RE: [xml] Bug 81360

[Jean Senellart <senellart systran fr>]

my 2 cents on this: the interpretation layer will not be so easy to do and not easily maintainable since the 
error, warning and debug messages are not so structured. if we considere the messages that xmlGenericError 
receive (there is about 1000 different calls to this function):

"xmlC14NVisibleNsStackAdd: cur is null.\n"
"SAX.resolveEntity(%s, %s)\n", publicId, systemId
"Cannot read file %s\n", filename
"xmlNewProp : malloc failed\n"
"xmlBufferAddHead: len < 0\n"
"XInclude: XPointer evaluation failed: #%s\n", fragment
"XInclude: XPointer is not a range: #%s\n", fragment
...

maybe first step would be to add some structure in those messages (let us say for instance... a xml format 
:-)?

does this make sense?

Jean

Murray Cumming Comneon com writes:
> > From: Daniel Veillard [mailto:veillard redhat com] 
> >   I stand firm that current libxml2 and libxslt messages are not
> > intended for end-users as is. You will need an interpretation layer.
> > that layer then can rely on gettext or whatever localization
> > mechanism you have available.
>
> And if that's really possible then maybe someone can do it in a reusable
> way. Sounds like a project to me.
>
> Murray Cumming
> murrayc usa net
> www.murrayc.com 
> _______________________________________________
> xml mailing list, project page  http://xmlsoft.org/
> xml gnome org
> http://mail.gnome.org/mailman/listinfo/xml