You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@lenya.apache.org by qMax <qm...@mediasoft.ru> on 2005/06/10 14:16:27 UTC
Re[2]: Translation catalogues synchronization.
Thursday, June 9, 2005, 11:36:34 AM, Gregor J. Rothfuss wrote:
GJR> qMax wrote:
GJR> very cool! are you aware of
GJR> http://svn.apache.org/viewcvs.cgi/cocoon/trunk/src/resources/dev/i18n/
Nope, I've never digged in cocoon trank.
GJR> maybe your script and those at that location need to be merged and
GJR> better documented.
IIUC those scripts are to maintain one big dictionary with all languages in
it and split it into separate for each language.
In this case there's no need to synchronize several dictionaries, since
main message list is in single file.
It could really simplify maintaining translations.
I'll play with it as i'll have a time.
Would it be ok if to create language-dictionaries from one source at
webapp-build stage ?
--
qMax
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@lenya.apache.org
For additional commands, e-mail: user-help@lenya.apache.org
Re[2]: Translation catalogues synchronization.
Posted by qMax <qm...@mediasoft.ru>.
Friday, June 10, 2005, 9:01:23 PM, Gregor J. Rothfuss wrote:
>> Would it be ok if to create language-dictionaries from one source at
>> webapp-build stage ?
I mean to have single multi-lingual dictionary in sources and split it
using xslt task only when building webapp.
Otherwise, it could be possible to patch i18n transformations so they can use
multilingual dictionary.
But it is rather up to cocoon-dev and there could be some reasons to have separate
dictionaries for every language.
GJR> i think what needs to happen first is to go though
GJR> the codebase, and externalize all strings. instead of
GJR> <i18n:text>Your account has been suspended</i18n:text> it should be
GJR> <i18:text key="ac.warning.accountsuspended"/>
GJR> that way, we can be pretty sure we got all the text out.
The markup2messages.xsl can actually extract plain text as well.
Nevertheless, i'd strongly suggest to have keys, since they make all messages to
be in context and this avoid alot of ambiguities in translations.
However, xsl cannot to extract all keys, because some messages are
implicitely referenced:
e.g. in menu2xhtml.xsl:
<span id="workflow-state"><i18n:text><xsl:value-of select="$workflowstate"/></i18n:text></span>
Constructions in form
<i18n:translate><i18n:text key="..."/><i18:param>...</i18n:param></i18n:translate>
are also problem for xsl.
These cases could only be entered manually.
--
qMax
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@lenya.apache.org
For additional commands, e-mail: user-help@lenya.apache.org
Re: Translation catalogues synchronization.
Posted by "Gregor J. Rothfuss" <gr...@apache.org>.
qMax wrote:
> IIUC those scripts are to maintain one big dictionary with all languages in
> it and split it into separate for each language.
> In this case there's no need to synchronize several dictionaries, since
> main message list is in single file.
i see
> It could really simplify maintaining translations.
>
> I'll play with it as i'll have a time.
>
> Would it be ok if to create language-dictionaries from one source at
> webapp-build stage ?
explain a bit more? i think what needs to happen first is to go though
the codebase, and externalize all strings. instead of
<i18n:text>Your account has been suspended</i18n:text> it should be
<i18:text key="ac.warning.accountsuspended"/>
that way, we can be pretty sure we got all the text out.
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@lenya.apache.org
For additional commands, e-mail: user-help@lenya.apache.org