�ݺ�ߣ

SPARQL 1.1
Update Language

SPARQL 1.1 Update

Overview
• SPARQL 1.1 Update Overview
• Graph Updates
–
–
–
–
–

Insert
Delete
Delete/Insert
Load
Clear

• Graph Management
– Create
– Drop
– Copy, Move, Add
SPARQL 1.1 Update

#2

SPARQL 1.1 Update Overview
• SPARQL 1.1 increases further the prominence of
named graphs in RDF
• It offers three ways to affect explicitly specified data
within a graph (or graphs):
– INSERT DATA
– DELETE DATA
– DELETE/INSERT … WHERE …

• It also offers two ways to further change the data
within a graph:
– LOAD – CLEAR

• It offers a set of graph management operations:
– CREATE – DROP – COPY – MOVE – ADD
SPARQL 1.1 Update

#3

INSERT DATA
• The most basic part of SPARQL 1.1 Update is to allow
data to be inserted into a store, e.g.:
INSERT DATA {ou:barry foaf:givenName "Barry";
foaf:familyName "Norton";
foaf:mbox <mailto:barry.norton@sti2.at>}

• While a default graph may be supported, it is common
to manage data in a named graph:
INSERT DATA { GRAPH <http://ontotext.com/ns#>
{onto:barry1 foaf:givenName "Barry";
foaf:familyName "Bishop";
foaf:mbox <barry.bishop@ontotext.com>}}
SPARQL 1.1 Update

#4

DELETE DATA
• SPARQL 1.1 Update also allows deletion of explicitly
described triples:
DELETE DATA {sti:barry foaf:mbox
<mailto:barry.norton@sti2.at>}

• Note that no blank nodes are allowed in these DELETE
queries since these match any resource in SPARQL
(Cf.:
SELECT ?mbox WHERE {_:barry foaf:mbox ?mbox}
mbox
<mailto:b.j.norton@open.ac.uk>
SPARQL 1.1 Update

)
#5

DELETE/INSERT
• SPARQL 1.1 extends the use of graph patterns for use
in parallel INSERT and DELETE, according to matches,
e.g.:
DELETE {?barry foaf:mbox <mailto:barry.norton@sti2.at>}
INSERT { GRAPH <http://ontotext.com/ns#>
{onto:barry foaf:givenName ?forename;
foaf:familyName ?surname;
foaf:mbox <mailto:barry.norton@ontotext.com>}}
WHERE {?barry foaf:mbox <mailto:barry.norton@sti2.at>;
foaf:givenName ?forename;
foaf:familyName ?surname}
}
SPARQL 1.1 Update

#6

DELETE WHERE
• SPARQL 1.1 offers a simpler form of INSERT/DELETE
when just deleting:
DELETE {?barry foaf:mbox <mailto:barry.norton@sti2.at>}
WHERE {?barry foaf:mbox <mailto:barry.norton@sti2.at>}

DELETE
WHERE {?barry foaf:mbox <mailto:barry.norton@sti2.at>}

SPARQL 1.1 Update

#7

USING/WITH
• SPARQL Update uses USING (NAMED) in role of
FROM (NAMED) with respect to WHERE clauses
INSERT { GRAPH <http://ontotext.com/ns#>
{?barry foaf:nick "Baz"}}
USING <http://ontotext.com/ns#>
WHERE {?barry foaf:givenName "Barry"}

• It also offers a simpler way to deal with the same
named graph throughout a DELETE/INSERT query
WITH <http://ontotext.com/ns#>
INSERT {?barry foaf:nick "Baz"}
WHERE {?barry foaf:givenName "Barry"}
SPARQL 1.1 Update

#8

LOAD/CLEAR
• SPARQL 1.1 allows an RDF graph to be loaded from a
URL, e.g.:
LOAD <http://xmlns.com/foaf/spec/20100809.rdf>

• The graph can also be loaded as a named graph, e.g.:
LOAD
<http://dl.dropbox.com/u/3954923/ontotext.rdf.ttl>
INTO <http://ontotext.com/ns#>

• A named graph can also be cleared of all triples, e.g.:
CLEAR GRAPH <http://ontotext.com/ns#>
SPARQL 1.1 Update

#9

Graph Management 1/2
• A new named graph can be explicitly created, e.g.:
CREATE GRAPH <http://ontotext.com/ns#>

• A new named graph can be dropped (different from
clearing if empty graphs are supported), e.g.:
DROP GRAPH <http://ontotext.com/ns#>

• We can also CLEAR, CREATE and DROP:
DEFAULT and ALL
• We also LOAD, CLEAR, CREATE and DROP with SILENT
SPARQL 1.1 Update

#10

Graph Management 2/2
• The recommendation for SPARQL Update may or
may not also include:
– COPY… TO …, e.g.:
COPY GRAPH <http://www.example.com/v1#>
TO GRAPH <http://www.example.com/v2#>

– MOVE… TO …, e.g.:
MOVE GRAPH <http://www.example.com/draft#>
TO GRAPH <http://www.example.com/v1#>

– ADD… TO …, e.g.:
ADD GRAPH <http://ontotext.com/ns#>
TO GRAPH <http://sirma.bg/ns#>
SPARQL 1.1 Update

#11

�ݺ�ߣ

ESWC SS 2012 - Monday Tutorial 2 Barry Norton: SPARQL 1.1 Update Language

More Related Content

ESWC SS 2012 - Monday Tutorial 2 Barry Norton: SPARQL 1.1 Update Language