "post_tags" : ["</strong>"] }
}
}
}
{
"took" : 1689,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 1,
"relation" : "eq"
},
"max_score" : null,
"hits" : [
{
"_index" : "yztest-4-6",
"_id" : "Ws4u5IYBHHgM2AP0Gisz",
"_score" : null,
"_source" : {
"f1" : "test doc"
},
"highlight" : {
"f1" : [
"<strong>test</strong> doc"
]
},
"sort" : [
"test doc"
]
}
]
}
}
