1 of 44

Graph data formats

SPARQL

This work is licensed under a Creative Commons Attribution 4.0 International License.

2 of 44

SPARQL & SPARQL endpoints

SPARQL

query language for RDF data

SPARQL endpoint

HTTP-based web service
input: SPARQL query
output: data

RDF
CSV
JSON
XML
...

typically including a user form

or use https://yasgui.triply.cc/

2

3 of 44

Existing SPARQL endpoint examples

There is a list on GitHub
Wikidata Query Service is quite popular

Wikidata is a community built database (like Wikipedia is a community built encyclopedia)

The official portal for European data

datasets from all over Europe

In Czechia

National Open Data Catalog
Registr práv a povinností (one of our base registries)
Czech Social Security Administration

User friendly SPARQL query editor Yasgui

can be used instead the less friendly Virtuoso SPARQL editor

3

4 of 44

RDF model: a triple, a statement

<http://example.com/index.html> <http://purl.org/dc/terms/creator> <http://example.com/staff/8574> .

4

http://example.com/index.html

http://example.com/staff/8574

http://purl.org/dc/terms/creator

subject (S)

predicate (P)�(property)

object (O)

Resource / Thing

5 of 44

SPARQL querying - variable

?s <http://purl.org/dc/terms/creator> <http://example.com/staff/8574> .

5

?s

http://example.com/staff/8574

http://purl.org/dc/terms/creator

subject (S)

predicate (P)�(property)

object (O)

Variable

Resource / Thing

6 of 44

SPARQL querying - graph pattern matching

?s�dcterms:creator�staff:8574 .

ex:index.html�dcterms:creator�staff:8574 .

6

?s

staff:8574

dcterms:creator

ex:index.html

staff:8574

dcterms:creator

It’s a match!

7 of 44

SPARQL querying - variable binding

?s�dcterms:creator�staff:8574 .

ex:index.html�dcterms:creator�staff:8574 .

7

?s

staff:8574

dcterms:creator

ex:index.html

staff:8574

dcterms:creator

?s <- ex:index.html

8 of 44

SPARQL querying - more triples

sis:stud1 sis:name "John" ;

sis:age 26 ;

rdf:type sis:Person .

sis:stud2 sis:name "Peter" ;

sis:age 30 ;

rdf:type sis:Person .

sis:stud3 sis:name "Martin" ;

sis:age 20 .

8

?stud rdf:type sis:Person ;

sis:name ?name ;

sis:age ?age .

?stud	?name	?age
sis:stud1	"John"	26
sis:stud2	"Peter"	30

Solutions table

9 of 44

SPARQL querying - optional graph pattern matching

sis:stud1 sis:name "John" ;

sis:age 26 ;

rdf:type sis:Person .�

sis:stud2 sis:name "Peter" ;

sis:age 30 ;

rdf:type sis:Person .�

sis:stud3 sis:name "Martin" ;

sis:age 20 .

9

?stud sis:name ?name ;

sis:age ?age .�OPTIONAL { � ?stud rdf:type ?type.

}

Solutions table

?stud	?name	?age	?type
sis:stud1	"John"	26	sis:Person
sis:stud2	"Peter"	30	sis:Person
sis:stud3	"Martin"	20	NOT BOUND

10 of 44

SPARQL querying - multiple optional graph patterns

?stud rdf:type ?type .

OPTIONAL {

?stud sis:age ?age .

}

OPTIONAL {

?stud sis:name ?name .

}

10

OPTIONAL { ?dataset dcterms:temporal ?temporal .

OPTIONAL { ?temporal dcat:startDate ?startDate . }

OPTIONAL { ?temporal dcat:endDate ?endDate . }

OPTIONAL { ?temporal schema:startDate ?schemaStartDate . }

OPTIONAL { ?temporal schema:endDate ?schemaEndDate . }

BIND(IF(BOUND(?startDate), ?startDate, ?schemaStartDate) AS ?finalStartDate)

BIND(IF(BOUND(?endDate), ?endDate, ?schemaEndDate) AS ?finalEndDate)

}

OPTIONAL { ?dataset dcat:contactPoint ?cp .

?cp a ?cptype.

OPTIONAL { ?cp vcard:fn ?cpfn . }

OPTIONAL { ?cp vcard:hasEmail ?cpemail . }

}

OPTIONAL { ?dataset foaf:page ?page . }

OPTIONAL { ?dataset dcterms:conformsTo ?conformsTo . }

OPTIONAL { ?dataset dcat:spatialResolutionInMeters ?spatialResolution . }

OPTIONAL { ?dataset dcat:temporalResolution ?temporalResolution . }

OPTIONAL { ?dataset dcterms:isPartOf ?topDataset . }

11 of 44

SPARQL querying - union graph pattern matching

sis:stud1 sis:name "John" ;

sis:age 26 ;

rdf:type sis:Person .�

sis:stud2 sis:name "Peter" ;

sis:age 30 ;

rdf:type sis:Person .

sis:stud3 sis:name "Martin" ;

sis:age 20 .

11

{�?stud sis:name ?name ;

sis:age ?age .�}�UNION�{ �?stud sis:name ?name ;

sis:age ?age ;� rdf:type ?type.

}

?stud	?name	?age	?type
sis:stud1	"John"	26	sis:Person
sis:stud2	"Peter"	30	sis:Person
sis:stud3	"Martin"	20	NOT BOUND
sis:stud1	"John"	26	NOT BOUND
sis:stud2	"Peter"	30	NOT BOUND

12 of 44

SPARQL - First complete query

PREFIX sis: <http://is.cuni.cz/studium/sis#>

SELECT ?name ?age

WHERE {

?stud a ?type ;

sis:name ?name ;

sis:age ?age .

}

12

?name	?age
"John"	26
"Peter"	30

Query result

13 of 44

SPARQL - Prologue

Prefix

Like in Turtle, but without @ and without "."

PREFIX name: <http://www.my.cz/>

…

WHERE {

name:local ?p ?o.

}

Relative URIs

Like in Turtle, but without @ and without "."

BASE <http://www.my.cz/>

…

WHERE {

<sis> <http://www.my.cz/sis> ?o

}

13

14 of 44

SPARQL - Blank nodes

Blank nodes in data

Distinct nodes within the document scope

Blank nodes in query results

Distinct nodes within the result scope

14

Blank node identifiers in the result may not correspond to original blank node identifiers from the given document

15 of 44

SPARQL - (named) graph patterns

PREFIX foaf: <http://xmlns.com/foaf/0.1/>

SELECT DISTINCT ?g

WHERE {

GRAPH ?g {

?s a foaf:Person .

}

PREFIX foaf: <http://xmlns.com/foaf/0.1/>

SELECT DISTINCT ?s

WHERE {

GRAPH <https://...> {

?s a foaf:Person .

}

15

16 of 44

SPARQL - literals

Language tags

"Praha"
"Praha"@cs
"Prague"@en

Typed literals

1 = "1"^^xsd:integer
1.5 = "1.5"^^xsd:decimal
1.0E6 = "1"^^xsd:double
true = "true"^^xsd:boolean

16

17 of 44

SPARQL - Simple query with FILTER

PREFIX sis: <http://is.cuni.cz/studium/sis#>

SELECT ?name ?age

WHERE {

?stud a ?type ;

sis:name ?name ;

sis:age ?age .

FILTER (?age > 27)

}

17

?name	?age
"John"	26
"Peter"	30

Query result

18 of 44

Example: Simple SPARQL query

Select URI and labels of all skos:ConceptSchemes from a given SPARQL endpoint, which are labeled „Paragraf“.

PREFIX skos: <http://www.w3.org/2004/02/skos/core#>

SELECT *

WHERE {

?s a skos:ConceptScheme ;

skos:prefLabel ?label .

FILTER (?label = "Paragraf"@cs)

}

18

19 of 44

SPARQL: logical expression operators

The usual operators

Arithmetic operators

Unary + –
Binary + – * /

Comparison operators

< <= >= >
= !=

Logical connectives

! && ||

RDF specific operators

Other operators

bound, isIri, isBlank, isLiteral

Access functions

str, lang, dataType

19

20 of 44

SPARQL - property paths example

Select first 100 inspections of CTIA, which have a fine as a result.�Return its value.

20

PREFIX schema: <http://schema.org/>

PREFIX gr: <http://purl.org/goodrelations/v1#>�

SELECT ?check ?fine

WHERE {

?check a schema:CheckAction ;

schema:result ?sanction .

?sanction schema:result ?price .

?price gr:hasCurrencyValue ?fine .

}

LIMIT 100

ex:Check

ex:Sanction

gr:PriceSpecification

40000

schema:result

gr:hasCurrencyValue

21 of 44

SPARQL - property paths example

Select first 100 inspections of CTIA, which have a fine as a result.�Return its value.

21

PREFIX schema: <http://schema.org/>

PREFIX gr: <http://purl.org/goodrelations/v1#>

SELECT ?check ?fine

WHERE {

?check a schema:CheckAction ;

schema:result/schema:result/gr:hasCurrencyValue ?fine .

}

LIMIT 100

ex:Check

ex:Sanction

gr:PriceSpecification

40000

schema:result

gr:hasCurrencyValue

22 of 44

SPARQL - property paths

Sequences

path1/path2

Alternatives

path1|path2

Groups

(path)

Negation

!item
!(item1|item2|…)

Occurrences

path*
path+
path?

Inverse paths (from object to subject)

^path

22

23 of 44

SPARQL - aggregation example

What was the highest issued fine?

PREFIX schema: <http://schema.org/>

PREFIX gr: <http://purl.org/goodrelations/v1#>

SELECT (MAX (?fine) AS ?max)

WHERE

{

?check a schema:CheckAction ;

schema:result/schema:result/gr:hasCurrencyValue ?fine .

}

23

24 of 44

SPARQL - aggregation

COUNT
SUM
AVG
MIN
MAX
GROUP_CONCAT(?x ; separator="|")
SAMPLE

24

25 of 44

SPARQL - federated query

PREFIX foaf: <http://xmlns.com/foaf/0.1/>

SELECT ?s ?name

WHERE {

?s a foaf:Person .

SERVICE <https://peoplenames.org/sparql> {

?s foaf:name ?name .

}

25

26 of 44

SPARQL - binding variables

PREFIX dct: <http://purl.org/dc/terms/>

PREFIX ns: <http://example.org/ns#>

SELECT ?title ?price

WHERE { ?product ns:price ?p ;

ns:discount ?discount ;

dct:title ?title .

BIND (?p * ( 1 - ?discount) AS ?price)

FILTER(?price < 20)

}

26

27 of 44

SPARQL - binding variables example

Select first 100 inspections of CTIA, which have a fine as a result.�Return its value and whether it is higher than 1000.

PREFIX schema: <http://schema.org/>

PREFIX gr: <http://purl.org/goodrelations/v1#>

SELECT ?check ?fine ?higher

WHERE {

?check a schema:CheckAction ;

schema:result/schema:result/gr:hasCurrencyValue ?fine .

BIND(IF(?fine > 1000, true, false) AS ?higher)

}

LIMIT 100

27

28 of 44

SPARQL - functions

if

IF(?x = 2, "yes", "no")

evaluates first parameter
If true, returns second parameter
If false, returns third parameter

coalesce

COALESCE(1/0, ?x, ...)

returns first parameter which does not result in error

exists { graph pattern }
not exists { graph pattern }
IN, NOT IN

2 IN (<http://example/iri>, "str", 2.0) = true

28

29 of 44

SPARQL - functions on RDF terms

isIRI
isBlank
isLiteral
isNumeric
str
lang
datatype
IRI
STRDT
STRLANG
UUID

29

UUID()	<urn:uuid:b9302fb5-642e-4d3b-af19-29a8f6d894c9>

STRLANG("chat", "en")	"chat"@en

STRDT("123", xsd:integer)	"123"^^<http://www.w3.org/2001/XMLSchema#integer>
STRDT("iiii", <http://ex/roman>)	"iiii"^^<http://ex/roman>

30 of 44

SPARQL - functions on strings

STRLEN
SUBSTR
UCASE, LCASE
STRSTARTS, STRENDS
CONTAINS
STRBEFORE, STRAFTER
ENCODE_FOR_URI
CONCAT
langMatches
REGEX
REPLACE

30

substr("foobar"@en, 4, 1)	"b"@en

strbefore("abc","b")	"a"
strbefore("abc"@en,"bc")	"a"@en
strbefore("abc"@en,"b"@cy)	error

encode_for_uri("Los Angeles")	"Los%20Angeles"

replace("abcd", "b", "Z")	"aZcd"
replace("abab", "B", "Z","i")	"aZaZ"
replace("abab", "B.", "Z","i")	"aZb"

31 of 44

SPARQL - another binding variables example

Create IRI for entities with IDs�the Frequency EU vocabulary

�PREFIX dcterms: <http://purl.org/dc/terms/>

PREFIX nkod: <https://data.gov.cz/slovník/nkod/>

SELECT ?frequency

WHERE

{

?dataset nkod:accrualPeriodicity ?freq .

BIND(IRI(CONCAT("http://publications.europa.eu/resource/authority/frequency/", ?freq)) AS ?frequency)

}

31

?freq
MONTHLY
ANNUAL
BIENNIAL
…

<http://publications.europa.eu/resource/authority/frequency/MONTHLY>�<http://publications.europa.eu/resource/authority/frequency/ANNUAL>�…

32 of 44

SPARQL - functions on numerics

abs
round
ceil
floor
RAND

32

33 of 44

SPARQL - functions on dates and times

now
year
month
day
hours
minutes
seconds
timezone
tz

33

timezone("2011-01-10T14:45:13.815-05:00"^^xsd:dateTime)	"-PT5H"^^xsd:dayTimeDuration

tz("2011-01-10T14:45:13.815-05:00"^^xsd:dateTime)	"-05:00"

now()	"2011-01-10T14:45:13.815-05:00"^^xsd:dateTime

day("2011-01-10T14:45:13.815-05:00"^^xsd:dateTime)	10

34 of 44

SPARQL - subqueries

Inner queries evaluated first.

PREFIX : <https://example.org/>

SELECT ?company ?maxFine

WHERE {

?company a schema:Organization .

{

SELECT ?company (MAX(?fine) AS ?maxFine)

WHERE {

?company :fine ?fine .

} GROUP BY ?company

}

34

35 of 44

SPARQL - VALUES example

35

@prefix dct: <http://purl.org/dc/terms/> .

@prefix : <http://example.org/book/> .

@prefix ns: <http://example.org/ns#> .

:book1 dct:title "SPARQL Tutorial"@en ;

ns:price 42 .

:book2 dct:title "The Semantic Web"@en ;

ns:price 23 .

PREFIX dct: <http://purl.org/dc/terms/>

PREFIX : <http://example.org/book/>

PREFIX ns: <http://example.org/ns#>

SELECT ?book ?title ?price

WHERE {

VALUES ?book { :book1 :book3 }

?book dct:title ?title ;

ns:price ?price .

}

book	title	price
<http://example.org/book/book1>	"SPARQL Tutorial"	42

36 of 44

SPARQL - VALUES example - codelist translation

PREFIX dcat: <http://www.w3.org/ns/dcat#>

PREFIX dcterms: <http://purl.org/dc/terms/>

PREFIX skos: <http://www.w3.org/2004/02/skos/core#>

PREFIX f: <http://publications.europa.eu/resource/authority/frequency/>

CONSTRUCT {

?dataset dcterms:accrualPeriodicity ?new.

?new a skos:Concept, dcterms:Frequency .

}

WHERE {

?dataset dcterms:accrualPeriodicity ?f .

VALUES (?f ?new) {

("R/P1M" f:MONTHLY)

("R/P1Y" f:ANNUAL)

("R/P2Y" f:BIENNIAL)

…

("jednorázově" f:NEVER)

("nikdy" f:NEVER)

}

36

<#dataset1> dcterms:accrualPeriodicity "R/P1M" .

<#dataset2> dcterms:accrualPeriodicity "R/P2Y" .

<#dataset1> dcterms:accrualPeriodicity f:MONTHLY .

f:MONTHLY a skos:Concept, dcterms:Frequency .

<#dataset2> dcterms:accrualPeriodicity f:BIENNIAL .

f:BIENNIAL a skos:Concept, dcterms:Frequency .

37 of 44

SPARQL - result forms - SELECT and ASK

SELECT

Result: Solutions table (e.g. in CSV or JSON)

ASK

Checks whether at least one result item exists
Result: True or false

Is there a CTIA inspection with a resulting fine higher than 5 000 000?��PREFIX schema: <http://schema.org/>�PREFIX gr: <http://purl.org/goodrelations/v1#>�PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>�ASK {�?check a schema:CheckAction ;� schema:result/schema:result/gr:hasCurrencyValue ?fine .�FILTER(?fine > 5000000)�}

37

38 of 44

SPARQL - result forms - CONSTRUCT

Result: RDF graph constructed from a template.

CONSTRUCT

{ ?s sis:name ?fullName }

WHERE {

?s sis:firstName ?n1 ;

sis:lastName ?n2 .

BIND(STRLANG(CONCAT(?n1, " ", ?n2), "cs") AS ?fullName)

}

38

39 of 44

SPARQL - result forms - DESCRIBE

Result: RDF graph about the given resource.

Specification non-normative - various triplestores implement it differently.

Example: DESCRIBE <http://www.my.cz/>

39

40 of 44

SPARQL SELECT - modifiers

ORDER BY

Orders items in the solutions table
Hierarchical ordering criteria

ASC = ascending (default)
DESC = descending

Unbound variable < blank node < URI < literal
Example

ORDER BY ASC(?name), DESC(?age)

40

41 of 44

SPARQL SELECT - ORDER BY example

PREFIX : <http://example.org/ns#>

PREFIX foaf: <http://xmlns.com/foaf/0.1/>

SELECT ?name

WHERE { ?x foaf:name ?name ;

:empId ?emp }

ORDER BY ?name DESC(?emp)

41

42 of 44

SPARQL SELECT - modifiers

LIMIT

Limits the number of items in the result sequence
Always should be preceded by ORDER BY modifier

Because RDF graph is a set with no ordering

Used as page size
Example: ORDER BY ?name LIMIT 10

OFFSET

Index of the first reported item from the sequence
Used for paging through results
Example: ORDER BY ?name LIMIT 10 OFFSET 20

42

43 of 44

SPARQL SELECT - GROUP BY, HAVING

PREFIX : <http://data.example/>

SELECT (AVG(?size) AS ?asize)

WHERE {

?x :size ?size

}

GROUP BY ?x

HAVING(AVG(?size) > 10)

43

44 of 44

SPARQL 1.1

W3C Recommendation

21 March 2013
Set of recommendations

SPARQL 1.1 Overview
SPARQL 1.1 Query Language
SPARQL 1.1 Update
SPARQL 1.1 Service Description
SPARQL 1.1 Federated Query
SPARQL 1.1 Query Results JSON Format
SPARQL 1.1 Query Results CSV and TSV Formats
SPARQL Query Results XML Format (Second Edition)
SPARQL 1.1 Entailment Regimes
SPARQL 1.1 Protocol
SPARQL 1.1 Graph Store HTTP Protocol

44