RE: [xml] UTF8Toisolat1() usage

["Henke, Markus" <Markus_Henke ordat com>]

> -----Original Message-----
> From: Geert Kloosterman [mailto:geertk ai rug nl]
> Sent: Tuesday, June 04, 2002 7:37 PM
> To: Henke, Markus
> Subject: RE: [xml] UTF8Toisolat1() usage
> 
> 
> >>>>> "Markus" == Henke, Markus <Markus_Henke ordat com> writes:
> 
>     > Hi Geert, all,
>     >> -----Original Message-----
>     >> From: Geert Kloosterman [mailto:geertk ai rug nl]
>     >> Sent: Monday, June 03, 2002 4:09 PM
>     >> To: xml gnome org
>     >> Subject: [xml] UTF8Toisolat1() usage
>     >> 
>     >> 
>     >> Hi all,
>     >> 
>     >> How can I convert the result from xmlNodeListGetString() to
>     >> iso-8859-1?  
>     >> 
>     >> I think I have to use UTF8Toisolat1(), but it isn't 
> exactly clear to
>     >> me how to use the function with respect to memory 
> allocation.  Do I
>     >> need to supply a buffer that's large enough myself, or does the
>     >> function allocate it for me?
>     >> 
>     >> Geert
>     >> 
> 
>     > It's been a while that i've looked at the corresponding code,
>     > but i'm quite sure that "UTF8Toisolat1()" won't handle 
> the buffer
>     > size for you.
>     > You can use "xmlCharEncOutFunc()" (get the encoding 
> handler using
>     > "xmlGetCharEncodingHandler()") to convert from UTF-8 to 
> ISO-8859-1,
>     > it uses xmlBuffer for in/out parameters and handles the 
> buffer size.
> 
> Thanks.  So the idiom to convert an UTF8 xmlChar * to an iso-8859-1
> char * is:
> 
>     - setup an encoding handler with xmlGetCharEncodingHandler()
>     - create input and output buffers with xmlBufferCreate()
>     - transfer the UTF8 xmlChar * to an xmlBufferPtr with 
> xmlBufferCat()
>     - translate to iso-8859-1 using xmlCharEncOutFunc()
>     - put the iso-8859-1 results in an xmlChar * using 
> xmlBufferContent()
>     (- cast the xmlChar * to char * ????)
> 
>     --> for iso-8859-1 output use xmlBufferDump() on the buffer
> 
> Am I correct?

Well, seems OK for me if you want to avoid any buffer allocation
by your own (and therefore it's more complex).
The last step (cast to char*) of course depends on what you intend
to do with the buffer...  :)
It's also possible to use UTF8ToIsolat1() if you could provide an
adequate allocated out-buffer. An (out-) buffer that's convertet
from UTF-8 to ISO-8859-1 will never need more than twice the space
as the corresponding in-buffer.


> Geert

Ciao, Markus