Sparq lreference 1.8-us

3. SPARQL Query Language Reference

SPARQL RDF Query Language Reference v1.8 Based on SPARQL Query Language 23 November 2005
<http://www.w3.org/TR/2005/WD-rdf-sparql-query-20051123/>.
Copyright © 2005, 2006 Dave Beckett.
Latest version: <http://www.dajobe.org/2005/04-sparql/> RDF Term: A part of an RDF Triple. An IRI, Blank Node or a Literal.
<uri> _:b1 “Literal”@en “abc123”^^my:datatype
Comments to: dave@dajobe.org
Query Variable: Identifiers for binding to RDF Terms in matches.
?a / $b or in lists: $name $title $place
1. RDF Model and SPARQL RDF Terms Syntax Anonymous Blank Nodes in a graph pattern act as variables that cannot be SELECTed
Query Variable: _:abc
Triple Pattern: An RDF Triple with Query Variables or blank nodes allowed in each term:
RDF Graph: A set of RDF Triples <http://example.org/abc> ?x “Hello”
?subject ?predicate ?object
RDF Triple: A triple (3-tuple) of:
Turtle abbreviations can be used for Triple Patterns, see Section 4.
Graph Pattern: A block that matches part of the queried RDF graph.
Subject: IRI
Basic A set of Triple Patterns binding RDF Terms in the graph to variables.
or Blank Node
Graph Pattern: Written as a {..} block with '.' separating the triple patterns:
{ <http://example.org/abc> ?y “Hello” .
Predicate: IRI ?subject $predicate “Literal” }
Group A graph pattern containing multiple graph patterns which must all match
Object: IRI or Blank Node Graph Pattern: { { ?person rdf:type foaf:Person }
or Literal { ?person foaf:name “Dave” } }
Optional A graph pattern which may fail to match and provide bindings but not
Graph Pattern: cause the entire query to fail. Written with OPTIONAL before a graph
URI: An absolute IRI which may include a # fragment. pattern.
<http://www.w3.org/> OPTIONAL { ?person foaf:nick ?nick }
<http://example.org/#fragment> Union A pair of graph patterns any of which may match and bind the same
<abc.rdf> Relative IRI resolved against base IRI. Graph Pattern: variables. Written with the UNION keyword between two graph patterns.
<> Base IRI, usually the query document IRI { ?node ex:name ?name } UNION
ex:name IRI shorthand using XML-style prefix ex and local name. { ?node vcard:FN ?name }
Declared with PREFIX (SPARQL) or @prefix (Turtle) Graph A keyword for specifying a graph name to use or to return a graph name
Graph Pattern: as a binding. Written with the GRAPH keyword before a graph pattern.
RDF Literal: A Unicode string with an optional language tag. GRAPH <http://example.org/myfoaf>
“hello” “bonjour”@fr { ?person foaf:name ?name }
GRAPH ?graph { ?person foaf:name ?name }
RDF Typed A Unicode string and datatype IRI for encoding datatypes. Value Constraints: A boolean expression in a graph pattern over query variables that
Literal: “abc”^^<http://example.org/myDatatype> constrains matched graph patterns.
abbreviated with an XML QName style as: { ?item ex:size $size . FILTER ($size < 10 ) }
“10”^^xsd:integer
Short forms for several common datatypes: 4. SPARQL Query Language Structure
-10 “-10”^^xsd:integer
1.2345 “1.2345”^^xsd:decimal Prologue (optional) BASE <iri>
true “true”^^xsd:boolean PREFIX prefix: <iri> (repeatable)
Query Result forms (required, pick 1) SELECT (DISTINCT)sequence of ?variable
Blank Node: A node in a graph with a local name. The scope of the name is the RDF graph.
SELECT (DISTINCT)*
_:node
DESCRIBE sequence of ?variable or <iri>
DESCRIBE *
2. Common RDF Namespaces and Prefixes CONSTRUCT { graph pattern }
ASK
Namespace Common Prefix Namespace URI
RDF rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# Query Dataset Sources (optional) Add triples to the background graph (repeatable):
Dublin Core dc: http://purl.org/dc/elements/1.1/ FROM <iri>
FOAF foaf: http://xmlns.com/foaf/0.1/ Add a named graph (repeatable):
XML Schema Datatypes xsd: http://www.w3.org/2001/XMLSchema# FROM NAMED <iri>
RDFS rdfs: http://www.w3.org/2000/01/rdf-schema# Graph Pattern (optional, required for ASK) WHERE { graph pattern [ FILTER expression ]}
OWL owl: http://www.w3.org/2002/07/owl# Query Results Ordering (optional) ORDER BY ...
Query Results Selection (optional) LIMIT n, OFFSET m

5. SPARQL Query Result Forms 8. Turtle RDF Syntax Reference (Turtle 2006-01-2121 <http://www.dajobe.org/2004/01/turtle/>)
Variable Bindings: A sequence of (set of variable bindings) for each query pattern match. Turtle (Terse RDF Triple Language) describes triples in an RDF graph and allows abbreviations.
SELECT * Triple Patterns in SPARQL can use the same abbreviations.
WHERE { $a rdf:type $b }
to ask for bindings for all variables mentioned in the query and RDF Terms:
SELECT $a ?b IRI < IRI > (<> is the base IRI, often the document IRI)
WHERE { $a rdf:type ?b } Literal: ”string” or ”string”@language or ^^< datatype IRI >
to list them explicitly Blank Node: _: name or [] for an anonymous blank node
RDF Graph:
Describe An RDF graph describing resources either given by URI @prefix operator: IRIs can be written as XML-style QNames by defining a prefix / IRI binding:
Resources: DESCRIBE <http://example.org/thing> @prefix dc: <http://purl.org/dc/elements/1.1/> .
or by binding variables using the same syntax as SELECT.
DESCRIBE ?person Triples: 3 RDF terms with whitespace separating them as necessary, and '.' between triples:
WHERE { ?person foaf:name “Dave” } <> dc:title "SPARQL Reference" .
Build an An RDF graph made by substituting variables into a triple template. <> dc:date “2006-02-06”^^xsd:dateTime .
RDF graph CONSTRUCT { ?a foaf:knows ?b }
, operator: Triples with the same subject and predicate may be abbreviated with ',':
WHERE { ?a ex:KnowsQuiteWell ?b }
Boolean: True if the query pattern could be answered. <http://example.org/book> dc:title “My Book”, “Mein Buch”@de .
ASK ; operator: Triples with the same subject may be abbreviated with ';':
WHERE { ?a rdf:type foaf:Person } <http://work.example.org/> dc:title “My Workplace”;
dc:publisher “My Employer” .
6. Query Results Ordering and Modifying
[ ... ] operator: A sequence of (predicate object) pairs may be put inside [ ... ] and a blank node
The optional modifications on query results are performed in the following order: subject will be assigned to them:
1. DISTINCT to ensure solutions in the sequence are unique <> dc:creator [ foaf:name “Dave”; foaf:homePage <http:...> ] .
1. ORDER BY ordering solutions sequences by variable, expression or extension function call: [] operator: A blank node:
ORDER BY DESC(?date) ?title ASC(?familyName) my:fn(?a) [] a ex:Book [ dc:title “Thing”; dc:description “On shelf” ] .
in descending order by date, by ascending title order, by familyName ascending, by extension a predicate: The common rdf:type QName may be abbreviated by the keyword a as a predicate:
function <> a Foaf:Document .
2. LIMIT n to restrict the number of solutions to n
3. OFFSET m to start the results in the solution from item m Decimal integers: Positive or negative decimal integers can be written as (type xsd:integer)
<> ex:sizeInBytes 12345 .
7. Values – datatypes, expressions and operators Decimal numbers: Positive or negative decimal numbers can be written as (type xsd:decimal)
<> ex:shoeSize 8.5 .
Supported datatypes: RDF Terms, xsd:boolean, xsd:string, xsd:double,
xsd:float, xsd:decimal, xsd:integer and xsd:dateTime ( ... ) collections: RDF collections can be written inside ( ... ) as space-separated lists of contents:
<> ex:contents ( ex:apple ex:banana ex:pear ) .
Logical operators: Logical: A|| B, A && B, !A, (A)
Comparison (A op B): =, !=, <, >, <=, >= 9. Example SPARQL Query
Arithmetic operators: Unary: +A, -A BASE <http://example.org/>
Binary (A op B): +, -, *, / PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
RDF operators: Boolean: BOUND(A), isIRI(A) / isURI(A), PREFIX foaf: <http://xmlns.com/foaf/0.1/>
isBlank(A), isLiteral(A) # This is a relative IRI to BASE above
String: STR(A), LANG(A), DATATYPE(A) PREFIX ex: <properties/1.0#>
String Match operator: REGEX(string expression, pattern expression
[,flags expression] ) SELECT DISTINCT $person ?name $age
pattern syntax is from XQuery 1.0 / XPath 2.0, FROM <http://rdf.example.org/personA.rdf>
FROM <http://rdf.example.org/personB.rdf>
XML Schema, similar to Perl. flags are s, m, i, x WHERE { $person a foaf:Person ;
Extension Functions and QName( expression, expression, ...) foaf:name ?name.
Explicit Type Casting: OPTIONAL { $person ex:age $age } .
Automatic Type from xsd:decimal to xsd:float FILTER (!REGEX(?name, “Bob”))
Promotion: from xsd:float to xsd:double }
ORDER BY ASC(?name) LIMIT 10 OFFSET 20

Sparq lreference 1.8-us

More Related Content

Viewers also liked (6)

Similar to Sparq lreference 1.8-us (20)

More from Ajay Ohri (20)

Sparq lreference 1.8-us