FT.SEARCH
FT.SEARCH index query [NOCONTENT] [VERBATIM] [NOSTOPWORDS] [WITHSCORES] [WITHPAYLOADS] [WITHSORTKEYS] [FILTER numeric_field min max [ FILTER numeric_field min max ...]] [GEOFILTER geo_field lon lat radius m | km | mi | ft [ GEOFILTER geo_field lon lat radius m | km | mi | ft ...]] [INKEYS count key [key ...]] [ INFIELDS count field [field ...]] [RETURN count identifier [AS property] [ identifier [AS property] ...]] [SUMMARIZE [ FIELDS count field [field ...]] [FRAGS num] [LEN fragsize] [SEPARATOR separator]] [HIGHLIGHT [ FIELDS count field [field ...]] [ TAGS open close]] [SLOP slop] [TIMEOUT timeout] [INORDER] [LANGUAGE language] [EXPANDER expander] [SCORER scorer] [EXPLAINSCORE] [PAYLOAD payload] [SORTBY sortby [ ASC | DESC] [WITHCOUNT]] [LIMIT offset num] [PARAMS nargs name value [ name value ...]] [DIALECT dialect]
- Available in:
- Redis Stack / Search 1.0.0
- Time complexity:
- O(N)
Search the index with a textual query, returning either documents or just ids
Required arguments
index
is index name. You must first create the index using FT.CREATE
.
query
is text query to search. If it's more than a single word, put it in quotes. Refer to Query syntax for more details.
Optional arguments
NOCONTENT
returns the document ids and not the content. This is useful if RediSearch is only an index on an external document collection.
VERBATIM
does not try to use stemming for query expansion but searches the query terms verbatim.
WITHSCORES
also returns the relative internal score of each document. This can be used to merge results from multiple instances.
WITHPAYLOADS
retrieves optional document payloads. See FT.CREATE
. The payloads follow the document id and, if WITHSCORES
is set, the scores.
WITHSORTKEYS
returns the value of the sorting key, right after the id and score and/or payload, if requested. This is usually not needed, and exists for distributed search coordination purposes. This option is relevant only if used in conjunction with SORTBY
.
FILTER numeric_attribute min max
limits results to those having numeric values ranging between min
and max
, if numeric_attribute is defined as a numeric attribute in FT.CREATE
.
min
and max
follow ZRANGE
syntax, and can be -inf
, +inf
, and use (
for exclusive ranges. Multiple numeric filters for different attributes are supported in one query.
GEOFILTER {geo_attribute} {lon} {lat} {radius} m|km|mi|ft
filter the results to a given radius
from lon
and lat
. Radius is given as a number and units. See GEORADIUS
for more details.
INKEYS {num} {attribute} ...
limits the result to a given set of keys specified in the list. The first argument must be the length of the list and greater than zero. Non-existent keys are ignored, unless all the keys are non-existent.
INFIELDS {num} {attribute} ...
filters the results to those appearing only in specific attributes of the document, like title
or URL
. You must include num
, which is the number of attributes you're filtering by. For example, if you request title
and URL
, then num
is 2.
RETURN {num} {identifier} AS {property} ...
limits the attributes returned from the document. num
is the number of attributes following the keyword. If num
is 0, it acts like NOCONTENT
.
identifier
is either an attribute name (for hashes and JSON) or a JSON Path expression (for JSON).
property
is an optional name used in the result. If not provided, the identifier
is used in the result.
SUMMARIZE ...
returns only the sections of the attribute that contain the matched text. See Highlighting for more information.
HIGHLIGHT ...
formats occurrences of matched text. See Highlighting for more information.
SLOP {slop}
is the number of intermediate terms allowed to appear between the terms of the query.
Suppose you're searching for a phrase hello world.
If some terms appear in-between hello and world, a SLOP
greater than 0
allows for these text attributes to match.
By default, there is no SLOP
constraint.
INORDER
requires the terms in the document to have the same order as the terms in the query, regardless of the offsets between them. Typically used in conjunction with SLOP
. Default is false
.
LANGUAGE {language}
use a stemmer for the supplied language during search for query expansion. If querying documents in Chinese, set to chinese
to
properly tokenize the query terms. Defaults to English. If an unsupported language is sent, the command returns an error.
See FT.CREATE
for the list of languages.
EXPANDER {expander}
uses a custom query expander instead of the stemmer. See Extensions.
SCORER {scorer}
uses a built-in or a user-provided scoring function.
EXPLAINSCORE
returns a textual description of how the scores were calculated. Using this option requires WITHSCORES
.
PAYLOAD {payload}
adds an arbitrary, binary safe payload that is exposed to custom scoring functions. See Extensions.
SORTBY {attribute} [ASC|DESC] [WITHCOUNT]
orders the results by the value of this attribute. This applies to both text and numeric attributes. Attributes needed for SORTBY
should be declared as SORTABLE
in the index, in order to be available with very low latency. Note that this adds memory overhead.
Sorting Optimizations: performance is optimized for sorting operations on DIALECT 4
in different scenarios:
- Skip Sorter - applied when there is no sort of any kind. The query can return after it reaches the
LIMIT
requested results. - Partial Range - applied when there is a
SORTBY
clause over a numeric field, with no filter or filter by the same numeric field, the query iterate on a range large enough to satisfy theLIMIT
requested results. - Hybrid - applied when there is a
SORTBY
clause over a numeric field and another non-numeric filter. Some results will get filtered, and the initial range may not be large enough. The iterator is then rewinding with the following ranges, and an additional iteration takes place to collect theLIMIT
requested results. - No optimization - If there is a sort by score or by non-numeric field, there is no other option but to retrieve all results and compare their values.
Counts behavior: optionalWITHCOUNT
argument returns accurate counts for the query results with sorting. This operation processes all results in order to get an accurate count, being less performant than the optimized option (default behavior on DIALECT 4
)
LIMIT first num
limits the results to the offset and number of results given. Note that the offset is zero-indexed. The default is 0 10, which returns 10 items starting from the first result. You can use LIMIT 0 0
to count the number of documents in the result set without actually returning them.
TIMEOUT {milliseconds}
overrides the timeout parameter of the module.
PARAMS {nargs} {name} {value}
defines one or more value parameters. Each parameter has a name and a value.
You can reference parameters in the query
by a $
, followed by the parameter name, for example, $user
. Each such reference in the search query to a parameter name is substituted by the corresponding parameter value. For example, with parameter definition PARAMS 4 lon 29.69465 lat 34.95126
, the expression @loc:[$lon $lat 10 km]
is evaluated to @loc:[29.69465 34.95126 10 km]
. You cannot reference parameters in the query string where concrete values are not allowed, such as in field names, for example, @loc
. To use PARAMS
, set DIALECT
to 2
or greater than 2
.
DIALECT {dialect_version}
selects the dialect version under which to execute the query. If not specified, the query will execute under the default dialect version set during module initial loading or via FT.CONFIG SET
command.
Return
FT.SEARCH returns an array reply, where the first element is an integer reply of the total number of results, and then array reply pairs of document ids, and array replies of attribute/value pairs.
- If
NOCONTENT
is given, an array is returned where the first element is the total number of results, and the rest of the members are document ids. - If a hash expires after the query process starts, the hash is counted in the total number of results, but the key name and content return as null.
Return multiple values
When the index is defined ON JSON
, a reply for a single attribute or a single JSONPath may return multiple values when the JSONPath matches multiple values, or when the JSONPath matches an array.
Prior to RediSearch v2.6, only the first of the matched values was returned. Starting with RediSearch v2.6, all values are returned, wrapped with a top-level array.
In order to maintain backward compatibility, the default behavior with RediSearch v2.6 is to return only the first value.
To return all the values, use DIALECT
3 (or greater, when available).
The DIALECT
can be specified as a parameter in the FT.SEARCH command. If it is not specified, the DEFAULT_DIALECT
is used, which can be set using FT.CONFIG SET
or by passing it as an argument to the redisearch
module when it is loaded.
For example, with the following document and index:
127.0.0.1:6379> JSON.SET doc:1 $ '[{"arr": [1, 2, 3]}, {"val": "hello"}, {"val": "world"}]'
OK
127.0.0.1:6379> FT.CREATE idx ON JSON PREFIX 1 doc: SCHEMA $..arr AS arr NUMERIC $..val AS val TEXT
OK
Notice the different replies, with and without DIALECT 3
:
127.0.0.1:6379> FT.SEARCH idx * RETURN 2 arr val
1) (integer) 1
2) "doc:1"
3) 1) "arr"
2) "[1,2,3]"
3) "val"
4) "hello"
127.0.0.1:6379> FT.SEARCH idx * RETURN 2 arr val DIALECT 3
1) (integer) 1
2) "doc:1"
3) 1) "arr"
2) "[[1,2,3]]"
3) "val"
4) "[\"hello\",\"world\"]"
Complexity
FT.SEARCH complexity is O(n) for single word queries. n
is the number of the results in the result set. Finding all the documents that have a specific term is O(1), however, a scan on all those documents is needed to load the documents data from redis hashes and return them.
The time complexity for more complex queries varies, but in general it's proportional to the number of words, the number of intersection points between them and the number of results in the result set.
Examples
Search for a term in every text attribute
Search for the term "wizard" in every TEXT attribute of an index containing book data.
127.0.0.1:6379> FT.SEARCH books-idx "wizard"
Search for a term in title attribute
Search for the term dogs in the title
attribute.
127.0.0.1:6379> FT.SEARCH books-idx "@title:dogs"
Search for books from specific years
Search for books published in 2020 or 2021.
127.0.0.1:6379> FT.SEARCH books-idx "@published_at:[2020 2021]"
Search for a restaurant by distance from longitude/latitude
Search for Chinese restaurants within 5 kilometers of longitude -122.41, latitude 37.77 (San Francisco).
127.0.0.1:6379> FT.SEARCH restaurants-idx "chinese @location:[-122.41 37.77 5 km]"
Search for a book by terms but boost specific term
Search for the term dogs or cats in the title
attribute, but give matches of dogs a higher relevance score (also known as boosting).
127.0.0.1:6379> FT.SEARCH books-idx "(@title:dogs | @title:cats) | (@title:dogs) => { $weight: 5.0; }"
Search for a book by a term and EXPLAINSCORE
Search for books with dogs in any TEXT attribute in the index and request an explanation of scoring for each result.
127.0.0.1:6379> FT.SEARCH books-idx "dogs" WITHSCORES EXPLAINSCORE
Search for a book by a term and TAG
Search for books with space in the title that have science
in the TAG attribute categories
.
127.0.0.1:6379> FT.SEARCH books-idx "@title:space @categories:{science}"
Search for a book by a term but limit the number
Search for books with Python in any TEXT
attribute, returning 10 results starting with the 11th result in the entire result set (the offset parameter is zero-based), and return only the title
attribute for each result.
127.0.0.1:6379> FT.SEARCH books-idx "python" LIMIT 10 10 RETURN 1 title
Search for a book by a term and price
Search for books with Python in any TEXT
attribute, returning the price stored in the original JSON document.
127.0.0.1:6379> FT.SEARCH books-idx "python" RETURN 3 $.book.price AS price
Search for a book by title and distance
Search for books with semantically similar title to Planet Earth. Return top 10 results sorted by distance.
127.0.0.1:6379> FT.SEARCH books-idx "*=>[KNN 10 @title_embedding $query_vec AS title_score]" PARAMS 2 query_vec <"Planet Earth" embedding BLOB> SORTBY title_score DIALECT 2
Search for a phrase using SLOP
Search for a phrase hello world. First, create an index.
127.0.0.1:6379> FT.CREATE memes SCHEMA phrase TEXT
OK
Add variations of the phrase hello world.
127.0.0.1:6379> HSET s1 phrase "hello world"
(integer) 1
127.0.0.1:6379> HSET s2 phrase "hello simple world"
(integer) 1
127.0.0.1:6379> HSET s3 phrase "hello somewhat less simple world"
(integer) 1
127.0.0.1:6379> HSET s4 phrase "hello complicated yet encouraging problem solving world"
(integer) 1
127.0.0.1:6379> HSET s5 phrase "hello complicated yet amazingly encouraging problem solving world"
(integer) 1
Then, search for the phrase hello world. The result returns all documents that contain the phrase.
127.0.0.1:6379> FT.SEARCH memes '@phrase:(hello world)' NOCONTENT
1) (integer) 5
2) "s1"
3) "s2"
4) "s3"
5) "s4"
6) "s5"
Now, return all documents that have one of fewer words between hello and world.
127.0.0.1:6379> FT.SEARCH memes '@phrase:(hello world)' NOCONTENT SLOP 1
1) (integer) 2
2) "s1"
3) "s2"
Now, return all documents with three or fewer words between hello and world.
127.0.0.1:6379> FT.SEARCH memes '@phrase:(hello world)' NOCONTENT SLOP 3
1) (integer) 3
2) "s1"
3) "s2"
4) "s3"
s5
needs a higher SLOP
to match, SLOP 6
or higher, to be exact. See what happens when you set SLOP
to 5
.
127.0.0.1:6379> FT.SEARCH memes '@phrase:(hello world)' NOCONTENT SLOP 5
1) (integer) 4
2) "s1"
3) "s2"
4) "s3"
5) "s4"
If you add additional terms (and stemming), you get these results.
127.0.0.1:6379> FT.SEARCH memes '@phrase:(hello amazing world)' NOCONTENT
1) (integer) 1
2) "s5"
127.0.0.1:6379> FT.SEARCH memes '@phrase:(hello encouraged world)' NOCONTENT SLOP 5
1) (integer) 2
2) "s4"
3) "s5"
127.0.0.1:6379> FT.SEARCH memes '@phrase:(hello encouraged world)' NOCONTENT SLOP 4
1) (integer) 1
2) "s4"
If you swap the terms, you can still retrieve the correct phrase.
127.0.0.1:6379> FT.SEARCH memes '@phrase:(amazing hello world)' NOCONTENT
1) (integer) 1
2) "s5"
But, if you use INORDER
, you get zero results.
127.0.0.1:6379> FT.SEARCH memes '@phrase:(amazing hello world)' NOCONTENT INORDER
1) (integer) 0
Likewise, if you use a query attribute $inorder
set to true
, s5
is not retrieved.
127.0.0.1:6379> FT.SEARCH memes '@phrase:(amazing hello world)=>{$inorder: true;}' NOCONTENT
1) (integer) 0
To sum up, the INORDER
argument or $inorder
query attribute require the query terms to match terms with similar ordering.
NEW!!! Polygon Search with WITHIN and CONTAINS operators
Query for polygons which contain a given geoshape or are within a given geoshape
First, create an index using GEOSHAPE
type with a FLAT
coordinate system:
127.0.0.1:6379> FT.CREATE idx SCHEMA geom GEOSHAPE FLAT
OK
Adding a couple of geometries using HSET
:
127.0.0.1:6379> HSET small geom 'POLYGON((1 1, 1 100, 100 100, 100 1, 1 1))'
(integer) 1
127.0.0.1:6379> HSET large geom 'POLYGON((1 1, 1 200, 200 200, 200 1, 1 1))'
(integer) 1
Query with WITHIN
operator:
127.0.0.1:6379> FT.SEARCH idx '@geom:[WITHIN $poly]' PARAMS 2 poly 'POLYGON((0 0, 0 150, 150 150, 150 0, 0 0))' DIALECT 3
1) (integer) 1
2) "small"
3) 1) "geom"
2) "POLYGON((1 1, 1 100, 100 100, 100 1, 1 1))"
Query with CONTAINS
operator:
127.0.0.1:6379> FT.SEARCH idx '@geom:[CONTAINS $poly]' PARAMS 2 poly 'POLYGON((2 2, 2 50, 50 50, 50 2, 2 2))' DIALECT 3
1) (integer) 2
2) "small"
3) 1) "geom"
2) "POLYGON((1 1, 1 100, 100 100, 100 1, 1 1))"
4) "large"
5) 1) "geom"
2) "POLYGON((1 1, 1 200, 200 200, 200 1, 1 1))"
See also
Related topics
History
- Starting with Redis version 2.0.0: Deprecated
WITHPAYLOADS
andPAYLOAD
arguments