[xquery-talk] XQuery - Grammar? Implementations?

[Ronald Bourret]

Hello,

[1] succeeds (correctly) because of the function conversion rules listed 
in section 3.1.5 (Function Calls). These "are used to convert an 
argument value or a return value to its expected type". The rule that 
applies to [1] is that, when a function argument has an atomic type, 
argument values are atomized. Atomization (section 2.4.2.) converts 
sequences of nodes to sequences of atomic values. For [1], the node 
value <X>Y</X> is converted to "Y".

[2] fails (correctly) because $a is declared to be an xs:string and 
<X>Y</X> is not a string. Section 3.8.1 (For and Let Clauses) states the 
following in the 14th paragraph (not counting examples):

    "If the type of a value bound to the variable does not match the 
declared type according to the rules for SequenceType matching, a type 
error is raised [err:XPTY0004]"

[3] succeeds correctly because casting explicitly converts <X>Y</X> to a 
string, which is the type expected by the function. Note that [3] does 
explicitly what [1] does implicitly.

-- Ron

Yoshi Okamoto wrote:
> Dear XQuery subscribers,
> 
> I am sometimes fighting to understand all the details of the XQuery grammar. If you have some time, I would be delighted if you could have a look at the following example: I have defined a function with a single string argument:
> 
>   [0]  declare function local:f($a as xs:string) { $a };
> 
> If I pass on an element as follows:
> 
>   [1]  let $a := <X>Y</X> return local:f($a)
> 
> it is accepted by most XQuery implementations which I tried (I am testing with Saxon, BaseX, MonetDB, Zorba, Qizx, Kawa and MXQuery). However, when I cast the element before passing it on:
> 
>   [2]  let $a as xs:string := <X>Y</X> return local:f($a)
> 
> some implementations (also Saxon, which I regard as reference implementation) return [XPTY0004] as error. A third version with an explicit cast:
> 
>   [3]  let $a := xs:string(<X>Y</X>) return local:f($a)
> 
> is finally accepted by all implementations.
> 
> If possible, can someone tell me what the difference are between the versions, especially [1] and [2]? I tried to find the answer in the specification, but I was not successful.
> 
> Thank you very much for your attention and help!
> 
> Best Regards,
> Yoshi Okamoto
> 
> 
>       
> _______________________________________________
> talk at x-query.com
> http://x-query.com/mailman/listinfo/talk
> 
> 
> ------------------------------------------------------------------------
> 
> 
> No virus found in this incoming message.
> Checked by AVG - http://www.avg.com 
> Version: 8.0.233 / Virus Database: 270.10.16/1926 - Release Date: 1/30/2009 5:31 PM
>