Logical Functions
This topic describes the logical functions in AnzoGraph DB.
- AND: Evaluates two logical expressions and returns true if both expressions are true.
- BOUND: Evaluates whether an RDF term type is bound.
- CASE: Evaluates a series of conditions and returns the matching result.
- COALESCE: Evaluates a number of expressions and returns the results for the first expression that is bound and does not raise an error.
- EXISTS: Evaluates whether the specified pattern exists.
- IF: Evaluates a condition and returns the specified result depending on the outcome of the test.
- IN: Evaluates whether the specified RDF term is found in any of the given test values.
- NOT: Evaluates whether the specified logical expression is not true.
- OR: Evaluates two logical expressions and returns true if at least one of the expressions is true.
- PARTITIONINDEX: Returns the zero-based index of the bucket in which the specified value falls.
- SAMETERM: Evaluates whether two RDF term type values are the same.
- UNBOUNDED: Extends the SPARQL UNDEF functionality to enable users to include an undefined value as a function argument.
AND
This function evaluates two logical expressions. If both expressions are true, the function returns true. If one or both arguments are false, the function returns false.
Syntax
AND(logical_expression1, logical_expression2)
| Argument |
Data Type |
Description |
| logical_expression1 |
evaluates to boolean |
The first logical expression to evaluate. |
| logical_expression2 |
evaluates to boolean |
The second logical expression to evaluate. |
Returns
| Data Type |
Description |
| boolean |
True if both conditions are true and false if either condition is false. |
BOUND
This function evaluates whether the specified RDF term has a value bound to it.
Syntax
BOUND(term)
| Argument |
Data Type |
Description |
| term |
RDF term |
The literal, URI, or blank node value to evaluate. |
Returns
| Data Type |
Description |
| boolean |
True if the term is bound and false if it is not. |
CASE
This function enables you to add IF/THEN logic to a query. A CASE expression evaluates a series of conditions and returns the matching result. You can use CASE expressions wherever expressions are valid in SPARQL queries.
Syntax
There are two variations of CASE statements: simple and generic. Use the simple form to compare the results of an expression with a series of tests and return a result when a test returns true. Use the generic form when evaluating a larger range of tests with multiple conditions.
Simple Form
CASE expression_to_compare
WHEN expression1 THEN result1
WHEN expression2 THEN result2
[ WHEN expressionN THEN resultN ]
[ ELSE result_when_false ]
END
| Argument |
Data Type |
Description |
| expression_to_compare |
evaluates to boolean |
The expression to evaluate and compare its results with the subsequent expressions. |
| expression1–N |
evaluates to boolean |
The expressions to evaluate against expression_to_compare. |
| result1–N |
any |
The result to return when the corresponding expression is true. |
| result_when_false |
any |
An optional value to be returned if none of the specified expressions are true. |
Generic Form
CASE WHEN condition1 THEN result1
WHEN condition2 THEN result2
[ WHEN conditionN THEN resultN ]
[ ELSE result_when_false ]
END
| Argument |
Data Type |
Description |
| condition1–N |
evaluates to boolean |
The conditions to test. |
| result1–N |
any |
The result to return when the corresponding condition passes. |
| result_when_false |
any |
An optional value to be returned if none of the specified conditions pass. |
Returns
| Data Type |
Description |
| The type of the specified result |
The specified results according to the evaluation of the conditions. |
Example
The following example uses a CASE statement to determine and report on whether the top 10 events (with the most tickets sold) sold out.
PREFIX tickit: <http://anzograph.com/tickit/>
SELECT ?event ?venue ?seats
(( CASE WHEN (?seats <= (sum(?qty))) then "yes"
WHEN (?seats > (sum(?qty))) then "no"
END ) as ?sold_out)
FROM <http://anzograph.com/tickit>
WHERE {
?sales tickit:qtysold ?qty .
?sales tickit:eventid ?eventid .
?eventid tickit:eventname ?event .
?eventid tickit:venueid ?venueid .
?venueid tickit:venuename ?venue .
?venueid tickit:venueseats ?seats.
}
GROUP BY ?event ?venue ?seats ?qty
ORDER BY desc(?qty)
LIMIT 10event | venue | seats | sold_out
------------------+------------------------------+-------+----------
Simple Plan | Hubert H. Humphrey Metrodome | 64035 | no
Black Crowes | Yankee Stadium | 52325 | no
Hot Tuna | Turner Field | 50091 | no
Marc Anthony | Georgia Dome | 71149 | no
Mark Knopfler | Edward Jones Dome | 66965 | no
Spoon | Dolphin Stadium | 74916 | no
Armando Manzanero | Texas Stadium | 65595 | no
Missy Higgins | Great American Ball Park | 42059 | no
Zombies | Lambeau Field | 72922 | no
Hannah Montana | Monster Park | 69843 | no
10 rows
COALESCE
This function evaluates a number of expressions and returns the results for the first expression that is bound and does not raise an error.
Syntax
COALESCE(expression1 [, expression2 ] [, expressionN ] )
| Argument |
Data Type |
Description |
| expression1–N |
RDF term |
The literal, URI, or blank node expressions to evaluate. |
Returns
| Data Type |
Description |
| RDF term |
The result of the first expression that is bound and does not error. |
EXISTS
This function evaluates whether the specified pattern exists in the data.
Syntax
EXISTS { graph_pattern }
Returns
| Data Type |
Description |
| boolean |
True if the pattern exists and false if it does not. |
IF
This function evaluates a condition and returns the specified result depending on the outcome of the test. If the condition evaluates to true, the first result is returned. If the condition evaluates to false, the second result is returned. And if the condition results in an error, the third result is returned.
Syntax
IF(logical_expression, true_result, false_result [, error_result ])
| Argument |
Data Type |
Description |
| logical_expression |
evaluates to boolean |
The condition that evaluates to true or false. |
| true_result |
RDF term |
The value that defines the result to return if the condition evaluates to true. |
| false_result |
RDF term |
The value that defines the result to return if the condition evaluates to false. |
| error_result |
RDF term |
An optional value that defines the result to return if the condition evaluates to an error. If the condition results in an error and error_result is not specified, logical_expression(error) is returned. |
Returns
| Data Type |
Description |
| RDF term |
The result based on the evaluation of the condition. |
IN
This function evaluates whether the specified RDF term type value is found in any of the given test values.
Syntax
IN(term, test_value1 [, test_value2 ] [, test_valueN])
| Argument |
Data Type |
Description |
| term |
RDF term |
The literal, URI, or blank node value to look for in the test values. |
| test_value1–N |
RDF term |
The literal, URI, or blank node values to look for the specified term in. |
Returns
| Data Type |
Description |
| boolean |
True if the given term is found in the test values and false if it is not. |
Example
The example below queries the sample Tickit data set to return the names of people who were born in the year 1975.
PREFIX tickit: <http://anzograph.com/tickit/>
SELECT ?birthday (concat(?fname, ?lname) AS ?name)
FROM <http://anzograph.com/tickit>
WHERE {
?s tickit:firstname ?fname .
?s tickit:lastname ?lname .
?s tickit:birthday ?birthday.
FILTER ((YEAR(?birthday)) IN (1975))
}
ORDER BY ?birthdaybirthday | name
-----------+---------------------
1975-01-01 | MaryamWeeks
1975-01-01 | OliverHammond
1975-01-01 | MacKenzieBaldwin
1975-01-01 | XenosBaxter
1975-01-01 | XanderWilson
1975-01-01 | HadleyBush
1975-01-01 | RajaTodd
1975-01-02 | AlecFitzgerald
1975-01-02 | QuinnBuckley
1975-01-03 | RhiannonBooth
1975-01-03 | LanaLeonard
1975-01-03 | DeirdreWheeler
...
763 rows
NOT
This function evaluates whether the specified logical expression is not true.
Syntax
NOT(logical_expression)
| Argument |
Data Type |
Description |
| logical_expression |
evaluates to boolean |
The condition to evaluate. |
Returns
| Data Type |
Description |
| boolean |
True if the condition is false and false if it is true. |
OR
This function evaluates two logical expressions. If at least one expression is true, the function returns true. If both expressions are false, the function returns false.
Syntax
OR(logical_expression1, logical_expression2)
| Argument |
Data Type |
Description |
| logical_expression1 |
evaluates to boolean |
The first logical expression to evaluate. |
| logical_expression2 |
evaluates to boolean |
The second logical expression to evaluate. |
Returns
| Data Type |
Description |
| boolean |
True if one or both conditions are true and false if both conditions are false. |
PARTITIONINDEX
This function returns the zero-based index of the bucket in which the specified value falls. Buckets start at the specified start value and are sized according to the specified interval. The first bucket is [start, start+interval). That means it is closed on the low end and open on the high end. PARTITIONINDEX returns less than 0 if the value does not fall into any bucket, such as when the given value is less than start or if the comparison is indeterminate for date and time data types.
Syntax
PARTITIONINDEX(value, start, interval)
| Argument |
Data Type |
Description |
| value |
literal |
The literal value for which to determine the zero-based index. |
| start |
literal |
The literal value that indicates the start of the first bucket. |
| interval |
literal |
The literal value that specifies the size of the bucket. |
Returns
| Data Type |
Description |
| long |
The zero-based index of the bucket in which the specified value exists. |
SAMETERM
This function evaluates whether two RDF term type values are the same.
Syntax
SAMETERM(term1, term2)
| Argument |
Data Type |
Description |
| term1 |
RDF term |
The first literal, URI, or blank node value to compare. |
| term2 |
RDF term |
The literal, URI, or blank node value to compare to term1. |
Returns
| Data Type |
Description |
| boolean |
True if the terms are the same and false if they are not. |
UNBOUNDED
This function is like the SPARQL UNDEF keyword but extends that functionality to enable users to include an undefined value as a function argument, as UNDEF is only supported in VALUES clauses.
Syntax
UNBOUNDED()
Returns
| Data Type |
Description |
| RDF term |
The specified result according to the evaluation of the condition. |
Example
The following example statement incorporates UNBOUNDED to return null if the specified condition (?x > 5) fails:
BIND(IF(?x > 5 , "Win", UNBOUNDED()) as ?testResult)
In this case, ?testResult is bound if ?x is greater than 5. If ?x is not greater than 5, ?testResult is not bound.