Hi David,

Thanks again for your mail.

>This is a different bug, from what I remember.  Unfortunately, the old
Bugzilla database is gone, so the diffs are not available, so I can't
verify it.
I confirm that version 1.10 fixes the problem in either case (i.e
Changing the encoding through the C++ API or Xsl stlysheet)


Cheers
Manuel



 

-----Original Message-----
From: David Bertoni [mailto:[EMAIL PROTECTED] 
Sent: jeudi, 13. novembre 2008 23:39
To: xalan-dev@xml.apache.org
Subject: Re: [Encoding] Xalan-c API alternative to <xsl:output encoding=
"xxx">

Diaz Manuel wrote:
> Thanks David for your answer.
> I will nevertheless close the thread on the Dev's list if you don't 
> mind.
> 
> 
>  Setting the XSLTResultTarget.setEncoding(xmlencoding) solved my 
> problem!
> 
> Now, as I am using the xalan-c 1.9 version, the ommit-xml-declaration 
> in the stylesheet has no effect whenever "xmlencoding" is set to 
> something different than UTF-8.
This was a big debate when the issue came up.  Technically, the
processor should never generate markup that's not well-formed.  If an
entity is not encoded in UTF-8 or UTF-16, then it requires an XML
declaration (or text declaration) with the proper encoding:

http://www.w3.org/TR/REC-xml/#charencoding

As a result, the request to omit the XML declaration was ignored in
older versions of Xalan-C.  However, there were requests to change this
behavior, under the assumption that a user might want to specify the
actual encoding through some other mechanism.  I was opposed to this,
but was out-voted by the Xalan-J team.

> As a result, I get multiple <?xml.. Entries in my output file.
This should never happen.  If you get more than one XML declaration,
it's a bug, and you should file a Jira issue so we can investigate.

> 
> I can see that this has been fixed in the 1.10 version (XALANC-557), 
> but just by curiosity...is there any workoround for this in 1.9?
This is a different bug, from what I remember.  Unfortunately, the old
Bugzilla database is gone, so the diffs are not available, so I can't
verify it.

Dave

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to