W3C home > Mailing lists > Public > public-i18n-its@w3.org > January to March 2006

Re: Directionality Scope/inheritance issue (same as translatability)

From: Felix Sasaki <fsasaki@w3.org>
Date: Sun, 26 Mar 2006 22:10:25 +0900
Message-ID: <442692C1.9050200@w3.org>
To: Yves Savourel <yves@opentag.com>
Cc: public-i18n-its@w3.org
Hi Yves,

Sorry, *I* missed s.t., see below.

Yves Savourel wrote:
> Felix, thanks for taking the time to write down the steps.
> 
> I'm still a bit unclear on how you apply a rule to the children of the its selection node.
> For example, in the case of p[2] you have:
> 
> <p trans="false" its:translate-global-4="no">some text <code its:translate-global-2="no">code <textInCode
> its:translate-global-1="yes">text in code</textInCode>code</code>.</p>
> 
> That is:
> 
> /myDoc/body/p[2]/: its:translate-global-4="no"
> /myDoc/body/p[2]/code: its:translate-global-2="no"
> /myDoc/body/p[2]/code/textInCode: its:translate-global-1="yes"
> 
> Which means (looking at your description for p[1]) that the content of <textInCode> would be translated.
> 
> But it's wrong since the trans='false' in <p> should overwrite it, if it is to behave like its:translate.
> 
> Am I missing something?

what I missed was to say: As a value is inherited, it looses the
information about its origin: it is just an inherited value. Just FYI, I
wrote the source of inheritance in brackets.

/myDoc/body/p[2]/: its:translate-global-4="no" inherited-value="default"
(no information available, hence: "default": elements are translated,
attributes not)

/myDoc/body/p[2]/code: its:translate-global-2="no" inherited-value="no"
(taken from its:translate-global-4="no" at /myDoc/body/p[2]/)

/myDoc/body/p[2]/code/textInCode: its:translate-global-1="yes"
inherited-value="no" (taken from its:translate-global-2="no" at
/myDoc/body/p[2]/code)

Inherited values have less priority than directly assigned values. That
is the reason why its:translate-global-4="no" from /myDoc/body/p[2]/
does not overwrite its:translate-global-1="yes" at
/myDoc/body/p[2]/code/textInCode .

the information about "xxx-global-1" versus "xxx-global-2" versus "..."
only comes into play if they match both the same node *directly*, e.g.
like //*[@trans='true']" (translate-global-3) and "//code"
(translate-global-2).

would that resolve the problem?

I am really wondering what Sebastian thinks about all this, from the
XSLT point of view.

Cheers,

Felix

> 
> Cheers,
> -yves
> 
> 
> 



Received on Sunday, 26 March 2006 13:10:41 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:43:06 UTC