W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > April to June 2012

Re: Editorial remarks on the definition of in-scope (ACTION-618 - part 2)

From: Andy Seaborne <andy.seaborne@epimorphics.com>
Date: Wed, 23 May 2012 09:58:20 +0100
Message-ID: <4FBCA6AC.8010702@epimorphics.com>
To: public-rdf-dawg@w3.org


On 23/05/12 08:32, Polleres, Axel wrote:
> Some more small editorial remarks on the definition table for in-scope variables(http://www.w3.org/2009/sparql/docs/query-1.1/rq25.xml#variableScope)

which are your definitions ... :-)

>
> 1) In the definition table in Section 18.2.1
>
> "
> (expr AS v) for BIND, SELECT and GROUP BY
>         v is in-scope
> "
>
> For clarification, I would prefer to split this into 3 rows in the table as follows:
>
>
> "
> BIND (expr AS v)
>        v is in-scope
>
> SELECT .. (expr AS v)  .. { P }
>        v is in-scope
>
> SELECT .. { P }  GROUP BY (expr AS v)
>        v is in-scope (due to the equivalence with  SELECT .. { P }  { ... BIND (expr AS v) } GROUP BY v) )
> "

Done.

>
> 2) In the definition table in Section 18.2.1
>
> "
> SELECT .. v .. { P }
>        v is in-scope if v is mentioned as a project variable
> "
>
> I think it is sufficient to write
> "
> SELECT .. v .. { P }
>        v is in-scope
> "
>

Done

> 3) In the definition table in Section 18.2.1
> "
> Group { P1 P2 ... }
>      v is in-scope if it is in-scope in one or more of P1, P2, ...
> "
>
> I think it is sufficient (and more consistent with the other definitions) to write
>
> "
> SELECT * { P }
> Group { P1 P2 ... }
>      v is in-scope in one or more of P1, P2, ...
> "

Not done - adding the SELECT stresses the "SELECT *" query level form 
and it's any group however nested (and it may be ASK etc)

How about just

{ P1 P2 ... }

or leave as-is.

(the "group" was added to emphasis which use of {} we are talking about 
but it is valid without and it works in the same in OPTIONAL{} so 
reading it as an exclusive or inclusive list gets the same meaning)

> 4) I'd suggest to add links to #variableScope for *all* appearances of the word "in-scope", for instance in section
>     18.2.5.1 ORDER BY
>
> s/
> where vars is the set of variables mentioned in the SELECT  clause or all named variables that are in-scope in the query if SELECT * used.
> /
> where vars is the set of variables mentioned in the SELECT clause or all named variables that are<a href="#variableScope">in-scope</a>  in the query if SELECT * used.
> /
>
> Similarly,
>   in section 18.2.5.2 Projection
> and
>   in section 10 Assignment	
>

Done where it makes sense.  There are only a couple of places.

Uses in the section #variableScope are not linked to the section because 
it that looks strange to me.

One use is a different "in-scope" is about FILTERs - made it "in scope".

>
> 5) Finally, in Section 18.2.1 I suggest to put the word "in-scope" in italic font, where it is defined, i.e.
>
> s/
> We define a variable to be in-scope if there is a way [...]
> /
> We define a variable to be<i>in-scope</i>  if there is a way [...]

Done.

>
>
>
> Best,
> Axel
>
Received on Wednesday, 23 May 2012 08:59:16 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:48 GMT