At static catalog files you don't need the CDATA
you can direct fill in
<catalogue xml:lang="en">
<message key="test">this is <b>bold</b></message>
</catalogue>
then the bold is *bold*at xsp with esql use <esql:get-xml column="text"/>
Is that what you are looking for?
Klaus
[EMAIL PROTECTED] wrote:
We're trying to migrate our existing application to cocoon. This application supports multiple languages so we stored the text in a DB. However the text in the DB can contain html tags like " this is <b>bold</b> " for English and " celui-ci c'est <b>gras</b> " for French. Now we want to use i18n to translate our pages. Therefor we were trying to generate our catalogue with XSP . As such we are looking to generate for instance:
For English
<catalogue xml:lang="en"> <message key="test"><![CDATA[this is <b>bold</b> ]]></message>
</catalogue>
For French
<catalogue xml:lang="fr"> <message key="test"><![CDATA[celui-ci c'est <b>gras</b>]]></message>
</catalogue>
So that's why we need a CDATA section. If I can't generate cdata my xsp will generate a catalogue like this:
<catalogue xml:lang="en"> <message key="test">this is <b>bold</b></message>
</catalogue>
However , struggling with this example ,I experienced another problem with i18n. Suppose I have a static catalogue (not generated with xsp this time) who looks like:
<catalogue xml:lang="en"> <message key="test"><![CDATA[this is <b>bold</b> ]]></message>
</catalogue>
When I know apply a i18n transformation I see that the <b>bold</b> has become <b>bold</b> in the resulting html ... Or to say it in xslt terms : when selecting the actual translation, the i18n transformation applied a <xsl:value-of select="."/> instead of a <xsl:value-of select="." disable-output-escaping="yes"/>. So I'm afraid that in our case the i18n transformer is useless, unless someone has an idee ?
-----Original Message-----
From: Marc Portier [mailto:[EMAIL PROTECTED] Sent: Tuesday, March 30, 2004 12:23
To: [EMAIL PROTECTED]
Subject: Re: Cdata section in xsp
[EMAIL PROTECTED] wrote:
I want to dynamically generate an xml catalogue for I18n using XSP. If my translations contain certain characters like '>' or '<' my catalogue xml file will not be valid unless I put every translation text in an CDATA section.
Joerg's point is that < and > will do as well to insert those in the characterstream...
whatever you put in originally as <![CDATA[ will go through the sax-stream as characters, and not as elementsl
but the serializer will not recreate cdata sections for those since it doesn't now about those any more and more importantly having them or not is just a syntactical-sugar thing
the serializer will however inject the chars by using the correct character-entities
bottom line you don't get cdata but you don't need them, do you?
-marc=
-----Original Message-----
From: Joerg Heinicke [mailto:[EMAIL PROTECTED] Sent: Tuesday, March 30, 2004 11:29
To: [EMAIL PROTECTED]
Subject: Re: Cdata section in xsp
On 30.03.2004 11:00, [EMAIL PROTECTED] wrote:
Hello,
How can I output a cdata section in an XSP ? I need the disable output
escaping feature of xslt.
First, please do not response to mails from other threads when starting new threads as this breaks thread view in thread-aware browsers.
Second, what exactly do you need it for? Every "normal" character in cdata can be added escaped as normal text and you will have the same at the end. And how shall "disable output escaping feature of xslt" stand in relation to this?
Joerg
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
