This makes sense. My biggest problem was that my function took multiple
parameters so I thought it wasn't really a function that would have been
affected by mapping.
Now I think I understand it.
Steve
On Fri, Apr 2, 2010 at 6:37 PM, Danny Sokolsky <danny.sokol...@marklogic.com
> wrote:
> Right. And the other point I was trying to make is that if the function
> takes a zero-or-one parameter, such as "xs:string?", then it is valid to
> pass in the empty sequence and function mapping will not happen on that
> parameter. Here is an example:
>
> xquery version "1.0-ml";
>
> declare function local:will-map($x as xs:string)
> {
> $x, "function will-map was run"
> };
>
> declare function local:no-map($x as xs:string?)
> {
> $x, "function no-map was run"
> };
>
> "function will-map with empty sequence arg:
> ",
> local:will-map(()),
> "function no-map with empty sequence arg:
> ",
> local:no-map(())
>
> =>
> function will-map with empty sequence arg:
>
> function no-map with empty sequence arg:
>
> function no-map was run
>
> Clear as mud?
> -Danny
>
> -----Original Message-----
> From: general-boun...@developer.marklogic.com [mailto:
> general-boun...@developer.marklogic.com] On Behalf Of Micah Dubinko
> Sent: Friday, April 02, 2010 5:07 PM
> To: General Mark Logic Developer Discussion
> Subject: Re: [MarkLogic Dev General] Unexpected behavior
>
> Close. :-)
>
> On Apr 2, 2010, at 4:15 PM, spig wrote:
>
> > So, let me make sure I understand.
> >
> > When I declare a function with any number of parameters, if one of those
> parameters requires a singleton, and I pass in an empty sequence, the
> function will not likely be called if function mapping is on. This makes
> sense to me now.
>
> With function mapping enabled, if a function takes a singleton argument,
> and you pass in a sequence, your function will be called n times, where n is
> the length of the sequence. This is true even when n is 0.
>
>
> >
> > And, by similar logic, it will not throw an error because the function
> requires a singleton parameter, and I have provided one, albeit the empty
> sequence. And, since the empty sequence is a valid singleton (is that true)
> it will not throw an error because the parameters are valid.
>
> Actually, empty sequence is never a valid instance of a singleton type. In
> cases like these the error fails to throw not because the parameter is
> valid, but because the function itself never gets called.
>
> This is easy to get mixed up. I still do at times. -m
>
>
> _______________________________________________
> General mailing list
> General@developer.marklogic.com
> http://xqzone.com/mailman/listinfo/general
> _______________________________________________
> General mailing list
> General@developer.marklogic.com
> http://xqzone.com/mailman/listinfo/general
>
_______________________________________________
General mailing list
General@developer.marklogic.com
http://xqzone.com/mailman/listinfo/general
