I'm wondering if it would be useful to provide such a function in apr, because:
- nearly all occurences of str(case)cmp are just interested in equality, but take the extra overhead for the -1/+1 comparison of strcmp - strcasecmp is locale dependent, even if you just want to use ascii. For example, in the turkish locale i and I are not the case compagnions of each other (there's a httpd bug report somewhere regarding to this). Deadly if you're trying to compare something with an I in it. So my proposal would be to write a str_eq and str_case_eq function, which just checks for equality, case translation is limited to us-ascii alphabet (via a simple table). What do you think? Would that fit into the apr concepts? where? nd
