Mark Wielaard <mjw@xxxxxxxxxx> writes: > On Wed, 2011-01-19 at 10:48 -0500, Stan Cox wrote: >> mysql defines min and max to be macros in my_global.h which conflicts >> with the definition of max and min in numeric_limits in <limits>. >> This use of max and min is described in the WG14 c++ standard under >> 18.3.1 Numeric limits. Perhaps the mysql maintainers wouldn't mind >> renaming theirs to MIN/MAX. I think the odds of that happening are barely distinguishable from zero, and the odds of it happening with a short turnaround time are exactly zero. They're going to say they don't use <limits> and are not going to promise to avoid conflicts with any random header file that might be out there; and besides who's going to promise them that MIN/MAX won't conflict with something else? Nor am I prepared to carry a patch that replaces each of the several hundred uses of those macros in the mysql sources. I think the only near-term fix is to turn off dtrace support in mysql. It wasn't there till last month, and has never been in any shipped Fedora release, so there's no real loss. But the real issue here is that systemtap has got to be more chary of what they expose into application namespace, or this same scenario is going to be played out again and again. regards, tom lane -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel
