Bug #28184 min() and max() macro defines in my_global.h conflict with c++ headers

Description:
my_global.h contains:
#if !defined(max)
#define max(a, b)       ((a) > (b) ? (a) : (b))
#define min(a, b)       ((a) < (b) ? (a) : (b))
#endif

which are highly problematic when building with upcoming gcc 4.2.0 because:

x86_64-pld-linux-g++ -DUNDEF_THREADS_HACK -DDEFAULT_MYSQL_HOME="\"/usr\"" -DDATADIR="\"/var/lib/mysql\"" -I. -I.. -I../include -I../include -I../regex     -O2 -fno-strict-aliasing -fwrapv -march=x86-64 -gdwarf-2 -g2  -felide-constructors -fno-rtti -fno-exceptions -fomit-frame-pointer   -fno-implicit-templates -fno-exceptions -fno-rtti  -MT mysqladmin.o -MD -MP -MF .deps/mysqladmin.Tpo -c -o mysqladmin.o mysqladmin.cc
sql_string.cc: In member function 'int String::reserve(uint32, uint32)':
sql_string.cc:665: error: 'max' was not declared in this scope
sql_string.cc: In function 'int stringcmp(const String*, const String*)':
sql_string.cc:751: error: 'min' was not declared in this scope
sql_string.cc: In function 'String* copy_if_not_alloced(String*, String*, uint32)':
sql_string.cc:768: error: 'min' was not declared in this scope

and that's because /usr/include/c++/4.2.0/limits contains:
  template<typename _Tp>
    struct numeric_limits : public __numeric_limits_base
    {
      /** The minimum finite value, or for floating types with
          denormalization, the minimum positive normalized value.  */
      static _Tp min() throw() { return static_cast<_Tp>(0); }
      /** The maximum finite value.  */
      static _Tp max() throw() { return static_cast<_Tp>(0); }

How to repeat:
build mysql 5.1.17 with gcc 4.2.0rc1

Suggested fix:
Change names of macros from min to mysql_min and max to mysql_max.

Workaround patch:
http://cvs.pld-linux.org/cgi-bin/cvsweb/SOURCES/mysql-min_max.patch?rev=1.2
but clean solution is to rename these everywhere in source code.

Thank you for a problem report. Verified just as described with code review.

Replaces min() max() with mysql_min() and mysql_max()

Attachment: mysql-min_max.patch (application/octet-stream, text), 142.99 KiB.

Compile tested. Replaces all (hope that I didn't miss anything in code that's not compiled in by default) occurences of min() and max() with non-conflicting mysql_min() and mysql_max().

Please review and merge since the bug is annoying for anyone using recent compiler (note, gcc 4.2.0 was already released some time ago).

see also http://lists.mysql.com/internals/34680 and references therein

Uppercasing to indicate that these are macros is a good idea. So after applying my patch:

find . -type f -exec sed -i -e 's|mysql_min|MYSQL_MIN|g' -e 's|mysql_max|MYSQL_MAX|g' '{}' ';'

This occurs now on Gentoo Linux, for which a 4.2.0 ebuild has been released recently.

% gcc --version
gcc (GCC) 4.2.0 (Gentoo 4.2.0 p1.4)

Affects the 5.1, 5.1-falcon, and 5.2 bk trees.

Also affects Debian "testing" and "unstable" as both have gcc 4.2.x as default compiler now (probably also some Ubuntu versions?)

We use smaller version mentioned earlier in commeents:
http://cvs.pld-linux.org/cgi-bin/cvsweb.cgi/~checkout~/SOURCES/mysql-min_max.patch?rev=1.6...
Works.

Fixed by

http://mysql.bkbits.net:8080/mysql-5.1/?PAGE=cset&REV=46dbec1c1lAGFuqFZQy58XVe4KSijg

Noted in 5.1.23, 6.0.4 changelogs.

MySQL now can be compiled with gcc 4.2.x. There was a problem 
involving a conflict with the min() and max() macros in my_global.h,
which now have been renamed to mysql_min() and mysql_max().

No, they weren't renamed, see the patch at the link

Removed the "which have been renamed" part from the changelog entry.

Is this still fixed?  I'm running into this issue using mysql-devel-5.5-1.3