2013-08-28 19:24:34 -04:00
[[search-request-sort]]
=== Sort
Allows to add one or more sort on specific fields. Each sort can be
reversed as well. The sort is defined on a per field level, with special
2015-08-24 09:37:30 -04:00
field name for `_score` to sort by score, and `_doc` to sort by index order.
2013-08-28 19:24:34 -04:00
2016-05-17 09:14:37 -04:00
Assuming the following index mapping:
[source,js]
--------------------------------------------------
PUT /my_index
{
"mappings": {
"my_type": {
"properties": {
"post_date": { "type": "date" },
2016-07-15 16:02:07 -04:00
"user": {
2016-05-18 04:03:38 -04:00
"type": "keyword"
2016-05-17 09:14:37 -04:00
},
"name": {
2016-05-18 04:03:38 -04:00
"type": "keyword"
2016-05-17 09:14:37 -04:00
},
"age": { "type": "integer" }
}
}
}
}
--------------------------------------------------
// CONSOLE
2013-08-28 19:24:34 -04:00
[source,js]
--------------------------------------------------
2016-05-17 09:14:37 -04:00
GET /my_index/my_type/_search
2013-08-28 19:24:34 -04:00
{
"sort" : [
2013-09-04 16:06:38 -04:00
{ "post_date" : {"order" : "asc"}},
2013-08-28 19:24:34 -04:00
"user",
{ "name" : "desc" },
{ "age" : "desc" },
"_score"
],
"query" : {
"term" : { "user" : "kimchy" }
}
}
--------------------------------------------------
2016-05-17 09:14:37 -04:00
// CONSOLE
// TEST[continued]
2013-08-28 19:24:34 -04:00
2015-08-24 09:37:30 -04:00
NOTE: `_doc` has no real use-case besides being the most efficient sort order.
So if you don't care about the order in which documents are returned, then you
should sort by `_doc`. This especially helps when <<search-request-scroll,scrolling>>.
2013-08-28 19:24:34 -04:00
==== Sort Values
The sort values for each document returned are also returned as part of
the response.
2015-02-19 12:04:14 -05:00
==== Sort Order
The `order` option can have the following values:
[horizontal]
`asc`:: Sort in ascending order
`desc`:: Sort in descending order
The order defaults to `desc` when sorting on the `_score`, and defaults
to `asc` when sorting on anything else.
2013-08-28 19:24:34 -04:00
==== Sort mode option
2013-09-03 15:27:49 -04:00
Elasticsearch supports sorting by array or multi-valued fields. The `mode` option
2013-08-28 19:24:34 -04:00
controls what array value is picked for sorting the document it belongs
2013-09-04 16:06:38 -04:00
to. The `mode` option can have the following values:
2013-08-28 19:24:34 -04:00
[horizontal]
2013-09-04 16:06:38 -04:00
`min`:: Pick the lowest value.
2013-08-28 19:24:34 -04:00
`max`:: Pick the highest value.
`sum`:: Use the sum of all values as sort value. Only applicable for
2013-09-04 16:06:38 -04:00
number based array fields.
2013-08-28 19:24:34 -04:00
`avg`:: Use the average of all values as sort value. Only applicable
for number based array fields.
2015-05-11 12:32:16 -04:00
`median`:: Use the median of all values as sort value. Only applicable
for number based array fields.
2013-08-28 19:24:34 -04:00
===== Sort mode example usage
In the example below the field price has multiple prices per document.
2016-05-17 09:14:37 -04:00
In this case the result hits will be sorted by price ascending based on
2013-08-28 19:24:34 -04:00
the average price per document.
[source,js]
--------------------------------------------------
2016-05-18 04:03:38 -04:00
PUT /my_index/my_type/1?refresh
2016-05-17 09:14:37 -04:00
{
"product": "chocolate",
"price": [20, 4]
}
2016-07-15 16:02:07 -04:00
POST /_search
2016-05-17 09:14:37 -04:00
{
2013-08-28 19:24:34 -04:00
"query" : {
2016-05-17 09:14:37 -04:00
"term" : { "product" : "chocolate" }
2013-08-28 19:24:34 -04:00
},
"sort" : [
{"price" : {"order" : "asc", "mode" : "avg"}}
]
2016-05-17 09:14:37 -04:00
}
2013-08-28 19:24:34 -04:00
--------------------------------------------------
2016-05-17 09:14:37 -04:00
// CONSOLE
2013-08-28 19:24:34 -04:00
2015-08-06 11:24:29 -04:00
[[nested-sorting]]
2013-08-28 19:24:34 -04:00
==== Sorting within nested objects.
2013-09-03 15:27:49 -04:00
Elasticsearch also supports sorting by
2013-08-28 19:24:34 -04:00
fields that are inside one or more nested objects. The sorting by nested
field support has the following parameters on top of the already
2013-09-04 16:06:38 -04:00
existing sort options:
2013-08-28 19:24:34 -04:00
2013-09-04 16:06:38 -04:00
`nested_path`::
2015-11-19 12:20:48 -05:00
Defines on which nested object to sort. The actual
sort field must be a direct field inside this nested object.
When sorting by nested field, this field is mandatory.
2013-08-28 19:24:34 -04:00
2013-09-04 16:06:38 -04:00
`nested_filter`::
2015-11-19 12:20:48 -05:00
A filter that the inner objects inside the nested path
2013-08-28 19:24:34 -04:00
should match with in order for its field values to be taken into account
by sorting. Common case is to repeat the query / filter inside the
nested filter or query. By default no `nested_filter` is active.
===== Nested sorting example
2015-09-09 12:39:23 -04:00
In the below example `offer` is a field of type `nested`.
2015-11-19 12:20:48 -05:00
The `nested_path` needs to be specified; otherwise, elasticsearch doesn't know on what nested level sort values need to be captured.
2013-08-28 19:24:34 -04:00
[source,js]
--------------------------------------------------
2016-05-17 09:14:37 -04:00
POST /_search
{
2013-08-28 19:24:34 -04:00
"query" : {
2016-05-17 09:14:37 -04:00
"term" : { "product" : "chocolate" }
2013-08-28 19:24:34 -04:00
},
"sort" : [
{
"offer.price" : {
"mode" : "avg",
"order" : "asc",
2015-09-09 12:39:23 -04:00
"nested_path" : "offer",
2013-08-28 19:24:34 -04:00
"nested_filter" : {
"term" : { "offer.color" : "blue" }
}
}
}
]
2016-05-17 09:14:37 -04:00
}
2013-08-28 19:24:34 -04:00
--------------------------------------------------
2016-05-17 09:14:37 -04:00
// CONSOLE
2013-08-28 19:24:34 -04:00
2013-09-03 15:27:49 -04:00
Nested sorting is also supported when sorting by
2013-08-28 19:24:34 -04:00
scripts and sorting by geo distance.
==== Missing Values
2013-09-04 16:06:38 -04:00
The `missing` parameter specifies how docs which are missing
the field should be treated: The `missing` value can be
set to `_last`, `_first`, or a custom value (that
2013-08-28 19:24:34 -04:00
will be used for missing docs as the sort value). For example:
[source,js]
--------------------------------------------------
2016-05-17 09:14:37 -04:00
GET /_search
2013-08-28 19:24:34 -04:00
{
"sort" : [
2016-05-17 09:14:37 -04:00
{ "price" : {"missing" : "_last"} }
2013-08-28 19:24:34 -04:00
],
"query" : {
2016-05-17 09:14:37 -04:00
"term" : { "product" : "chocolate" }
2013-08-28 19:24:34 -04:00
}
}
--------------------------------------------------
2016-05-17 09:14:37 -04:00
// CONSOLE
2013-08-28 19:24:34 -04:00
2013-09-03 15:27:49 -04:00
NOTE: If a nested inner object doesn't match with
2013-08-28 19:24:34 -04:00
the `nested_filter` then a missing value is used.
==== Ignoring Unmapped Fields
By default, the search request will fail if there is no mapping
2014-07-25 10:57:53 -04:00
associated with a field. The `unmapped_type` option allows to ignore
fields that have no mapping and not sort by them. The value of this
parameter is used to determine what sort values to emit. Here is an
example of how it can be used:
2013-08-28 19:24:34 -04:00
[source,js]
--------------------------------------------------
2016-05-17 09:14:37 -04:00
GET /_search
2013-08-28 19:24:34 -04:00
{
"sort" : [
2016-05-17 09:14:37 -04:00
{ "price" : {"unmapped_type" : "long"} }
2013-08-28 19:24:34 -04:00
],
"query" : {
2016-05-17 09:14:37 -04:00
"term" : { "product" : "chocolate" }
2013-08-28 19:24:34 -04:00
}
}
--------------------------------------------------
2016-05-17 09:14:37 -04:00
// CONSOLE
2013-08-28 19:24:34 -04:00
2014-07-25 10:57:53 -04:00
If any of the indices that are queried doesn't have a mapping for `price`
then Elasticsearch will handle it as if there was a mapping of type
`long`, with all documents in this index having no value for this field.
2015-08-06 11:24:29 -04:00
[[geo-sorting]]
2013-08-28 19:24:34 -04:00
==== Geo Distance Sorting
Allow to sort by `_geo_distance`. Here is an example:
[source,js]
--------------------------------------------------
2016-05-17 09:14:37 -04:00
GET /_search
2013-08-28 19:24:34 -04:00
{
"sort" : [
{
"_geo_distance" : {
"pin.location" : [-70, 40],
"order" : "asc",
2014-08-14 08:54:34 -04:00
"unit" : "km",
"mode" : "min",
"distance_type" : "sloppy_arc"
2013-08-28 19:24:34 -04:00
}
}
],
"query" : {
"term" : { "user" : "kimchy" }
}
}
--------------------------------------------------
2016-05-17 09:14:37 -04:00
// CONSOLE
2013-08-28 19:24:34 -04:00
2014-08-14 08:54:34 -04:00
`distance_type`::
2015-05-05 04:03:15 -04:00
How to compute the distance. Can either be `sloppy_arc` (default), `arc` (slightly more precise but significantly slower) or `plane` (faster, but inaccurate on long distances and close to the poles).
2014-08-14 08:54:34 -04:00
2013-08-28 19:24:34 -04:00
Note: the geo distance sorting supports `sort_mode` options: `min`,
`max` and `avg`.
The following formats are supported in providing the coordinates:
===== Lat Lon as Properties
[source,js]
--------------------------------------------------
2016-05-17 09:14:37 -04:00
GET /_search
2013-08-28 19:24:34 -04:00
{
"sort" : [
{
"_geo_distance" : {
"pin.location" : {
"lat" : 40,
"lon" : -70
},
"order" : "asc",
"unit" : "km"
}
}
],
"query" : {
"term" : { "user" : "kimchy" }
}
}
--------------------------------------------------
2016-05-17 09:14:37 -04:00
// CONSOLE
2013-08-28 19:24:34 -04:00
===== Lat Lon as String
Format in `lat,lon`.
[source,js]
--------------------------------------------------
2016-05-17 09:14:37 -04:00
GET /_search
2013-08-28 19:24:34 -04:00
{
"sort" : [
{
"_geo_distance" : {
2015-12-16 06:52:28 -05:00
"pin.location" : "40,-70",
2013-08-28 19:24:34 -04:00
"order" : "asc",
"unit" : "km"
}
}
],
"query" : {
"term" : { "user" : "kimchy" }
}
}
--------------------------------------------------
2016-05-17 09:14:37 -04:00
// CONSOLE
2013-08-28 19:24:34 -04:00
===== Geohash
[source,js]
--------------------------------------------------
2016-05-17 09:14:37 -04:00
GET /_search
2013-08-28 19:24:34 -04:00
{
"sort" : [
{
"_geo_distance" : {
"pin.location" : "drm3btev3e86",
"order" : "asc",
"unit" : "km"
}
}
],
"query" : {
"term" : { "user" : "kimchy" }
}
}
--------------------------------------------------
2016-05-17 09:14:37 -04:00
// CONSOLE
2013-08-28 19:24:34 -04:00
===== Lat Lon as Array
Format in `[lon, lat]`, note, the order of lon/lat here in order to
conform with http://geojson.org/[GeoJSON].
[source,js]
--------------------------------------------------
2016-05-17 09:14:37 -04:00
GET /_search
2013-08-28 19:24:34 -04:00
{
"sort" : [
{
"_geo_distance" : {
"pin.location" : [-70, 40],
"order" : "asc",
"unit" : "km"
}
}
],
"query" : {
"term" : { "user" : "kimchy" }
}
}
--------------------------------------------------
2016-05-17 09:14:37 -04:00
// CONSOLE
2013-08-28 19:24:34 -04:00
2014-07-30 12:48:36 -04:00
==== Multiple reference points
Multiple geo points can be passed as an array containing any `geo_point` format, for example
2014-08-08 05:25:14 -04:00
[source,js]
--------------------------------------------------
2014-07-30 12:48:36 -04:00
"pin.location" : [[-70, 40], [-71, 42]]
2015-12-16 06:52:28 -05:00
"pin.location" : [{"lat": 40, "lon": -70}, {"lat": 42, "lon": -71}]
2014-08-08 05:25:14 -04:00
--------------------------------------------------
2014-07-30 12:48:36 -04:00
and so forth.
2014-08-14 08:54:34 -04:00
The final distance for a document will then be `min`/`max`/`avg` (defined via `mode`) distance of all points contained in the document to all points given in the sort request.
2014-07-30 12:48:36 -04:00
2013-08-28 19:24:34 -04:00
==== Script Based Sorting
Allow to sort based on custom scripts, here is an example:
[source,js]
--------------------------------------------------
2016-05-17 09:14:37 -04:00
GET /_search
2013-08-28 19:24:34 -04:00
{
"query" : {
2016-05-17 09:14:37 -04:00
"term" : { "user" : "kimchy" }
2013-08-28 19:24:34 -04:00
},
"sort" : {
"_script" : {
"type" : "number",
2015-05-12 05:37:22 -04:00
"script" : {
2016-06-27 09:55:16 -04:00
"lang": "painless",
"inline": "doc['field_name'].value * params.factor",
2015-05-12 05:37:22 -04:00
"params" : {
"factor" : 1.1
}
2013-08-28 19:24:34 -04:00
},
"order" : "asc"
}
}
}
--------------------------------------------------
2016-05-17 09:14:37 -04:00
// CONSOLE
2013-08-28 19:24:34 -04:00
==== Track Scores
When sorting on a field, scores are not computed. By setting
`track_scores` to true, scores will still be computed and tracked.
[source,js]
--------------------------------------------------
2016-05-17 09:14:37 -04:00
GET /_search
2013-08-28 19:24:34 -04:00
{
"track_scores": true,
"sort" : [
2016-03-24 07:04:31 -04:00
{ "post_date" : {"order" : "desc"} },
2013-08-28 19:24:34 -04:00
{ "name" : "desc" },
{ "age" : "desc" }
],
"query" : {
"term" : { "user" : "kimchy" }
}
}
--------------------------------------------------
2016-05-17 09:14:37 -04:00
// CONSOLE
2013-08-28 19:24:34 -04:00
==== Memory Considerations
When sorting, the relevant sorted field values are loaded into memory.
This means that per shard, there should be enough memory to contain
them. For string based types, the field sorted on should not be analyzed
/ tokenized. For numeric types, if possible, it is recommended to
2015-06-08 12:07:52 -04:00
explicitly set the type to narrower types (like `short`, `integer` and
2013-08-28 19:24:34 -04:00
`float`).