Re: [libxml++] patch to allow building without glibmm

[Stefan Seefeld <seefeld sympatico ca>]

Vadim Zeitlin wrote:
> On Fri, 28 Oct 2005 16:16:08 +0200 (CEST) Murray Cumming <murrayc murrayc com> wrote:
>
> MC> It matters in the application if the document contains UTF-8 and you try
> MC> to do the_string.substr(2) or the_string[3] or convert to upper or lower
> MC> case in your document, or sort the strings.
>
>  Of course, I realize this. But libxml++ doesn't do anything of this
> (again, as far as I could see) in its code, so why does it matter to it?
> Any application which needs to do any of the tasks above will need
> Glib::ustring or something else (no, Glib is not the only possibility
> here). But there is no intrinsic dependency of libxml++ on ustring AFAICS,
> i.e. libxml++ just needs a buffer of chars and nothing more.
>
> MC> I don't believe that anybody can guarantee that their system will never
> MC> need to support UTF-8.
>
>  This is a separate question and I don't want to argue about it. What I do
> want to argue about is that any application which needs to support UTF-8
> could still do it perfectly well even if libxml++ used std::string.

FWIW, I have been proposing my own C++ wrapper around libxml2 as a new API
for boost (http://www.boost.org), and the discussion is mainly focussed on
how to interface between this XML API and an (external) unicode library.
You can follow the discussion here:

http://lists.boost.org/Archives/boost/2005/10/96129.php

May be you can find some inspiration for libxml++ there.

Regards,
		Stefan