W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > July to September 2010

Re: Review of SPARQL 1.1 Query (Property paths, negation, select expression, testing values)

From: Andy Seaborne <andy.seaborne@epimorphics.com>
Date: Fri, 24 Sep 2010 17:33:49 +0100
Message-ID: <4C9CD2ED.4030000@epimorphics.com>
To: W3C SPARQL Working Group <public-rdf-dawg@w3.org>


On 23/09/10 14:02, Andy Seaborne wrote:
> Matt - thank you very much for the review and comments.
>
> Split up here to help the editors track comments so we don't loose things.
>
> Andy
>
> -------------------------
>
> Section 8.1:
>
> Question: is NOT EXISTS { ... } equivalent to !EXISTS { ... } ?

Yes, functionally.  FILTER NOT EXISTS {  .. } is legal synatx.  Using ! 
needs ().

> Section 8.2:
>
> In the first paragraph would it be better to use "left-hand side" and
> "right-hand side" instead of "one side" and "other side" ?

Done.

>
> Typos:
> "evaluates both [it's] arguments" => [its]

Done.

>
> Section 9 introduction:
>
> "any given path expression" at the end of the second paragraph should be
> removed.

Done.

>
> "Cycles in the graph and in paths are possible" -- from the definition of
> arbitrary length path, it looks like cycles are not possible in
> aribtrary length paths. Is it only fixed-length paths that may contain
> cycles?

For arbitrary length paths, it's only intermediate nodes that are only 
traversed once.  If it's an end node, that node might also be an 
intermediate node and the matching can go round the cycle.

:a :p :b .
:b :p :b .

and { :a :p+ :b } matches twice.

> Section 9.2:
>
> Typos:
> "Example: Find nodes connected but not by [rdf;type]" => [rdf:type]

Done.

> "Note: This path expression does not [gaurantee] the order of the
> results" => [guarantee]

Done.

>
> Section 9.3.5:
>
> Does the term "intermediate node" used in the definition of
> ArbitraryLengthPath need to be formally defined?
>

[[
An intermediate node is one which
	does not form an endpoint of the path match.
]]
> Typo:
> "intermediate nodes in the graph are [trarversed] once only." =>
> [traversed]

Done.

> Section 9.3.6.2:
>
> Typo:
> "and [deteching] when a graph node" => [detecting]

Done.

>
> Section 15.1.2:
>
> Typo:
> "Variables can [be] also be used in expressions if they are introduced
> [as to the] earlier, syntactically in the same SELECT clause:" => remove
> the identified text

Reworded as in relation to a comment by Greg:
[[
New variables can be also be used in expressions if they are introduced 
earlier,
syntactically, in the same SELECT clause:
]]

>
> Section 16.4.14:
>
> COALESCE(5,?x) returns 2 ... shouldn't this return 5?

Done.

>
> Section 16.4.15:
>
> Typo:
> "interprets it as a effective boolean value, [the]" => [then]

Done.

> Section 16.4.16:
>
> I don't understand this sentence: "Errors in comparisons cause the IN
> expression to raise an error if the RDF term being tested is not found
> to be in the list elsewhere in the list of terms" I understood the
> sentence after looking at the examples but not after reading it. I think
> it would be clearer if "to be in the list" was removed.

Done.

>
> Section 16.4.21:
>
> should "str"^^<http://example/romanNumeral> be
> "iiii"^^<http://example/romanNumeral> ?

Done.

>
> Section 16.4.22:
>
> Should "123"@en be "chat"@en ?

Done.


Edits noted have been committed to CVS rq25 v1.90

	Thanks,
	Andy
Received on Friday, 24 September 2010 16:34:27 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:43 GMT