You are viewing the documentation for Blueriq 17. Documentation for other versions is available in our documentation directory.
EXISTS
This function determines whether an instance of a specified entity exists, optionally meeting certain criteria.
Syntax
EXISTS instances [WHERE condition]
- instances - Instances to search.
- condition - boolean expression that represents the criterion the instance has to meet.
Return type
- boolean
Examples
Suppose the following data model.
Person.name | Person.gender | Person.age |
---|---|---|
“Kim” | “f” | 23 |
“Rick” | “m” | 35 |
“Bob” | “m” | 42 |
“John” | “m” | 19 |
“Mary” | “f” | 33 |
EXISTS Person
= TRUEEXISTS Person WHERE ( Person.age < 18 )
= FALSEEXISTS Person WHERE ( Person.gender = “m” AND Person.age > 35 )
= TRUE
The return value of a boolean can be TRUE
, FALSE
or UNKNOWN
.
EACH
This function determines whether all instances of a specified entity meet a certain criteria
Syntax
EACH instances WHERE condition
- instances - A collection of instances to search.
- condition - Boolean expression that represents the criterion the instance has to meet.
Return type
- boolean
Examples
Suppose the following data model.
Person.name | Person.gender | Person.age |
---|---|---|
“Kim” | “f” | 23 |
“Rick” | “m” | 35 |
“Bob” | “m” | 42 |
“John” | “m” | 19 |
“Mary” | “f” | 33 |
EACH Person WHERE ( Person.age < 18 )
= FALSEEACH Person WHERE ( Person.age > 18 )
= TRUEEACH Person WHERE ( Person.age < 20 )
= FALSEEACH Person WHERE ( Person.gender = “m” OR Person.age > 35 )
= FALSEEACH Person WHERE ( Person.gender = “m” OR Person.age > 20 )
= TRUE
ALL
Use this function to create a collection of all instances of a specified entity.
Syntax
ALL entity
- entity - Entity of which to collect all its instances.
Return type
- collection of entity instances
Examples
Suppose the following data model. The Parent
and Child
entities both have Person
as base entity.
Parent instance | Child instance |
---|---|
Parent_1 | Child_1 |
Parent_1 | Child_2 |
Parent_1 | Child_3 |
Parent_2 | Child_4 |
ALL Parent
= [ Parent_1 , Parent_2 ]ALL Child
= [ Child_1 , Child_2 , Child_3 , Child_4 ]ALL Person
= [ Parent_1 , Parent_2 , Child_1 , Child_2 , Child_3 , Child_4 ]
COLLECT FROM [WHERE]
Use this function to create a collection of entity or attribute instances (meeting certain criteria).
Syntax
COLLECT entity | attribute FROM collection [WHERE expression]
- entity or attribute - Entity or attribute to collect.
- collection - A collection of entity instances.
- expression - Boolean expression that represents the criterion the instance has to meet.
Return type
- collection of entity instances
- collection of attribute values
Examples
Suppose the following model. Entity Teacher
has a multivalued relation with entity Child
via the relation Teacher.teaches_Children
.
Teacher instance | Child instance | Child.name | Child.hobbies |
---|---|---|---|
Teacher_1 | Child_1 | “Kim” | “Reading”, “Dancing” |
Teacher_1 | Child_2 | “Rick” | “Tennis”, “Dancing” |
Teacher_1 | Child_3 | “Bob” | “Painting”, “Basketball”, “Reading” |
Teacher_2 | Child_1 | “Kim” | “Reading”, “Dancing” |
Teacher_2 | Child_3 | “Bob” | “Painting”, “Basketball”, “Reading” |
Teacher_2 | Child_4 | “Mary” | “Football” |
COLLECT Child.name FROM ALL Child
= [ “Kim”, “Rick”, “Bob” and “Mary” ]COLLECT Child FROM Teacher[Teacher_2].teaches_Children
= [Child_1
,Child_3
andChild_4 ]
COLLECT Child.hobbies FROM Teacher[Teacher_1].teaches_Children
= [ “Reading”, “Dancing”, “Tennis”, “Painting” and “Basketball” ]COLLECT Child.name FROM ALL Child WHERE ( Child.hobbies = “Reading” )
= [ “Kim” and “Bob” ]COLLECT Child.hobbies FROM ALL Child WHERE ( Child.name = “Mary” )
= [ “Football” ]
COLLECT FROM NAMED [WHERE]
Use this version of the COLLECT FROM [WHERE] function for complex nested selections in which it is necessary to use an alias.
Syntax
COLLECT entity | attribute FROM collection NAMED alias [WHERE expression]
- entity or attribute - Entity or attribute to collect. This should contain the alias, e.g. alias.Name.
- collection - A collection of entity instances.
- alias - A name for the collection.
- expression - Boolean expression that represents the criterion the instance has to meet.
Return type
- collection of entity instances
- collection of attribute values
Examples
Suppose the following model with only instances of entity Person
. Person.hasChildren
is a relation from Person to Person.
Person instance | Person.Name | Person.Age | Person.hasChildren |
---|---|---|---|
Person_1 | “Kim” | 16 | |
Person_2 | “Rick” | 38 | Person_3, Person_5 |
Person_3 | “Bob” | 8 | |
Person_4 | “Julia” | 42 | Person_1 |
Person_5 | “Sam” | 3 | |
Person_6 | “Joan” | 34 | Person_3, Person_5 |
To retrieve all the parent names, we use this expression:
COLLECT Parent.name FROM COLLECT Person FROM ALL Person WHERE ( Person.hasChildren != ? ) NAMED Parent
This expression results in “Rick”, “Julia” and “Joan”.
To retrieve all children names, we use this expression:
COLLECT Child FROM ( COLLECT Person.hasChildren FROM ALL Person WHERE ( Person.hasChildren != ? ) ) NAMED Child
This expression results in a collection of three instances of Person
(i.e Person_1
, Person_3
and Person_5
, with names Kim, Bob and Sam).
To retrieve all children younger than 15, we use this expression:
COLLECT Child FROM ( COLLECT Person.hasChildren FROM ALL Person WHERE ( Person.hasChildren != ? ) ) NAMED Child WHERE ( Child.Age < 15 )
This expression results in a collection of two instances of Person
(i.e Person_3
and Person_5
, with names Bob and Sam).
To retrieve the children’s names with a parent older than 40 years:
COLLECT Child.Name FROM ( COLLECT Person.hasChildren FROM ALL Person WHERE ( Person.hasChildren != ? AND Person.Age > 40 ) ) NAMED Child
This expression results in a collection with only the name “Kim”.
You can make it as complex as you like, for example:
COLLECT Child.Name FROM ( COLLECT Parent.hasChildren FROM ( COLLECT Person FROM ALL Person WHERE ( Person.hasChildren != ? ) ) NAMED Parent WHERE ( Parent.Age > 40 ) ) NAMED Child WHERE ( Child.Age < 18 )
This expression results also in a collection with only the name “Kim”.
In some occasions, a local variable could be a good alternative for using an alias. E.g. if you want a collection of all persons except the person that is currently active, this would be simple and transparent:
thisPerson := Person ; COLLECT Person FROM ALL Person WHERE ( Person != thisPerson )
UNPACK
Use this function to extract the value from a single valued collection or list.
Syntax
UNPACK collection/list
- collection/list - A collection or list of one entity or attribute instance.
Return type
- entity instance
- attribute value of any type
Examples
Suppose the following data model.
Person.name | Person.SequenceNumber |
---|---|
“Bob” | 654 |
“Jane” | 523 |
“Mary” | 667 |
“Rick” | 500 |
“Ron” | 490 |
“Jenny” | 765 |
UNPACK ( COLLECT Person.name FROM ALL Person WHERE ( Person.SequenceNumber = MIN ( COLLECT Person.SequenceNumber FROM ALL Person ) ) )
= “Ron”- In case a second entry "Ron","490" exists, the expression
UNPACK ( COLLECT Person.name FROM ALL Person WHERE ( Person.SequenceNumber = MIN ( COLLECT Person.SequenceNumber FROM ALL Person ) ) )
will fail, because theUNPACK
cannot resolve a list with two elements. To solve this theUNIQUE
function has to be used:UNPACK ( UNIQUE ( COLLECT Person.name FROM ALL Person WHERE ( Person.SequenceNumber = MIN ( COLLECT Person.SequenceNumber FROM ALL Person ) ) ) )
= "Ron". UNPACK ( COLLECT Person.name FROM ALL Person WHERE ( Person.SequenceNumber = MAX ( COLLECT Person.SequenceNumber FROM ALL Person ) ) )
= “Jenny”
SIZE
This function determines the size of a collection.
Syntax
SIZE ( collection )
- collection - A collection of attribute or entity instances. This can be an expression or a relation attribute for instance.
Return type
- integer
Example
Suppose you have a Parent
and a Child
entity, where Parent
has a multivalued relation with Child
via the relation Parent.has_Children
. With this model the following instances are created:
Parent instance | Child instance | Child.name | Child.hobbies |
---|---|---|---|
Parent_1 | Child_1 | “Kim” | “Reading”, “Dancing” |
Parent_1 | Child_2 | “Rick” | “Tennis”, “Dancing” |
Parent_1 | Child_3 | “Bob” | “Painting”, “Basketball”, “Reading” |
Parent_2 | Child_4 | “Mary” | “Football” |
Then:
- if
Parent_1
is active,SIZE ( Parent.has_Children )
= 3 - if
Parent_2
is active,SIZE ( Parent.has_Children )
= 1 - without an active
Parent
instance,SIZE ( Parent.has_Children )
results in an error - if
Child_1
is active,SIZE ( Child.hobbies )
= 2 - if
Child_3
is active,SIZE ( Child.hobbies )
= 3 - if
Child_4
is active,SIZE ( Child.hobbies )
= 1 SIZE ( ? )
= 0
UNIQUE
The UNIQUE
function filters duplicate items from a collection. An expression resulting in a collection, never contains duplicate values. A subexpression however, can contain duplicates.
Syntax
UNIQUE ( collection )
- collection - A collection of attribute or entity instances.
Examples
Suppose the following model.
Person instance | Person.name |
---|---|
Person_1 | “Kim” |
Person_2 | “Rick” |
Person_3 | “Bob” |
Person_4 | “Rick” |
COLLECT Person.name FROM ALL Person
= "Kim”, “Rick”, “Bob” (a result never contains duplicate values)SIZE ( COLLECT Person.name FROM ALL Person )
= 4 (a subexpression can contain duplicate values)SIZE ( UNIQUE ( COLLECT Person.name FROM ALL Person ) )
= 3 (the collection holds three unique values)
SUBSET OF
This function returns TRUE
if the items in a collection are all present in another collection.
Syntax
collection1 SUBSET OF collection2
- collection1 - The collection that is tested to be a subset of the second collection.
- collection2 - The collection that is tested to hold all the items in the first collection.
Return type
- boolean
Examples
('a', 'b', 'c') SUBSET OF ('a', 'b', 'c', 'd')
= TRUE('a', 'b', 'c', 'd') SUBSET OF ('a', 'b', 'c')
= FALSE(Person.hobbies) SUBSET OF ([“Tennis”, “Soccer”, “Music”])
= TRUE ifPerson.hobbies = “Tennis”, “Soccer”
(Person.hobbies) SUBSET OF ([“Tennis”, “Soccer”, “Music”])
= UNKNOWNif Person.hobbies = ?
(Person.hobbies) SUBSET OF ([“Tennis”, “Soccer”, ?])
= FALSE ifPerson.hobbies = “Tennis”, “Soccer”
('a', 'b', 'c') SUBSET OF (['a', 'b', 'c'])
will result in an error.
UNION
Adds two collections of the same base type to a new collection.
Syntax
UNION ( collection1 , collection2 )
- collection1 - First collection to be added to the new collection.
- collection2 - Second collection to be added to the new collection.
Return type
- collection
Examples
Suppose you have a Parent
and a Child
entity, where Parent
has a multivalued relation with Child
via the relation Parent.has_Children
. With this model the following instances are created:
Parent instance | Child instance | Child.name | Child.hobbies |
---|---|---|---|
Parent_1 | Child_1 | “Kim” | “Reading”, “Dancing” |
Parent_1 | Child_2 | “Rick” | “Tennis”, “Dancing” |
Parent_1 | Child_3 | “Bob” | “Painting”, “Basketball”, “Reading” |
Parent_2 | Child_4 | “Mary” | “Football” |
UNION ( Parent[Parent_1].has_Children , Parent[Parent_2].Has_Children )
results in a collection of allChild
instancesUNION ( Parent[Parent_1].has_Children.name , Parent[Parent_2].Has_Children.name )
= “Kim”, “Rick”, “Bob”, “Mary”- if
Child_2
is active,UNION ( Child.hobbies , “Reading” )
= “Tennis”, “Dancing”, “Reading” UNION ( Child[Child_1].hobbies , Child[Child_2].hobbies )
= “Reading”, “Dancing”, “Tennis”
INTERSECTION
This function determines the intersection of two collections. It returns a collection containing the items that are present in both specified collections.
Syntax
INTERSECTION ( collection1 , collection2 )
- collection1, collection2 - Collections to be intersected. These collections must be of the same base type.
Return type
- collection
Examples
Suppose the following model. Entity Teacher
has a multivalued relation with entity Child
via the relation Teacher.teaches_Children
.
Teacher instance | Child instance | Child.name | Child.hobbies |
---|---|---|---|
Teacher_1 | Child_1 | “Kim” | “Reading”, “Dancing” |
Teacher_1 | Child_2 | “Rick” | “Tennis”, “Dancing” |
Teacher_1 | Child_3 | “Bob” | “Painting”, “Basketball”, “Reading” |
Teacher_2 | Child_1 | “Kim” | “Reading”, “Dancing” |
Teacher_2 | Child_3 | “Bob” | “Painting”, “Basketball”, “Reading” |
Teacher_2 | Child_4 | “Mary” | “Football” |
INTERSECTION ( Teacher[Teacher_1].teaches_Children , Teacher[Teacher_2].teaches_Children )
results in a collection ofChild
instancesChild_1
andChild_3
INTERSECTION ( Teacher[Teacher_1].teaches_Children.name , Teacher[Teacher_2].teaches_Children.name )
= “Kim”, “Bob”INTERSECTION ( Child[Child_1].hobbies , Child[Child_3].hobbies )
= “Reading”INTERSECTION ( Child[Child_2].hobbies , Child[Child_3].hobbies )
results in an empty list
DIFFERENCE
This function returns a collection containing all the items from collection1 that are not present in collection 2.
Syntax
DIFFERENCE ( collection1 , collection2 )
- collection1, collection2 - Collections to be compared. These collections must be of the same base type.
Return type
- collection
Examples
DIFFERENCE ( [ "a", "b", "c"] , [ "c", "d", "e" ] )
= [ "a" , "b" ]DIFFERENCE ( [ "nv" , "bv" ] , [ "NV" ] )
= [ "bv" ]DIFFERENCE ( 1 , 1 )
results in an empty list
SYMMETRIC_DIFFERENCE
This function determines the symmetric difference between two collections. It returns a collection with the elements of the provided collections which are in either one of the collections, but not in both.
Syntax
SYMMETRIC_DIFFERENCE ( collection1 , collection2 )
- collection1, collection2 - Collections to be compared. These collections must be of the same base type.
Return type
- collection
Examples
SYMMETRIC_DIFFERENCE ( [ "a", "b", "c"] , [ "c", "d", "e" ] )
= [ "a", "b", "d", "e" ]SYMMETRIC_DIFFERENCE ( [ "nv" , "bv" ] , [ "NV" ] )
= [ "bv" ]SYMMETRIC_DIFFERENCE ( 1 , 1 )
results in an empty list
A note on collections and duplicates
An expression resulting in a collection does not contain duplicates. Please be aware however, that intermediary results of a COLLECT
statement can contain duplicates. You have to be aware of this when using the SIZE
or UNPACK
function.
This is best illustrated with the following examples.
Person instance | Person.Name | Person.Age |
---|---|---|
Person_1 | Kim | 24 |
Person_2 | Rick | 25 |
Person_3 | Bob | 25 |
COLLECT Person.Age FROM ALL Person
= [ 24, 25 ]SIZE ( COLLECT Person.Age FROM ALL Person )
= 3 (the intermediary collection is [ 24, 25, 25 ]SIZE ( UNIQUE ( COLLECT Person.Age FROM ALL Person ) )
= 2 (the duplicates in the intermediary collection are filtered by theUNIQUE
function)
Now an example with the UNPACK function. We leave out the first instance from the previous example.
Person instance | Person.Name | Person.Age |
---|---|---|
Person_2 | Rick | 25 |
Person_3 | Bob | 25 |
COLLECT Person.Age FROM ALL Person
= [ 25 ]UNPACK ( COLLECT Person.Age FROM ALL Person )
will fail, because the collection contains 2 elements [25,25]UNPACK ( UNIQUE ( COLLECT Person.Age FROM ALL Person ) )
= 25
Only the intermediary results of a COLLECT
statement can contain duplicates. The functions UNIQUE
, UNION
, INTERSECTION
, DIFFERENCE
and SYMMETRIC_DIFFERENCE
always return collections without duplicates.