逗号分割的字符串，如何进行分组统计
在使用Elasticsearch的时候，经常会遇到类似标签的需求，比如给学生信息打标签，并且使用逗号分割的字符串进行存储，后期如果遇到需要根据标签统计学生数量的需求，则可以使用如下的命令进行处理。 前两个代码段落分别是 mapping、setting的配置，第三个代码段是请求命令，第四个代码段是分组结果。
"studentTags": {
"type": "text",
"analyzer": "comma",
"search_analyzer": "comma"
}
{
"analysis": {
"filter": {
},
"analyzer": {
"comma": {
"type": "pattern",
"pattern":","
}
},
"char_filter": {
},
"tokenizer": {
}
}
}
{
"size": 0,
"aggs": {
"group_by_field": {
"terms": {
"script": {
"source": "if (params['_source']['studentTags'] != null) { params['_source']['studentTags'].splitOnToken(',') }",
"lang": "painless"
},
"size": 10
}
}
}
}
{
"took": 9,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 27,
"relation": "eq"
},
"max_score": null,
"hits": []
},
"aggregations": {
"group_by_field": {
"doc_count_error_upper_bound": 0,
"sum_other_doc_count": 0,
"buckets": [
{
"key": "A",
"doc_count": 19
},
{
"key": "B",
"doc_count": 18
},
{
"key": "C",
"doc_count": 13
},
{
"key": "D",
"doc_count": 12
}
]
}
}
}
Nested对象，如何进行分组统计
在使用Elasticsearch的时候，如果遇到nested对象，并且想对nested对象进行分组统计的话，可以按照如下方式进行处理。 第一个代码段落分别是 mapping，第二个代码段是请求命令，第三个代码段是分组结果。
"mathScoreNestedList": {
"type": "nested",
"properties": {
"score": {
"type": "integer"
},
"halfYear": {
"type": "keyword"
}
}
},
{
"size": 0,
"aggs": {
"labels_nested": {
"nested": {
"path": "mathScoreNestedList"
},
"aggs": {
"nested_score": {
"terms": {
"field": "mathScoreNestedList.halfYear"
}
}
}
}
}
}
{
"took": 32,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 27,
"relation": "eq"
},
"max_score": null,
"hits": []
},
"aggregations": {
"labels_nested": {
"doc_count": 7,
"nested_score": {
"doc_count_error_upper_bound": 0,
"sum_other_doc_count": 0,
"buckets": [
{
"key": "202201",
"doc_count": 5
},
{
"key": "202207",
"doc_count": 2
}
]
}
}
}
}
