'Re: [xml] "Control over encoding declaration (prolog and meta)'

[Kasimier Buchcik <kbuchcik 4commerce de>]

Hi,

on 1/14/2004 6:27 PM Daniel Veillard wrote:
> On Wed, Jan 14, 2004 at 05:28:55PM +0100, Kasimier Buchcik wrote:
>
> > I'm trying to serialize documents to UTF-16LE, with full control over 
> > the prolog - I need the encoding declaration to differ from the target 
> > encoding. Doing so I used the static function "xmlDocContentDumpOutput" 
> > as a template, the prolog is constructed and "xmlNodeDumpOutput" is 
> > called (with the target encoding) on every child of the document node. 
> > It works all fine except for one case: if the document is a XHTML 
> > document, "xmlNodeDumpOutput" calls "xhtmlNodeDumpOutput", this sets the 
> > meta encoding (with "htmlSetMetaEncoding") to the target encoding. Here 
> > I lose the control over the declared encoding.
>
> [...]
>
> > Are there any other possibilities to gain full control over the encoding 
> > declaration if serializing?
>
>
>   There is far more simple. Redefine the encoding handler for the advertized
> encoding declaration and provide the functions generating the bits you want !
>    http://xmlsoft.org/html/libxml-encoding.html#xmlNewCharEncodingHandler
>    and
>    http://xmlsoft.org/html/libxml-encoding.html#xmlRegisterCharEncodingHandler
> user provided handler will override the default ones. And you won't have
> to make all this mess or change the code.
>
>   http://xmlsoft.org/encoding.html#extend

Would this approach be thread-save? I would expect this procedure to 
temporarily change the encoding handler for e.g. ISO-8859-1 to UTF-16LE, 
but if I'm serializing an other document to ISO-8859-1 at the same time, 
I would get false results. Or is the registration of encoding handlers 
somehow implemented per-thread like?


Thanks,

Kasimier