Re: [xmlroff] xmlroff's "story"?

[Stefan S. <se...@sy...>]

Tony Graham wrote:
> Stefan Seefeld <se...@sy...> writes:
> 
>>Mauro C. wrote:
>>
>>>--- Tony Graham <Ton...@Su...> ha scritto:
>>>
>>>>I started the "What needs to be done to make xmlroff
>>>>a success?"
>>>
>>>IMO a wrapper exetension for PHP and other scripting
>>>language.
>>
>>What would that wrapper do that can't be done by calling
>>system('xmlroff ...') ?
> 
> 
> If the scripting language's XML DOM implementation is a thin layer over
> libxml2's XMLdoc, then it becomes possible to pass a DOM straight to the
> thinly wrapped libfo.

I understand that. What I fail to see is that there is indeed a need for
this level of optimization (or whatever else a programmatic interface to
libfo provides), at this time.
I don't doubt that this is nice-to-have. What I'm not sure about is whether
it really deserves to be on a wishlist for xmlroff when other items such
as feature-completeness should figure with far more scores in terms of
priority.
The user of such an API still needs to preprocess his dom tree (possibly by
means of the libxslt API by Daniel Veillard) to apply the libfo-compat.xsl
stylesheet, right ?


Regards,
		Stefan