W3C home > Mailing lists > Public > public-multilingualweb-lt@w3.org > February 2013

Re: ACTION-447: Make a batch transformation of the test suite to xliff

From: Felix Sasaki <fsasaki@w3.org>
Date: Fri, 22 Feb 2013 16:40:25 +0100
Message-ID: <51279169.4040704@w3.org>
To: Yves Savourel <ysavourel@enlaso.com>
CC: public-multilingualweb-lt@w3.org
Am 22.02.13 16:01, schrieb Yves Savourel:
> Hi Felix,
>> Why I wrote "nearly": in the below you write about HTML5
>> (and its successor, I assume) - but what about XHTML and the
>> other HTML flavours? I am having here again the Linguaserve workflow
>> in mind. Do we then require (or suggest via SHOULD) Linguaserve to
>> process against the default rule set in the XHTML workflow?
> I would think yes. Since 2.0 is addressing HTML now we should do it for XHTML as well.
> I would assume the same default rules would apply.
>> Also, a detail: the rules would have no influence on precedence and overriding,
>> etc. right? - they are kind of assuming that each HTML element has an
>> "invisible linked or external rules file" - which has the lowest precedence compared
>> to other rules, but a higher one than inheritance? E.g.
> Yes. It corresponds to a non-linked rules file that any ITS processor can use as defined in the section 5.6:
> http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#associating-its-with-existing-markup
> "Global rules can be associated with a given XML document using different means:
> - By using an rules element in the document itself:
>    - with the rules directly inside the document, as shown in Example 24
>    - with a link to an external rules file using the XLink href attribute, as shown in Example 19
> - By associating the rules and the document through a tool-specific mechanism. For example, for a command-line tool: providing the paths of both the XML document to process and its corresponding external rules file."
> it's evaluated first, then all other rules are applied and override it if needed as defined per ITS.
>> <span its-within-text="no"><span> ...</span></span>
>> the inner span would be overriden by the
>> <its:withinTextRule withinText="yes"
>>   selector="//h:abbr | //h:acronym | //h:br | //h:cite | //h:code | //h:dfn
>>   | //h:kbd | //h:q | //h:samp | //h:span | //h:strong | //h:var | //h:b | //h:em
>>   | //h:big | //h:hr | //h:i | //h:small | //h:sub | //h:sup | //h:tt | //h:del
>>   | //h:ins | //h:bdo | //h:img | //h:a | //h:font | //h:center | //h:s | //h:strike
>>   | //h:u | //h:isindex" />
>> Taken from
>> http://www.w3.org/TR/xml-i18n-bp/#relating-its-plus-xhtml
> Mmm... You mean the reverse: The local its-within-text="no" markup would override the global (default) rules, right?

Actually: no. In my example there is a span nexted in the span with 
local its-within-text="no". Global rules have higher precedence than 
inherited values. So I thought that the nested span would be nested="yes".

> This is why knowing the default is important for the author.
> The example of the keywords in <meta> is a good illustration:
> If the default rules say that it's translatable, a translate='no' in <html> will not apply to it because the meta/@content node is already explicitly set to translate by the default rules (so translate='no' is not inherited because it has already been set). One would need to explicitly set translate='no' for //meta/@content[@name='keywords'] to not translate it.

That makes a lot of sense.


Received on Friday, 22 February 2013 15:40:56 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:32:04 UTC