マルチバリューフィールド（Multivalued fields）

ES|QL マルチバリューフィールド

ES|QL はマルチバリューフィールドからの読み取りが可能です:

Python

resp = client.bulk(
   index="mv",
   refresh=True,
   operations=[
   {
   "index": {}
   },
   {
   "a": 1,
   "b": [
   2,
   1
   ]
   },
   {
   "index": {}
   },
   {
   "a": 2,
   "b": 3
   }
   ],
)
print(resp)
resp1 = client.esql.query(
   query="FROM mv | LIMIT 2",
)
print(resp1)

Js

const response = await client.bulk({
  index: "mv",
  refresh: "true",
  operations: [
   {
   index: {},
   },
   {
   a: 1,
   b: [2, 1],
   },
   {
   index: {},
   },
   {
   a: 2,
   b: 3,
   },
  ],
});
console.log(response);
const response1 = await client.esql.query({
  query: "FROM mv | LIMIT 2",
});
console.log(response1);

Console

POST /mv/_bulk?refresh
{ "index" : {} }
{ "a": 1, "b": [2, 1] }
{ "index" : {} }
{ "a": 2, "b": 3 }
POST /_query
{
  "query": "FROM mv | LIMIT 2"
}

マルチバリューフィールドは JSON 配列として返されます:

Console-Result

{
  "columns": [
   { "name": "a", "type": "long"},
   { "name": "b", "type": "long"}
  ],
  "values": [
   [1, [1, 2]],
   [2,      3]
  ]
}

マルチバリューフィールド内の値の相対的な順序は未定義です。通常は昇順になりますが、それに依存しないでください。

重複値

keyword のような一部のフィールドタイプは、書き込み時に重複値を削除します:

Python

resp = client.indices.create(
   index="mv",
   mappings={
   "properties": {
   "b": {
   "type": "keyword"
   }
   }
   },
)
print(resp)
resp1 = client.bulk(
   index="mv",
   refresh=True,
   operations=[
   {
   "index": {}
   },
   {
   "a": 1,
   "b": [
   "foo",
   "foo",
   "bar"
   ]
   },
   {
   "index": {}
   },
   {
   "a": 2,
   "b": [
   "bar",
   "bar"
   ]
   }
   ],
)
print(resp1)
resp2 = client.esql.query(
   query="FROM mv | LIMIT 2",
)
print(resp2)

Js

const response = await client.indices.create({
  index: "mv",
  mappings: {
   properties: {
   b: {
   type: "keyword",
   },
   },
  },
});
console.log(response);
const response1 = await client.bulk({
  index: "mv",
  refresh: "true",
  operations: [
   {
   index: {},
   },
   {
   a: 1,
   b: ["foo", "foo", "bar"],
   },
   {
   index: {},
   },
   {
   a: 2,
   b: ["bar", "bar"],
   },
  ],
});
console.log(response1);
const response2 = await client.esql.query({
  query: "FROM mv | LIMIT 2",
});
console.log(response2);

Console

PUT /mv
{
  "mappings": {
   "properties": {
   "b": {"type": "keyword"}
   }
  }
}
POST /mv/_bulk?refresh
{ "index" : {} }
{ "a": 1, "b": ["foo", "foo", "bar"] }
{ "index" : {} }
{ "a": 2, "b": ["bar", "bar"] }
POST /_query
{
  "query": "FROM mv | LIMIT 2"
}

そして、ES|QL はその削除を確認します:

Console-Result

{
  "columns": [
   { "name": "a", "type": "long"},
   { "name": "b", "type": "keyword"}
  ],
  "values": [
   [1, ["bar", "foo"]],
   [2,          "bar"]
  ]
}

しかし、long のような他のタイプは重複を削除しません。

Python

resp = client.indices.create(
   index="mv",
   mappings={
   "properties": {
   "b": {
   "type": "long"
   }
   }
   },
)
print(resp)
resp1 = client.bulk(
   index="mv",
   refresh=True,
   operations=[
   {
   "index": {}
   },
   {
   "a": 1,
   "b": [
   2,
   2,
   1
   ]
   },
   {
   "index": {}
   },
   {
   "a": 2,
   "b": [
   1,
   1
   ]
   }
   ],
)
print(resp1)
resp2 = client.esql.query(
   query="FROM mv | LIMIT 2",
)
print(resp2)

Js

const response = await client.indices.create({
  index: "mv",
  mappings: {
   properties: {
   b: {
   type: "long",
   },
   },
  },
});
console.log(response);
const response1 = await client.bulk({
  index: "mv",
  refresh: "true",
  operations: [
   {
   index: {},
   },
   {
   a: 1,
   b: [2, 2, 1],
   },
   {
   index: {},
   },
   {
   a: 2,
   b: [1, 1],
   },
  ],
});
console.log(response1);
const response2 = await client.esql.query({
  query: "FROM mv | LIMIT 2",
});
console.log(response2);

Console

PUT /mv
{
  "mappings": {
   "properties": {
   "b": {"type": "long"}
   }
  }
}
POST /mv/_bulk?refresh
{ "index" : {} }
{ "a": 1, "b": [2, 2, 1] }
{ "index" : {} }
{ "a": 2, "b": [1, 1] }
POST /_query
{
  "query": "FROM mv | LIMIT 2"
}

そして、ES|QL もそれを確認します:

Console-Result

{
  "columns": [
   { "name": "a", "type": "long"},
   { "name": "b", "type": "long"}
  ],
  "values": [
   [1, [1, 2, 2]],
   [2,    [1, 1]]
  ]
}

これはすべてストレージ層でのことです。重複した long を保存し、それを文字列に変換すると、重複は残ります:

Python

resp = client.indices.create(
   index="mv",
   mappings={
   "properties": {
   "b": {
   "type": "long"
   }
   }
   },
)
print(resp)
resp1 = client.bulk(
   index="mv",
   refresh=True,
   operations=[
   {
   "index": {}
   },
   {
   "a": 1,
   "b": [
   2,
   2,
   1
   ]
   },
   {
   "index": {}
   },
   {
   "a": 2,
   "b": [
   1,
   1
   ]
   }
   ],
)
print(resp1)
resp2 = client.esql.query(
   query="FROM mv | EVAL b=TO_STRING(b) | LIMIT 2",
)
print(resp2)

Js

const response = await client.indices.create({
  index: "mv",
  mappings: {
   properties: {
   b: {
   type: "long",
   },
   },
  },
});
console.log(response);
const response1 = await client.bulk({
  index: "mv",
  refresh: "true",
  operations: [
   {
   index: {},
   },
   {
   a: 1,
   b: [2, 2, 1],
   },
   {
   index: {},
   },
   {
   a: 2,
   b: [1, 1],
   },
  ],
});
console.log(response1);
const response2 = await client.esql.query({
  query: "FROM mv | EVAL b=TO_STRING(b) | LIMIT 2",
});
console.log(response2);

Console

PUT /mv
{
  "mappings": {
   "properties": {
   "b": {"type": "long"}
   }
  }
}
POST /mv/_bulk?refresh
{ "index" : {} }
{ "a": 1, "b": [2, 2, 1] }
{ "index" : {} }
{ "a": 2, "b": [1, 1] }
POST /_query
{
  "query": "FROM mv | EVAL b=TO_STRING(b) | LIMIT 2"
}

Console-Result

{
  "columns": [
   { "name": "a", "type": "long"},
   { "name": "b", "type": "keyword"}
  ],
  "values": [
   [1, ["1", "2", "2"]],
   [2,      ["1", "1"]]
  ]
}

関数

特に文書化されていない限り、関数はマルチバリューフィールドに適用されると null を返します。

Python

resp = client.bulk(
   index="mv",
   refresh=True,
   operations=[
   {
   "index": {}
   },
   {
   "a": 1,
   "b": [
   2,
   1
   ]
   },
   {
   "index": {}
   },
   {
   "a": 2,
   "b": 3
   }
   ],
)
print(resp)

Ruby

response = client.bulk(
  index: 'mv',
  refresh: true,
  body: [
   {
   index: {}
   },
   {
   a: 1,
   b: [
   2,
   1
   ]
   },
   {
   index: {}
   },
   {
   a: 2,
   b: 3
   }
  ]
)
puts response

Js

const response = await client.bulk({
  index: "mv",
  refresh: "true",
  operations: [
   {
   index: {},
   },
   {
   a: 1,
   b: [2, 1],
   },
   {
   index: {},
   },
   {
   a: 2,
   b: 3,
   },
  ],
});
console.log(response);

Console

POST /mv/_bulk?refresh
{ "index" : {} }
{ "a": 1, "b": [2, 1] }
{ "index" : {} }
{ "a": 2, "b": 3 }

Python

resp = client.esql.query(
   query="FROM mv | EVAL b + 2, a + b | LIMIT 4",
)
print(resp)

Js

const response = await client.esql.query({
  query: "FROM mv | EVAL b + 2, a + b | LIMIT 4",
});
console.log(response);

Console

POST /_query
{
  "query": "FROM mv | EVAL b + 2, a + b | LIMIT 4"
}

Console-Result

{
  "columns": [
   { "name": "a",   "type": "long"},
   { "name": "b",   "type": "long"},
   { "name": "b + 2", "type": "long"},
   { "name": "a + b", "type": "long"}
  ],
  "values": [
   [1, [1, 2], null, null],
   [2,      3,    5,    5]
  ]
}

この制限を回避するには、フィールドを次のいずれかの単一値に変換します:

• MV_AVG
• MV_CONCAT
• MV_COUNT
• MV_MAX
• MV_MEDIAN
• MV_MIN
• MV_SUM

Python

resp = client.esql.query(
   query="FROM mv | EVAL b=MV_MIN(b) | EVAL b + 2, a + b | LIMIT 4",
)
print(resp)

Js

const response = await client.esql.query({
  query: "FROM mv | EVAL b=MV_MIN(b) | EVAL b + 2, a + b | LIMIT 4",
});
console.log(response);

Console

POST /_query
{
  "query": "FROM mv | EVAL b=MV_MIN(b) | EVAL b + 2, a + b | LIMIT 4"
}

Console-Result

{
  "columns": [
   { "name": "a",   "type": "long"},
   { "name": "b",   "type": "long"},
   { "name": "b + 2", "type": "long"},
   { "name": "a + b", "type": "long"}
  ],
  "values": [
   [1, 1, 3, 2],
   [2, 3, 5, 5]
  ]
}