RE: [all] Question on mapping best practices

Indeed. You are right Shaun.
That illuminates that solution then.

And that left the other one, which Felix's recommend.
So I guess my question is answered.

Thanks.
-yves

-----Original Message-----
From: Shaun McCance [mailto:shaunm@gnome.org] 
Sent: Friday, October 26, 2012 12:15 PM
To: public-multilingualweb-lt@w3.org
Subject: Re: [all] Question on mapping best practices

On Fri, 2012-10-26 at 11:25 -0600, Yves Savourel wrote:
> Hi all,
> 
> While mapping ITS to XLIFf we ran into cases of mapping that may occur 
> elsewhere and for which an ITS guideline may be helpful.
> 
> Here is the case:
> 
> The localization note has two pieces of information:  a) the text of 
> the note and b) a type (description|alert)
> 
> When mapping an inline note to XLIFF we can use this:
> 
> <mrk mtype='x-itsNote' comment='[text of the note]'
> its:locNoteType='alert'>...</mrk>
> 
> Or this:
> 
> <mrk mtype='x-itsNote' comment='[text of the note]'
> ZZZ:locNoteType='alert'>...</mrk>
> 
> The comment attribute is where XLIFF is expected to put the note, and 
> because there is no equivalent to the note type we use a non-XLIFF 
> attribute there. The question is can/should we use the ITS attribute 
> or another one?
> 
> In both cases if we want to process the file with an ITS processor, we 
> have to use global rules:
> 
> <its:locNoteRule selector="//mrk[@its:locNoteType='alert']"
>  locNotePointer="@comment" locNoteType='alert'/>
> 
> <its:locNoteRule selector="//mrk[@its:locNoteType='description']"
>  locNotePointer="@comment" locNoteType='description'/>
> 
> or
> 
> <its:locNoteRule selector="//mrk[@ZZZ:locNoteType='alert']"
>  locNotePointer="@comment" locNoteType='alert'/>
> 
> <its:locNoteRule selector="//mrk[@ZZZ:locNoteType='description']"
>  locNotePointer="@comment" locNoteType='description'/>
> 
> I think both would work.
> But we're not sure if the best thing to do for the local attribute is 
> to use a native ITS attribute or define a new namespace and use 
> something from there.

Wouldn't the first case not work because overriding is complete?
An ITS processor would see the local its:locNoteType attribute and be forced to use local only for the Localization Note data category. It doesn't matter that you matched and reapplied the local info in the global rule. The local attribute would cause the global rule to have no effect on that node.

Or am I misunderstanding overriding?

--
Shaun

Received on Friday, 26 October 2012 18:27:20 UTC