See also: IRC log
<LeeF> ACTION: ericP to eyeball the list of uncovered facets and prune ones that are covered [DONE] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action01]
<LeeF> ACTION: LeeF mark dataset and graphs tests approved [DONE] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action02]
<LeeF> ACTION: LeeF mark construct tests approved [DONE] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action03]
<LeeF> ACTION: LeeF mark i18n tests as approved [DONE] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action04]
<LeeF> ACTION: LeeF to propose a grouping of facets into higher level features [DONE] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action05]
<LeeF> ACTION: LeeF to write a ?o/$o test [DONE] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action06]
<LeeF> ACTION: AndyS to write a BASE test [DONE] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action07]
<LeeF> ACTION: chimezie to eyeball the algebraic forms and find any that do not have corresponding tests, attempt to do it by Aug 7 [DONE] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action08]
<LeeF> ACTION: ericP to run ASK tests 1, 4, and 7 by Aug 7 [DONE] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action09]
<LeeF> Scribe: chimezie
<AndyS> AndyS sends regrets for Aug 14 meeting
Chimezie sends regrets for Aug 14 as well
<LeeF> minutes from 31 July
<LeeF> approved, chimezie seconding
LeeF: we need canonical URIs for test reporting
<LeeF> .../manifest.ttl#test-name
<LeeF> .../tests/r2#test-name
<LeeF> http://www.w3.org/2001/sw/DataAccess/tests/r2#construct-4
<LeeF> http://www.w3.org/2001/sw/DataAccess/tests/data-r2/construct/manifest.ttl#construct-4
<LeeF> http://www.w3.org/2001/sw/DataAccess/tests/data-r2/construct/manifest#construct-4
ericP: It would be akward not to provide RDF/XML formats for manifest files
LeeF: should this apply to all our files in turtle format?
ericP: yes
<LeeF> qt:data <foo.ttl>
LeeF: that would be a large refactoring effort
<AndyS> Transform is <#testfrag> to base:test.
AndyS: we only need to change the name of the test not the query or the data
AndyS: what does it dereference to?
ericP: suggestion: keep ttl files. Have a copy of manifest.rdf served from manifest# with fragments for test names
<AndyS> data format (Turtle)
<LeeF> ericP: would like to have manifest.ttl point to foo.ttl and manifest.rdf point to foo.rdf
AndyS: precedent for not going down turtle route?
I think the DAWG is in a better position to use turtle since there is a close relationship between SPARQL and Turtle
<LeeF> Data examples in the spec are all in Turtle
scribe: conversation with Phillip (sp?) ..
<LeeF> I also think that maybe it's worthwhile to note that we're not trying to pub;lish our test suite as a /TR/ path doc
<LeeF> (unlike RDF, OWL, GRDDL, ...)
<AndyS> Could use N-TRIPLES which is what RDF-core developed so is more W3C, if still sub-REC.
<LeeF> http://www.w3.org/2001/sw/DataAccess/tests/data-r2/construct/manifest#construct-4
LeeF: we are in agreement that we prefer the latter test URI convention
<LeeF> <basic/manifest.ttl>
<LeeF> <#spoo-1>
<LeeF> basic/manifest.ttl#spoo-1
<LeeF> basic/manifest
AndyS: super manifest can name manifest.ttl and use consistent test URI convention throughout
<LeeF> @prefix base: <manifest#>
<LeeF> base:spoo-1
<AndyS> Will break running from CVS (file:)
<scribe> ... ongoing conversation of modifying manifests to comply with test name convention ...
AndyS: could we convert manifests to RDF/XML as a final process
<LeeF> AndyS: turn <#...> into prefix names
<ericP> cat /home/eric/WWW/2001/sw/DataAccess/tests/data-r2/distinct/manifest.ttl | perl -pe "s/.ttl/.rdf/g" | ./algae -i turtle -d - --sClass rdfxml "SELECT * { ?s ?p ?o }"
<LeeF> ACTION: LeeF to update the EARL document and all manifest files to use manifest#testname URIs [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action10]
<LeeF> LeeF: Schedule worries if we try to provide parallel test suite in RDF/XML world
<LeeF> ACTION: LeeF to ask CG about RDF/XML version of test suite [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action11]
LeeF: it can still be useful to the community to ask for implementors to use the test framework (as it is)
<LeeF> missing algebraic forms
<LeeF> ACTION: chimezie to write tests to cover missing algebraic forms as per http://lists.w3.org/Archives/Public/public-rdf-dawg/2007JulSep/0056.html [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action12]
<LeeF> scribenick: ericP
<LeeF> eric's ASK results
<LeeF> PROPOSE: Approve tests ask-1, ask-4, ask-7 in http://www.w3.org/2001/sw/DataAccess/tests/data-r2/ask/
<LeeF> resolved, AndyS seconding
<LeeF> basic tests
<LeeF> PROPOSE: Approve all the tests in http://www.w3.org/2001/sw/DataAccess/tests/data-r2/basic/
<LeeF> resolved, SteveH seconding
<LeeF> #distinct-start-1
<LeeF> #distinct-star-1
<LeeF> PROPOSE: Approve http://www.w3.org/2001/sw/DataAccess/tests/data-r2/distinct/manifest#distinct-star-1
<LeeF> resolved, ericP seconding
<LeeF> ACTION: LeeF to mark all tests approved in the 2007-08-07 meeting as approved [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action13]
<LeeF> updated coverage
LeeF: down to 16 facets not
covered
... ericP, did you find anything in the shadows?
ericP: nope, just got to ASK last night
LeeF: anyone who's concearned about coverage should propose test areas by next week
ericP: second
AndyS: always want more tests, but more important to give the implementors more time
Souri: by 14th? sounds good
SteveH: hope to compare DAWG
tests to our internal tests by tomorrow
... will notifiy list of something comes up
<LeeF> JJC on invalid numeric literals
LeeF: boolean value of ill-typed boolean is TRUE. prolly not intentional
<AndyS> Example FILTER ("foo"^^:myNumericType < 5 )
LeeF: seems like a bug that is fixable without another LC
-> http://www.w3.org/TR/rdf-sparql-query/#OperatorMapping operator mapping
<AndyS> "Xyz"^^xsd:integer is not a number.
<LeeF> """
<LeeF> If the argument is a numeric type or a typed literal with a datatype derived from a numeric type, the EBV is false if the operand value is NaN or is numerically equal to zero; otherwise the EBV is true.
<LeeF> """
http://www.w3.org/TR/xquery/#id-ebv
<LeeF> """
<LeeF> #
<LeeF> If its operand is a singleton value of any numeric type or derived from a numeric type, fn:boolean returns false if the operand value is NaN or is numerically equal to zero; otherwise it returns true.
<LeeF> """
<age xsi:datatype="xsd:integer">seven</age>
<LeeF> fn:boolean(//age)
(!"yep"^^my:boolean)
type error
[[If the argument is a numeric type or a typed literal with a datatype derived from a numeric type, the EBV is false if the operand value is NaN or is numerically equal to zero; otherwise the EBV is true.]]
<LeeF> ACTION: ericP to incorporate text to fix EBV of illformed numeric literals bug in spec [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action14]
<LeeF> ACTION: ericP to poke IETF folks about registering SPARQL media types (esp. application/sparql-query) [CONTINUES] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action15]
<LeeF> ACTION: ericP to write explanatory text saying that, like xsd:inteter and xsd:dateTime, the relative order of simple literals and xsd:strings is not defined here [CONTINUES] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action16]
<LeeF> ACTION: ericP to answer "EBV of invalid numeric literals" email from Arjohn [CONTINUES] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action17]
<LeeF> ACTION: ericP to requisition /TR/SPARQL as a switcher document pointing to the real specs [CONTINUES] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action18]
<LeeF> ACTION: EliasT to come up with feature list for protocol testing [CONTINUES] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action19]
<LeeF> ACTION: LeeF and ericP to write tests for uncovered facets [CONTINUES] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action20]
<LeeF> ACTION: ericP to write a test showing that langMatches doens't do extended matching [CONTINUES] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action21]
<LeeF> ACTION: AndyS or LeeF to mark non-SELECT tests using :QueryForm classes, and to move those URIs to the qt: namespace [CONTINUES] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action22]
<LeeF> ACTION: Eric+Andy to identify and collect mf:requires URIs and put in documentation [CONTINUES] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action23]
<LeeF> ACTION: ericP, AndyS to add the mf:requires labels to the manifest namespace document [CONTINUES] [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action24]
<Souri> I got to go too, bye
LeeF: what do we do with XML results format?
<LeeF> ACTION: LeeF to get rid of any distinct and ordered attributes [recorded in http://www.w3.org/2007/08/07-dawg-minutes.html#action25]
ericP: Ralph: best to publish XML Results Format as a CR when we tell the world that we are done with the tests
LeeF: how would you feel about a decision to release the test suite while you aren't around?
AndyS: that's fine
<LeeF> adjourned.