Re: g_utf8_offset_to_pointer() optimization

[David Malcolm <dmalcolm redhat com>]

On Thu, 2005-11-03 at 01:59 +0100, Luis Menina wrote:

[snip]

It would be great if this test could live in CVS somewhere.

If so, would it help if you do something like:
#define TEST_FUNCTION  g_utf8_offset_to_pointer2

and change that in one place to avoid having to keep updating the
function call in both places? (I'm paranoid that the validity test will
get out-of-sync with the timing test)

You could even write a Makefile that builds the program 5 different
times, passing -D TEST_FUNCTION with 5 different values each time, and
runs the tests.  Dunno if that's useful...

Dave