Questo contenuto non è disponibile nella lingua selezionata.
Chapter 2. Creating Ickle queries
Data Grid provides an Ickle query language that lets you create relational and full-text queries.
2.1. Ickle queries
To use the API, call the cache .query()
method and provide the query string.
For instance:
// Remote Query using protobuf Query<Transaction> q = remoteCache.query("from sample_bank_account.Transaction where amount > 20"); // Embedded Query using Java Objects Query<Transaction> q = cache.query("from org.infinispan.sample.Book where price > 20"); // Execute the query QueryResult<Book> queryResult = q.execute();
A query will always target a single entity type and is evaluated over the contents of a single cache. Running a query over multiple caches or creating queries that target several entity types (joins) is not supported.
Executing the query and fetching the results is as simple as invoking the execute()
method of the Query
object. Once executed, calling execute()
on the same instance will re-execute the query.
2.1.1. Pagination
You can limit the number of returned results by using the Query.maxResults(int maxResults)
. This can be used in conjunction with Query.startOffset(long startOffset)
to achieve pagination of the result set.
// sorted by year and match all books that have "clustering" in their title // and return the third page of 10 results Query<Book> query = cache.query("FROM org.infinispan.sample.Book WHERE title like '%clustering%' ORDER BY year").startOffset(20).maxResults(10)
If you don’t explicitly set the maxResults
for a query instance, Data Grid limits the number of results returned by the query to 100
. You can change the default limit by setting the query.default-max-results
cache property.
2.1.2. Number of hits
The QueryResult
object includes the .hitCount()
method, which returns a hit count value that represents the total number of results from a query, regardless of any pagination parameter.
Additionally, QueryResult
object contains a boolean value returned by the .isExact()
method which indicates whether the hit count number is exact or a lower bound. The hit count is only available for indexed queries for performance reasons.
2.1.2.1. Hit count accuracy
You can limit the required accuracy of hit counts by setting hit-count-accuracy
attribute. When dealing with large data sets, precise hit counts can impact performance. Setting a limit to the hit count accuracy, lets you achieve faster query responses while ensuring that the provided hit counts remain sufficiently accurate for your application’s needs.
The default accuracy of the hit-count-accuracy
attribute is limited to 10000
. This means that for any query, Data Grid provides exact hit count up to maximum of 10,000. If the effective hit count is higher than 10,000, Data Grid returns a lower bound estimate of the count. You can change the default limit by setting the query.hit-count-accuracy
cache property. Alternatively, it can be set on each query instance.
When the actual hit count exceeds the limit set by the hit-count-accuracy
, the .isExact()
method or the hit_count_exact
JSON field will be false
, indicating that the returned hit count is an estimation. Setting this value to Integer.MAX
would return accurate results for any query, but this can severely impact query performance.
For optimal performance set the property value slightly above the expected hit count. If you do not require accurate hit counts, set it to a low value.
2.1.3. Iteration
The Query
object has the .iterator()
method to obtain the results lazily. It returns an instance of CloseableIterator
that must be closed after usage.
The iteration support for Remote Queries is currently limited, as it will first fetch all entries to the client before iterating.
2.1.4. Named query parameters
Instead of building a new Query object for every execution it is possible to include named parameters in the query which can be substituted with actual values before execution. This allows a query to be defined once and be efficiently executed many times. Parameters can only be used on the right-hand side of an operator and are defined when the query is created by supplying an object produced by the org.infinispan.query.dsl.Expression.param(String paramName)
method to the operator instead of the usual constant value. Once the parameters have been defined they can be set by invoking either Query.setParameter(parameterName, value)
or Query.setParameters(parameterMap)
as shown in the examples below.
// Defining a query to search for various authors and publication years Query<Book> query = cache.query("SELECT title FROM org.infinispan.sample.Book WHERE author = :authorName AND publicationYear = :publicationYear").build(); // Set actual parameter values query.setParameter("authorName", "Doe"); query.setParameter("publicationYear", 2010); // Execute the query List<Book> found = query.execute().list();
Alternatively, you can supply a map of actual parameter values to set multiple parameters at once:
Setting multiple named parameters at once
Map<String, Object> parameterMap = new HashMap<>(); parameterMap.put("authorName", "Doe"); parameterMap.put("publicationYear", 2010); query.setParameters(parameterMap);
A significant portion of the query parsing, validation and execution planning effort is performed during the first execution of a query with parameters. This effort is not repeated during subsequent executions leading to better performance compared to a similar query using constant values instead of query parameters.
2.1.5. Query execution
The Query
API provides two methods for executing Ickle queries on a cache:
-
Query.execute()
runs a SELECT statement and returns a result. -
Query.executeStatement()
runs a DELETE statement and modifies data.
You should always invoke executeStatement()
to modify data and invoke execute()
to get the result of a query.
2.2. Ickle query language syntax
The Ickle query language is subset of the JPQL query language, with some extensions for full-text.
The parser syntax has some notable rules:
- Whitespace is not significant.
- Wildcards are not supported in field names.
- A field name or path must always be specified, as there is no default field.
-
&&
and||
are accepted instead ofAND
orOR
in both full-text and JPA predicates. -
!
may be used instead ofNOT
. -
A missing boolean operator is interpreted as
OR
. - String terms must be enclosed with either single or double quotes.
- Fuzziness and boosting are not accepted in arbitrary order; fuzziness always comes first.
-
!=
is accepted instead of<>
. -
Boosting cannot be applied to
>
,>=
,<
,<=
operators. Ranges may be used to achieve the same result.
2.2.1. Filtering operators
Ickle support many filtering operators that can be used for both indexed and non-indexed fields.
Operator | Description | Example |
---|---|---|
| Checks that the left operand is equal to one of the elements from the Collection of values given as argument. |
|
| Checks that the left argument (which is expected to be a String) matches a wildcard pattern that follows the JPA rules. |
|
| Checks that the left argument is an exact match of the given value. |
|
| Checks that the left argument is different from the given value. |
|
| Checks that the left argument is greater than the given value. |
|
| Checks that the left argument is greater than or equal to the given value. |
|
| Checks that the left argument is less than the given value. |
|
| Checks that the left argument is less than or equal to the given value. |
|
| Checks that the left argument is between the given range limits. |
|
2.2.2. Boolean conditions
Combining multiple attribute conditions with logical conjunction (and
) and disjunction (or
) operators in order to create more complex conditions is demonstrated in the following example. The well known operator precedence rule for boolean operators applies here, so the order of the operators is irrelevant. Here and
operator still has higher priority than or
even though or
was invoked first.
# match all books that have "Data Grid" in their title # or have an author named "Manik" and their description contains "clustering" FROM org.infinispan.sample.Book WHERE title LIKE '%Data Grid%' OR author.name = 'Manik' AND description like '%clustering%'
Boolean negation has highest precedence among logical operators and applies only to the next simple attribute condition.
# match all books that do not have "Data Grid" in their title and are authored by "Manik" FROM org.infinispan.sample.Book WHERE title != 'Data Grid' AND author.name = 'Manik'
2.2.3. Nested conditions
Changing the precedence of logical operators is achieved with parenthesis:
# match all books that have an author named "Manik" and their title contains # "Data Grid" or their description contains "clustering" FROM org.infinispan.sample.Book WHERE author.name = 'Manik' AND ( title like '%Data Grid%' OR description like '% clustering%')
2.2.4. Projections with SELECT statements
In some use cases returning the whole domain object is overkill if only a small subset of the attributes are actually used by the application, especially if the domain entity has embedded entities. The query language allows you to specify a subset of attributes (or attribute paths) to return - the projection. If projections are used then the QueryResult.list()
will not return the whole domain entity but will return a List
of Object[]
, each slot in the array corresponding to a projected attribute.
# match all books that have "Data Grid" in their title or description # and return only their title and publication year SELECT title, publicationYear FROM org.infinispan.sample.Book WHERE title like '%Data Grid%' OR description like '%Data Grid%'
2.2.4.1. Project cache entry version
It is possible to project the cache entry version, using the version
projection function.
# return the title, publication year and the cache entry version SELECT b.title, b.publicationYear, version(b) FROM org.infinispan.sample.Book b WHERE b.title like '%Data Grid%'
2.2.4.2. Project cache entry value
It is possible to project the cache entry value together with other projections. It can be used for instance to project the cache entry value together with the cache entry version in the same Object[]
returned hit.
# return the cache entry value and the cache entry version SELECT b, version(b) FROM org.infinispan.sample.Book b WHERE b.title like '%Data Grid%'
2.2.4.3. Project the score
If the query is indexed, it is possible to project the score obtained by each matching together with other projections. It can be used for instance to project the cache entry value together with the score in the same Object[]
returned hit.
# return the cache entry value and the the score of the matching SELECT b, score(b) FROM org.infinispan.sample.Book b WHERE b.title like '%Data Grid%'
Sorting
Ordering the results based on one or more attributes or attribute paths is done with the ORDER BY
clause. If multiple sorting criteria are specified, then the order will dictate their precedence.
# match all books that have "Data Grid" in their title or description # and return them sorted by the publication year and title FROM org.infinispan.sample.Book WHERE title like '%Data Grid%' ORDER BY publicationYear DESC, title ASC
2.2.5. Grouping and aggregation
Data Grid has the ability to group query results according to a set of grouping fields and construct aggregations of the results from each group by applying an aggregation function to the set of values that fall into each group. Grouping and aggregation can only be applied to projection queries (queries with one or more field in the SELECT clause).
The supported aggregations are: avg
, sum
, count
, max
, and min
.
The set of grouping fields is specified with the GROUP BY
clause and the order used for defining grouping fields is not relevant. All fields selected in the projection must either be grouping fields or else they must be aggregated using one of the grouping functions described below. A projection field can be aggregated and used for grouping at the same time. A query that selects only grouping fields but no aggregation fields is legal. Example: Grouping Books by author and counting them.
SELECT author, COUNT(title) FROM org.infinispan.sample.Book WHERE title LIKE '%engine%' GROUP BY author
A projection query in which all selected fields have an aggregation function applied and no fields are used for grouping is allowed. In this case the aggregations will be computed globally as if there was a single global group.
Aggregations
You can apply the following aggregation functions to a field:
Aggregation function | Description |
---|---|
|
Computes the average of a set of numbers. Accepted values are primitive numbers and instances of |
|
Counts the number of non-null rows and returns a |
|
Returns the greatest value found. Accepted values must be instances of |
|
Returns the smallest value found. Accepted values must be instances of |
|
Computes the sum of a set of Numbers. If there are no non-null values the result is |
Field Type | Return Type |
---|---|
Integral (other than BigInteger) | Long |
Float or Double | Double |
BigInteger | BigInteger |
BigDecimal | BigDecimal |
Evaluation of queries with grouping and aggregation
Aggregation queries can include filtering conditions, like usual queries. Filtering can be performed in two stages: before and after the grouping operation. All filter conditions defined before invoking the groupBy()
method will be applied before the grouping operation is performed, directly to the cache entries (not to the final projection). These filter conditions can reference any fields of the queried entity type, and are meant to restrict the data set that is going to be the input for the grouping stage. All filter conditions defined after invoking the groupBy()
method will be applied to the projection that results from the projection and grouping operation. These filter conditions can either reference any of the groupBy()
fields or aggregated fields. Referencing aggregated fields that are not specified in the select clause is allowed; however, referencing non-aggregated and non-grouping fields is forbidden. Filtering in this phase will reduce the amount of groups based on their properties. Sorting can also be specified similar to usual queries. The ordering operation is performed after the grouping operation and can reference any of the groupBy()
fields or aggregated fields.
2.2.6. DELETE statements
You can delete entities from Data Grid caches with the following syntax:
DELETE FROM <entityName> [WHERE condition]
-
Reference only single entities with
<entityName>
. DELETE queries cannot use joins. - WHERE conditions are optional.
DELETE queries cannot use any of the following:
- Projections with SELECT statements
- Grouping and aggregation
- ORDER BY clauses
Invoke the Query.executeStatement()
method to execute DELETE statements.
Additional resources
2.3. Full-text queries
You can perform full-text searches with the Ickle query language.
2.3.1. Fuzzy queries
To execute a fuzzy query add ~
along with an integer, representing the distance from the term used, after the term. For instance
FROM sample_bank_account.Transaction WHERE description : 'cofee'~2
2.3.2. Range queries
To execute a range query define the given boundaries within a pair of braces, as seen in the following example:
FROM sample_bank_account.Transaction WHERE amount : [20 to 50]
2.3.3. Phrase queries
A group of words can be searched by surrounding them in quotation marks, as seen in the following example:
FROM sample_bank_account.Transaction WHERE description : 'bus fare'
2.3.4. Proximity queries
To execute a proximity query, finding two terms within a specific distance, add a ~
along with the distance after the phrase. For instance, the following example will find the words canceling and fee provided they are not more than 3 words apart:
FROM sample_bank_account.Transaction WHERE description : 'canceling fee'~3
2.3.5. Wildcard queries
To search for "text" or "test", use the ?
single-character wildcard search:
FROM sample_bank_account.Transaction where description : 'te?t'
To search for "test", "tests", or "tester", use the *
multi-character wildcard search:
FROM sample_bank_account.Transaction where description : 'test*'
2.3.6. Regular expression queries
Regular expression queries can be performed by specifying a pattern between /
. Ickle uses Lucene’s regular expression syntax, so to search for the words moat
or boat
the following could be used:
FROM sample_library.Book where title : /[mb]oat/
2.3.7. Boosting queries
Terms can be boosted by adding a ^
after the term to increase their relevance in a given query, the higher the boost factor the more relevant the term will be. For instance to search for titles containing beer and wine with a higher relevance on beer, by a factor of 3, the following could be used:
FROM sample_library.Book WHERE title : beer^3 OR wine
2.4. Vector search queries
You can perform vector kNN searches with the Ickle query language using the special operator <->
to define predicates.
This is an example of kNN query:
from play.Item i where i.myVector <-> [7,7,7]~3
This query will find the items that have the myVector
fields that are within 3
nearest neighbourhood from the vector [7,7,7]
.
Notice that in order to use this kind of search the entity, in our example play.Item
, has to be @Indexed
and the field, in our example myVector
, should be annotated with @Vector
.
We support two kinds of vector field types:
-
byte
/Byte
(to work with byte vectors) -
float
/Float
(to work with float vectors)
You can have different vector fields on the same entity, but in any case you can have only one vector predicate on your queries.
2.4.1. Vector search parameters
Both the k-value and the vector can be passed as query parameter. The k-value scalar can be expressed with the usual placeholder :k
in the Ickle text. For the vector we can use either a placeholder for each term of the vector:
Query<Item> query = cache.query("from play.Item i where i.floatVector <-> [:a,:b,:c]~:k"); query.setParameter("a", 1); query.setParameter("b", 4.3); query.setParameter("c", 3.3); query.setParameter("k", 4);
Or a placeholder can be used for the entire vector:
Query<Item> query = cache.query("from play.Item i where i.floatVector <-> [:a]~:k"); query.setParameter("a", new float[]{7.1f, 7.0f, 3.1f}); query.setParameter("k", 3);
2.4.2. Score projection with vector search
Is very common also to return the score of the computation, using the score projection. In the case of vector search the query will be like the following:
Query<Object[]> query = cache.query("select i, score(i) from play.Item i where i.floatVector <-> [:a]~:k"); query.setParameter("a", new float[]{7.1f, 7.0f, 3.1f}); query.setParameter("k", 3); List<Object[]> resultList = query.list();
In this case the first element of each array will contain the entity, and the second element will contain the score of the matching.
2.4.3. Filtering entities
Instead of applying the kNN search to the entire population of entities of a given type, you can limiting the searching set applying classic predicates (match, full-text-search, range, …) to the kNN queries, defining what we call a filtering clause.
A filtering clause can contain any kind of predicates with the only exception of kNN predicates that cannot be included.
For instance the following query:
Query<Object[]> query = remoteCache.query( "select score(i), i from Item i where i.floatVector <-> [:a]~:k filtering (i.buggy : 'cat' or i.text : 'code')"); query.setParameter("a", new float[]{7, 7, 7}); query.setParameter("k", 3);
Will return the nearest 3 items from the point [7,7,7]
selecting only the items that have a text containing the term cat
or code
.
The filtering queries is a way to apply the classic indexed search to the new vector search.
2.4.4. Vector field attributes
It is always required to specify the dimension of the vector field.
The other mapping attributes are optional, since Infinispan will have a default for each of them. You can configure them, for instance, to tune the desired accuracy / performance.
2.4.4.1. Similarity
Different VectorSimilarity
algorithms are supported
Value | Distance | Score | Note |
---|---|---|---|
| This is the Data Grid default | ||
| To use this similarity efficiently, both index and search vectors must be normalized | ||
| This similarity does not require vector normalization | ||
|
This similarity cannot be of |
2.4.4.2. Beanwidth
Changing the beamWidth
you can modify the the size of the dynamic list used during k-NN graph creation. It affects how vectors are stored. Higher values lead to a more accurate graph but slower indexing speed. The Data Grid default is 512.
2.4.4.3. Max Connections
The number of neighbors each node will be connected to in the HNSW graph. Modifying this value will have an impact on memory consumption. It is recommended to keep this value between 2 and 100. The Data Grid default is 16.