Re: [xml] enhancing the error handling API

[Daniel Veillard <veillard redhat com>]

On Tue, Mar 18, 2003 at 12:11:48PM -0500, Stefan Seefeld wrote:
> Daniel Veillard wrote:
>
> > > No thread specific variables! Only global ones, this is the point.
> >
> >
> >   No no, I stand firm on this. The error "global" variables are already
> > handled in thered specific ways if compiled with --with-threads , I don't
> > expect to change that for the new infrastrcuture.
>
> I think what is important to note is that parsing is an incremental
> process,

  right 

> so you need some state where you can collect data to be able
> to determine what the logical error is.

  right it's called a parser context !

> That's why you need thread
> specific storage even on the level of libxml2.

  I don't understand how you draw your conclusion, errors are associated
to parsing context when parsing, and those can be thread specific or not
is simply doesn't matter !

Daniel

-- 
Daniel Veillard      | Red Hat Network https://rhn.redhat.com/
veillard redhat com  | libxml GNOME XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/