TestSuiteCoverage

From SPARQL Working Group
Revision as of 15:47, 24 January 2012 by Lfeigenb (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

This page records the status of the tests for each area.

If you find anything missing, please add it here so it does not get overlooked.

Query

  • Need tests for:
    • IF (gtw: added two tests / 2011.08.13)
    • COALESCE (currently used in group04, but not thorough) (gtw: added one test / 2011.08.15)
    • Errors in aggregate evaluation
    • Hash functions on Unicode literals (gtw: added tests / 2011.08.13)
    • Positive EXISTS tests (gtw: added two tests / 2012.01.17)

Implementations

The following implementers plan to support this:

  • WG members:
    • Andy: full
    • Chimezie: will not support all features of Query (which?)
    • Greg: full
    • Lee: part of 1.1 query, will implement the rest eventually
    • Olivier: full
    • Paul: query by first quarter next year
    • Steve: ? (all features?)
    • Birte: ? (all features?)
    • Carlos: ?
    • Oracle: ?
  • Others:
    • Sesame (Jeen): ?
    • OpenLink (Orri,IvanM): ?

Update

Covered:

  • INSERT DATA, DELETE DATA, DELETE-INSERT
  • DROP
  • CLEAR
  • ADD, MOVE, COPY (also missing in the update syntax tests, cf http://www.w3.org/2009/sparql/meeting/2011-04-05), need all variants with and without GRAPH keyword and with DEFAULT as source/destination, also syntax tests for SILENT variants.

Not covered:

  • LOAD
  • CREATE?
  • Do we test SILENT wherever it is allowed? (i.e., LOAD, LOAD INTO, CLEAR, DROP, CLEAR/DROP NAMED/ALL/DEFAULT, CREATE, DROP, COPY, MOVE, ADD) in the syntax tests, at least LOAD seems to be missing. - done (ACTION-517)
  • Do we negative syntax test cases for Turtle bnode shortcuts in DELETE DATA? (normal DELETE is already covered)
  • time-allowed, we should move all negativeSyntaxtests to the syntax-tests subfolder of the test suite (currently spread over several folders)

Implementations

The following implementers plan to support this:

  • WG members:
    • Andy: full
    • Chimezie: ?
    • Greg: full
    • Lee: full ()
    • Olivier: full
    • Paul: update by first quarter next year
    • Steve: ? (all features?)
    • Birte: ?
    • Carlos: ?
    • Oracle: ?
  • Others:
    • Sesame (Jeen): ?
    • OpenLink (Orri,IvanM): ?

Protocol

No tests yet.

Implementations

The following implementers plan to support this:

  • WG members:
    • Andy: full
    • Chimezie: ?
    • Greg: full
    • Lee: full
    • Olivier: don't implement protocol
    • Paul: ?
    • Steve: ?
    • Birte: ?
    • Carlos: ?
    • Oracle: ?
  • Others:
    • Sesame (Jeen): ?
    • OpenLink (Orri,IvanM): ?

Service Description

Implementations

The following implementers plan to support this:

  • WG members:
    • Andy: full
    • Chimezie: ?
    • Greg: full
    • Lee: full
    • Olivier: ?
    • Paul: ?
    • Steve: ?
    • Birte: ?
    • Carlos: ?
    • Oracle: ?
  • Others:
    • Sesame (Jeen): ?
    • OpenLink (Orri,IvanM): ?

Graph Store HTTP Protocol

No tests yet.

Implementations

The following implementers plan to support this:

  • WG members:
    • Andy: full
    • Chimezie: ?
    • Greg: full, eventually (KjetilK was working on this, but may not finish for LC testing)
    • Lee: full
    • Olivier: ?
    • Paul: ?
    • Steve: ?
    • Birte: ?
    • Carlos: ?
    • Oracle: ?
  • Others:
    • Sesame (Jeen): ?
    • OpenLink (Orri,IvanM): ?

Entailment

Covered:

Not covered:

  • RDF:
    • We have a test for rdf:_n occuring in the data (:rdf02), but should also have the other way around, ie. asking for an _:n not appearing in the data as a negative test
    • It seems we have no tests checking axiomatic triples in the query pattern, do we? Something like the following query against the empty dataset (Likewise, for other entailment regimes and their sets of axiomatic triples.
  ASK { rdf:type rdf:type rdf:Property .
        rdf:subject rdf:type rdf:Property .
        rdf:predicate rdf:type rdf:Property .
        rdf:object rdf:type rdf:Property .
        rdf:first rdf:type rdf:Property .
        rdf:rest rdf:type rdf:Property .
        rdf:value rdf:type rdf:Property .
        rdf:_1 rdf:type rdf:Property .
        rdf:_2 rdf:type rdf:Property .
        rdf:nil rdf:type rdf:List . }
  • D-Ent seems to miss some standard datatype map test case, such as
  data:  :s :p: "1"^^xsd:integer .
  query: ASK {:s :p: "1.0"^^xsd:decimal .}
 or alike.
  • OWL Direct-Semantics: ???
  • OWL RDF-based semantics seems not covered.
  • RIF seems not covered at all (we could probably look into RIF test cases)

Implementations

The following implementers plan to support this:

  • WG members:
    • Andy: full?
    • Chimezie: full
    • Greg: no
    • Lee: full?
    • Olivier: full?
    • Paul: ?
    • Steve: ?
    • Birte: full?
    • Carlos: ?
    • Oracle: ?
  • Others:
    • Sesame (Jeen): ?
    • OpenLink (Orri,IvanM): ?

Federated Query

  • Remove or mark as optional (via mf:requires) :service5 (testing variable-endpoint service patterns)

Covered:

  • SILENT, NESTED SERVICE, SERVICE VAR, BININDGS
  • JOIN of SERVICE and triples, SERVICE and SERVICE, triples and OPTIONAL SERVICE

Not covered:

  • FILTERS with SERVICE, more than 3 SERVICEs

Implementations

The following implementers plan to support this:

  • WG members:
    • Andy: full
    • Chimezie: ?
    • Greg: full
    • Lee: full?
    • Olivier: no
    • Paul: ?
    • Steve: ?
    • Birte: full?
    • Carlos: full?
    • Oracle: ?
  • Others:
    • Sesame (Jeen): ?
    • OpenLink (Orri,IvanM): ?

JSON Results

Test cases cover:

  • different types of literals (plain, string, other datatypes, blank nodes, URI-refs)
  • unbound variables
  • ASK queries, both true and false as answers.

Test cases don't cover:

  •  ???

Test cases:

Implementations

The following implementers plan to support this:

  • WG members:
    • Andy: full
    • Chimezie: ?
    • Greg: full
    • Lee: ?
    • Olivier: full
    • Paul: ?
    • Steve: ?
    • Birte: ?
    • Carlos: ?
    • Oracle: ?
  • Others:
    • Sesame (Jeen): ?
    • OpenLink (Orri,IvanM): ?

TSV/CSV Results

Test cases cover:

  • different types of literals (plain, string, other datatypes, blank nodes, URI-refs)
  • unbound variables

Test cases don't cover:

  •  ???

Test cases:

Implementations

The following implementers plan to support this:

  • WG members:
    • Andy: full
    • Chimezie: ?
    • Greg: full
    • Lee: part of 1.1 query, will implement the rest eventually
    • Olivier: full?
    • Paul: ?
    • Steve: ?
    • Birte: ?
    • Carlos: ?
    • Oracle: ?
  • Others:
    • Sesame (Jeen): ?
    • OpenLink (Orri,IvanM): ?