Re: reconsider non primary p:output on p:compare

It has been decided between

http://www.w3.org/TR/2007/WD-xproc-20070706/
when it was still spellt "p:equal"
and

http://www.w3.org/TR/2007/WD-xproc-20070920/

Here is Norm's answer to a related question, many months later :
http://lists.w3.org/Archives/Public/public-xml-processing-model-wg/2008May/0043.html

Which makes me think that it has been declared non primary to force
the user to explicit that she/he wants the result of the compare and
not one of the document itself

Which makes sense to me, at least

Because there is the option : "fail-if-not-equal"

We have two use case
1) fail-if-not-equal = false and then you're supposed to consume the
result and hence it would be practical for you to have the output
declared primary

2) fail-if-not-equal = true and then you're NOT suppose to consume the
result and hence it helps you detect a bug faster

I reasonnably think that the spec should stay as it is :
* make bugs easier to find even if it needs to user to be more explicit

My two cents

Xmlizer


On Thu, Apr 2, 2009 at 3:30 AM, Norman Walsh <ndw@nwalsh.com> wrote:
> James Fuller <james.fuller.2007@gmail.com> writes:
>> wondering if p:compare p:output should really be non-primary ... from
>> a DWIM pov I think that we will want to 80% of the time use a result
>>
>> thx for considering, Jim Fuller
>
> I can't immediately lay my hands on the minutes where we made this
> decision. In retrospect it looks like a pretty arbitrary choice to me,
> with my current intuition leaning in Jim's direction.
>
>                                        Be seeing you,
>                                          norm
>
> --
> Norman Walsh <ndw@nwalsh.com> | Through space the universe grasps me
> http://nwalsh.com/            | and swallows me up like a speck;
>                              | through thought I grasp it.-- Pascal
>

Received on Thursday, 2 April 2009 07:48:54 UTC