On Thu, Jan 25, 2007 at 02:52:10PM -0600, David Hagood wrote:
> I'd like to suggest that the various xmlHashScan functions be modified so
> that the function pointer passed to them can return a bool indicating
> whether the scan should continue or not.
>
> I have a case where I have to basically reverse a hash scan - given
> something about the data hashed, locate a string that maps to that data.
>
> So what I might have is:
>
> struct something
> {
> int a;
> int b;
> };
>
> and a hash mapping a bunch of strings to different instances of something.
> Now, I want to find a string which maps to a "something" with a .a value
> of 12.
>
> Now, unless I have some other entity besides the hash which lists all the
> possible somethings, the only way to find a "something" with .a=12 would
> be to create a hash scanner function that checks each item in the hash and
> records the first item which has .a=12 - but then I am going to iterate
> over the whole rest of the hash after having found my object.
>
> If the hash scanner function could return TRUE = (continue scanning) or
> FALSE = (stop scanning), then I could stop the search when I've found the
> data.
That would be a good idea, except it break both API compatibility and
ABI compatibility with previous versions, so that's just impossible. And
I'm not sure it's worth adding a new API at this point.
Daniel
--
Red Hat Virtualization group http://redhat.com/virtualization/
Daniel Veillard | virtualization library http://libvirt.org/
[EMAIL PROTECTED] | libxml GNOME XML XSLT toolkit http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/
_______________________________________________
xml mailing list, project page http://xmlsoft.org/
[email protected]
http://mail.gnome.org/mailman/listinfo/xml
