Docs: Clarified the purpose of the parent_id query

This commit is contained in:
Clinton Gormley 2016-04-14 11:25:26 +02:00
parent c22d93f7cc
commit acec464eb8
1 changed files with 24 additions and 12 deletions

View File

@ -3,19 +3,33 @@
added[5.0.0]
The `parent_id` query can be used to find a child document pointing to a particular parent id.
The actual underlying Lucene field that is used to store to what parent id a child document is referring to
is determined by the child type's `_parent` field. This query helps by selecting the right field based
on the specified child type. Example:
The `parent_id` query can be used to find child documents which belong to a particular parent:
[source,js]
--------------------------------------------------
{
"parent_id" : {
"type" : "blog_tag",
"id" : "1"
"parent_id" : {
"type" : "blog_tag",
"id" : "1"
}
}
--------------------------------------------------
The above is functionally equivalent to using the following
<<query-dsl-has-parent-query, `has_parent`>> query, but performs
better as it does not need to do a join:
[source,js]
--------------------------------------------------
{
"has_parent": {
"type": "blog",
"query": {
"term": {
"_id": "1"
}
}
}
}
--------------------------------------------------
@ -24,8 +38,6 @@ on the specified child type. Example:
This query has two required parameters:
[horizontal]
`type`::
The child type. This must be a type with `_parent` field.
`type`:: The **child** type. This must be a type with `_parent` field.
`id`::
The required parent id select documents must referrer to.
`id`:: The required parent id select documents must referrer to.