On Fri, Dec 14, 2007 at 08:22:24AM -0500, Daniel Veillard wrote:
On Fri, Dec 14, 2007 at 11:06:36AM +0000, Richard W.M. Jones wrote:
> OK, how about this?
>
> Rich.
>
> + for (p = str; *p; ++p) {
> + /* Want to escape only A-Z and 0-9. This may not work on
> EBCDIC. */
> + if (isascii (*p) && isalnum (*p))
> + grow_size++;
> + else
> + grow_size += 3; /* %ab */
> + }
Huh ? How much time and context informations are needed to understand
what that code does. Compare to the time needed for the explicit ranges.
I really don't see any progress there, and still object to the EBCDIC
argument, on such machine you usually don't even get c99 like compilers.
This makes no sense to me, really !
Add /* Not EBCDIC safe */ comment on top of the range if you really
believe someone is gonna cut and paste libvirt escaping code into a
different context, but please keep the semantic of the code clear, it's
#1 IMHO for maintainance.
In a nutshell, I fine with the original version of that patch,
Daniel
--
Red Hat Virtualization group
http://redhat.com/virtualization/
Daniel Veillard | virtualization library
http://libvirt.org/
veillard(a)redhat.com | libxml GNOME XML XSLT toolkit
http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine
http://rpmfind.net/