一般可能會比較 solr 有 pivot table 的這個用法，
其實在elastic search 裡面可以使用 facet search 做替代。
在 1.x以後 加入新的 Aggregations ，我們可以使用他來產生 pivot table。
ex:
1.先index 一些 records
curl -XPOST 'http://localhost:9200/twitter/tweet/' -d '{ "user": "kimchy", "post_date": "2009-11-15T14:12:12", "message": "trying out Elasticsearch", "ip": "123.0.0.2", "groups": "B" }'
{
"user": "kimchy",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.2",
"groups": "A"
}
{
"user": "pc",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.5",
"groups": "C"
}
{
"user": "kimchy",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.2",
"groups": "C"
}
{ "user": "kimchy", "post_date": "2009-11-15T14:12:12", "message": "trying out Elasticsearch", "ip": "123.0.0.1", "groups": "A" }
{
"user": "pc",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.5",
"groups": "A"
}
{
"user": "pc",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.2",
"groups": "C"
}
{
"user": "kimchy2",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.2",
"groups": "A"
}
curl -XPOST http://localhost:9200/twitter/_search?pretty=true -D '
{
"aggregations": {
"maincategory": {
"aggregations": {
"main2category": {
"aggregations": {
"main3category": {
"terms": {
"field": "user"
}
}
},
"terms": {
"field": "ip"
}
}
},
"terms": {
"field": "groups"
}
}
}
}
'
得到的結果
{
"took": 66,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"failed": 0
},
"hits": {
"total": 8,
"max_score": 1,
"hits": [
{
"_index": "twitter",
"_type": "tweet",
"_id": "XgQFw9ncTf2LhIOdb3Jw8g",
"_score": 1,
"_source": {
"user": "kimchy",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.2",
"groups": "B"
}
},
{
"_index": "twitter",
"_type": "tweet",
"_id": "BEN-DAVLSpacVvViGl1RxQ",
"_score": 1,
"_source": {
"user": "kimchy",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.2",
"groups": "A"
}
},
{
"_index": "twitter",
"_type": "tweet",
"_id": "GByKrOpJRnCeWtVpvKo1bA",
"_score": 1,
"_source": {
"user": "pc",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.5",
"groups": "A"
}
},
{
"_index": "twitter",
"_type": "tweet",
"_id": "5ExMEa4TQMKyLyBR-61CUg",
"_score": 1,
"_source": {
"user": "kimchy",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.1",
"groups": "A"
}
},
{
"_index": "twitter",
"_type": "tweet",
"_id": "gMHOz5DqRaiL15dIVh_nzA",
"_score": 1,
"_source": {
"user": "pc",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.2",
"groups": "C"
}
},
{
"_index": "twitter",
"_type": "tweet",
"_id": "XJHgowtKQ26zvUj2Md4b4g",
"_score": 1,
"_source": {
"user": "kimchy2",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.2",
"groups": "A"
}
},
{
"_index": "twitter",
"_type": "tweet",
"_id": "2JjnAJxrRJ6JPCNLhlgYfQ",
"_score": 1,
"_source": {
"user": "kimchy",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.2",
"groups": "C"
}
},
{
"_index": "twitter",
"_type": "tweet",
"_id": "_xPuE9bPTr2p2IKIpVDKSw",
"_score": 1,
"_source": {
"user": "pc",
"post_date": "2009-11-15T14:12:12",
"message": "trying out Elasticsearch",
"ip": "123.0.0.5",
"groups": "C"
}
}
]
},
"aggregations": {
"maincategory": {
"buckets": [
{
"key": "a",
"doc_count": 4,
"main2category": {
"buckets": [
{
"key": "123.0.0.2",
"doc_count": 2,
"main3category": {
"buckets": [
{
"key": "kimchy",
"doc_count": 1
},
{
"key": "kimchy2",
"doc_count": 1
}
]
}
},
{
"key": "123.0.0.1",
"doc_count": 1,
"main3category": {
"buckets": [
{
"key": "kimchy",
"doc_count": 1
}
]
}
},
{
"key": "123.0.0.5",
"doc_count": 1,
"main3category": {
"buckets": [
{
"key": "pc",
"doc_count": 1
}
]
}
}
]
}
},
{
"key": "c",
"doc_count": 3,
"main2category": {
"buckets": [
{
"key": "123.0.0.2",
"doc_count": 2,
"main3category": {
"buckets": [
{
"key": "kimchy",
"doc_count": 1
},
{
"key": "pc",
"doc_count": 1
}
]
}
},
{
"key": "123.0.0.5",
"doc_count": 1,
"main3category": {
"buckets": [
{
"key": "pc",
"doc_count": 1
}
]
}
}
]
}
},
{
"key": "b",
"doc_count": 1,
"main2category": {
"buckets": [
{
"key": "123.0.0.2",
"doc_count": 1,
"main3category": {
"buckets": [
{
"key": "kimchy",
"doc_count": 1
}
]
}
}
]
}
}
]
}
}
}
DevMynd Blog: GeoHash Grid Aggregation with ElasticSearch
http://www.devmynd.com/blog/2014-2-geohash-grid-aggregation-with-elasticsearch
Elasticsearch's New Aggregations
https://www.found.no/foundation/elasticsearch-aggregations/
Elasticsearch Aggregations Overview | Chris Simpson - Software Developer
http://chrissimpson.co.uk/elasticsearch-aggregations-overview.html
Aggregations
http://www.elasticsearch.org/guide/en/elasticsearch/reference/master/search-aggregations.html
沒有留言:
張貼留言