目录
为什么要一定要为子文档指定routing
The routing value is mandatory because parent and child documents must be indexed on the same shard
A document is routed to a particular shard in an index using the following formula:
shard_num = hash(_routing) % num_primary_shards
The default value used for _routing is the document’s _id.
所以在写入数据的时候使用父文档的_id来作为routing的值，来保证子文档和父文档在同一个分片中。
索引定义
{
"mappings": {
"test_20191127": {
"properties": {
"created_at": {
"type": "date",
"format": "yyyy-MM-dd HH:mm:ss"
},
"ip": {
"type": "keyword",
"normalizer": "standard_lowercase"
},
"cve_id": {
"type": "keyword",
"normalizer": "standard_lowercase"
},
"apply": {
"type": "keyword",
"normalizer": "standard_lowercase"
},
"join_field": {
"type": "join",
"relations": {
"ip": [
"cve",
"apply"
]
}
}
}
}
},
"settings": {
"index": {
"search": {
"slowlog": {
"threshold": {
"fetch": {
"warn": "1s",
"trace": "200ms",
"debug": "500ms",
"info": "800ms"
},
"query": {
"warn": "10s",
"trace": "500ms",
"debug": "2s",
"info": "5s"
}
}
}
},
"number_of_shards": "10",
"max_result_window": "10000000",
"mapper": {
"dynamic": "false"
},
"unassigned": {
"node_left": {
"delayed_timeout": "1m"
}
},
"analysis": {
"normalizer": {
"standard_lowercase": {
"filter": [
"lowercase"
],
"type": "custom"
}
}
},
"number_of_replicas": "0"
}
}
}写入数据
$create = \Elasticsearch\ClientBuilder::create();
$create->setHosts([['host' => "10.249.43.93", 'user' => 'xxxx', 'pass' => 'xxxx']]);
$client = $create->build();
for ($i = 1; $i <= 10; $i++) {
$id = $i;
$parent = [
'ip' => $i,
"join_field" => 'ip',
];
$cve = [
'ip' => $i,
"join_field" => ['name' => 'cve', "parent" => $id],
"cve_id" => 'CVE-' . $i,
];
$apply = [
'ip' => $i,
"join_field" => ['name' => 'apply', "parent" => $id],
"apply" => 'APPLY-' . $i,
];
/**
* $params['index'] = (string) The name of the index (Required)
* ['type'] = (string) The type of the document (Required)
* ['id'] = (string) Specific document ID (when the POST method is used)
* ['consistency'] = (enum) Explicit write consistency setting for the operation
* ['op_type'] = (enum) Explicit operation type
* ['parent'] = (string) ID of the parent document
* ['refresh'] = (boolean) Refresh the index after performing the operation
* ['replication'] = (enum) Specific replication type
* ['routing'] = (string) Specific routing value
* ['timeout'] = (time) Explicit operation timeout
* ['timestamp'] = (time) Explicit timestamp for the document
* ['ttl'] = (duration) Expiration time for the document
* ['version'] = (number) Explicit version number for concurrency control
* ['version_type'] = (enum) Specific version type
* ['body'] = (array) The document
*
*/
//插入父文档
$client->index(['index' => 'test_20191127', 'type' => 'test_20191127', 'id' => $id, 'body' => $parent]);
//插入子文档，并指定routing为父文档ID
$client->index(['index' => 'test_20191127', 'type' => 'test_20191127', 'routing' => $id, 'body' => $cve]);
$client->index(['index' => 'test_20191127', 'type' => 'test_20191127', 'routing' => $id, 'body' => $apply]);
}数据概览
{
"took": 0,
"timed_out": false,
"_shards": {
"total": 10,
"successful": 10,
"skipped": 0,
"failed": 0
},
"hits": {
"total": 30,
"max_score": 1.0,
"hits": [
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "5",
"_score": 1.0,
"_source": {
"ip": 5,
"join_field": "ip"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "UX7Hq24B5GVBR06e807I",
"_score": 1.0,
"_routing": "5",
"_source": {
"ip": 5,
"join_field": {
"name": "cve",
"parent": 5
},
"cve_id": "CVE-5"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "Un7Hq24B5GVBR06e807Y",
"_score": 1.0,
"_routing": "5",
"_source": {
"ip": 5,
"join_field": {
"name": "apply",
"parent": 5
},
"apply": "APPLY-5"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "10",
"_score": 1.0,
"_source": {
"ip": 10,
"join_field": "ip"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "W37Hq24B5GVBR06e9E5v",
"_score": 1.0,
"_routing": "10",
"_source": {
"ip": 10,
"join_field": {
"name": "cve",
"parent": 10
},
"cve_id": "CVE-10"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "XH7Hq24B5GVBR06e9E54",
"_score": 1.0,
"_routing": "10",
"_source": {
"ip": 10,
"join_field": {
"name": "apply",
"parent": 10
},
"apply": "APPLY-10"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "1",
"_score": 1.0,
"_source": {
"ip": 1,
"join_field": "ip"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "SX7Hq24B5GVBR06e804m",
"_score": 1.0,
"_routing": "1",
"_source": {
"ip": 1,
"join_field": {
"name": "cve",
"parent": 1
},
"cve_id": "CVE-1"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "Sn7Hq24B5GVBR06e8041",
"_score": 1.0,
"_routing": "1",
"_source": {
"ip": 1,
"join_field": {
"name": "apply",
"parent": 1
},
"apply": "APPLY-1"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "3",
"_score": 1.0,
"_source": {
"ip": 3,
"join_field": "ip"
}
}
]
}
}查询所有父文档
要确保每个父文档都拥有这个子文档才可以， 查询语法如下：
{
"query": {
"has_child" : {
"type" : "apply",
"query" : {
"match_all":{}
}
}
}
}结果如下：
{
"took": 0,
"timed_out": false,
"_shards": {
"total": 10,
"successful": 10,
"skipped": 0,
"failed": 0
},
"hits": {
"total": 10,
"max_score": 1.0,
"hits": [
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "5",
"_score": 1.0,
"_source": {
"ip": 5,
"join_field": "ip"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "10",
"_score": 1.0,
"_source": {
"ip": 10,
"join_field": "ip"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "1",
"_score": 1.0,
"_source": {
"ip": 1,
"join_field": "ip"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "3",
"_score": 1.0,
"_source": {
"ip": 3,
"join_field": "ip"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "8",
"_score": 1.0,
"_source": {
"ip": 8,
"join_field": "ip"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "9",
"_score": 1.0,
"_source": {
"ip": 9,
"join_field": "ip"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "2",
"_score": 1.0,
"_source": {
"ip": 2,
"join_field": "ip"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "4",
"_score": 1.0,
"_source": {
"ip": 4,
"join_field": "ip"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "6",
"_score": 1.0,
"_source": {
"ip": 6,
"join_field": "ip"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "7",
"_score": 1.0,
"_source": {
"ip": 7,
"join_field": "ip"
}
}
]
}
}
根据子文档过滤得出父文档
查询语法：
{
"query": {
"has_child" : {
"type" : "apply",
"query" : {
"term":{
"apply":"APPLY-5"
}
}
}
}
}
结果：
{
"took": 0,
"timed_out": false,
"_shards": {
"total": 10,
"successful": 10,
"skipped": 0,
"failed": 0
},
"hits": {
"total": 1,
"max_score": 1.0,
"hits": [
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "5",
"_score": 1.0,
"_source": {
"ip": 5,
"join_field": "ip"
}
}
]
}
}
根据父文档过滤得出子文档
查询如下
{
"size":100,
"query": {
"has_parent" : {
"parent_type" : "ip",
"query" : {
"term":{
"ip":"5"
}
}
}
}
}
结果如下：
{
"took": 0,
"timed_out": false,
"_shards": {
"total": 10,
"successful": 10,
"skipped": 0,
"failed": 0
},
"hits": {
"total": 2,
"max_score": 1.0,
"hits": [
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "6n73q24B5GVBR06e7k_w",
"_score": 1.0,
"_routing": "5",
"_source": {
"type": "cve",
"ip": 5,
"join_field": {
"name": "cve",
"parent": 5
},
"cve_id": "CVE-5"
}
},
{
"_index": "test_20191127",
"_type": "test_20191127",
"_id": "6373q24B5GVBR06e708C",
"_score": 1.0,
"_routing": "5",
"_source": {
"type": "apply",
"ip": 5,
"join_field": {
"name": "apply",
"parent": 5
},
"apply": "APPLY-5"
}
}
]
}
}