Re: Question on domain in XLIFF

Actually there is no issue, you are right - sorry for the noise.

- Felix

Am 18.11.2014 um 12:58 schrieb Yves Savourel <ysavourel@enlaso.com>:

> Hi Felix,
>  
> I’m not sure I understand the issue (if there is one).
> I think the rule you came up with is the way to go.
> There are no other Domain-related data in the rules for the ITS module, so placing it anywhere in that rules file would be fine, no? Maybe I’m missing something.
>  
> Cheers,
> -yves
>  
> From: Felix Sasaki [mailto:fsasaki@w3.org] 
> Sent: Monday, November 17, 2014 10:34 PM
> To: public-i18n-its-ig
> Subject: Question on domain in XLIFF
>  
> Hi all,
>  
> looking at Yves’ comment at
> https://www.w3.org/International/its/wiki/XLIFF_2.0_Mapping#Re-writing_the_ITS_namespace
>  
> „The usage of the namespace urn:oasis:names:tc:xliff:its:2.1 is only due to XLIFF 2.0 validation constraints.
> Note YS: it's also because the ITS namespace needs at time to be completed: when a data category uses the XLIFF markup and is missing some features (we would not be able to use the ITS namespace for this); and when ITS local rules are missing things, like a domain attribute"
>  
> I was wondering about domain local. As Yves writes, this is not available in ITS. So an ITS implementation will not look for local attributes. How should the ITS processor then work with local domain attributes in XLIFF? One answer could be to write a global domain rule like
>  
>  <its:domainRule selector="//xlf:mrk[@type='any‘ and @itsm:domains]" domainPointer="@itsm:domains" /> 
>  
> Using global rules for domain then means that this rule has to come last, so that it has precedence compared to other domain related global rules and in that way acts like a local markup. 
>  
> Thoughts?
>  
> - Felix