Description:
------------
This is related to bug 32211. The fix specified a documentation fix was required. It seems that this should be fixed in the code base, so that hexadecimal and decimal integers behave identically. Regardless, the documentation doesn't completely describe the behaviour below.

Description:

The manual (http://ie2.php.net/int#language.types.integer.overflow) implies that hexadecimal integers above 2147483647 do not get converted to a float, but rather, are truncated to an int.

In actual fact, only numbers above 0x100000000 get truncated. Hexadecimal numbers between 0x80000000 and 0xffffffff are converted to floats, in the same way as if they had been specified in decimal.



Reproduce code:
---------------
        echo "\nHexdecimal ints near ULONG_MAX(truncated to int(LONG_MAX) after ULONG_MAX)\n";
        echo "0xfffffffd\t= "; var_dump(0xfffffffd);
        echo "0xfffffffe\t= "; var_dump(0xfffffffe);
        echo "0xffffffff\t= "; var_dump(0xffffffff);
        echo "0x100000000\t= "; var_dump(0x100000000);
        echo "0x100000001\t= "; var_dump(0x100000001);
        echo "0x100000002\t= "; var_dump(0x100000002);

        echo "\nDecimal ints near ULONG_MAX(floats as far as the eye can see)\n";
        echo "4294967293\t= "; var_dump(4294967293);
        echo "4294967294\t= "; var_dump(4294967294);
        echo "4294967295\t= "; var_dump(4294967295);
        echo "4294967296\t= "; var_dump(4294967296);
        echo "4294967297\t= "; var_dump(4294967297);
        echo "4294967298\t= "; var_dump(4294967298);


Expected result:
----------------
Hexdecimal ints near ULONG_MAX(truncated to int(LONG_MAX) after ULONG_MAX)
0xfffffffd      = float(4294967293)
0xfffffffe      = float(4294967294)
0xffffffff      = float(4294967295)
0x100000000     = float(4294967296)
0x100000001     = float(4294967297)
0x100000002     = float(4294967298)

Decimal ints near ULONG_MAX(floats as far as the eye can see)
4294967293      = float(4294967293)
4294967294      = float(4294967294)
4294967295      = float(4294967295)
4294967296      = float(4294967296)
4294967297      = float(4294967297)
4294967298      = float(4294967298)

Actual result:
--------------
Hexdecimal ints near ULONG_MAX(truncated to int(LONG_MAX) after ULONG_MAX)
0xfffffffd      = float(4294967293)
0xfffffffe      = float(4294967294)
0xffffffff      = float(4294967295)
0x100000000     = int(2147483647)
0x100000001     = int(2147483647)
0x100000002     = int(2147483647)

Decimal ints near ULONG_MAX(floats as far as the eye can see)
4294967293      = float(4294967293)
4294967294      = float(4294967294)
4294967295      = float(4294967295)
4294967296      = float(4294967296)
4294967297      = float(4294967297)
4294967298      = float(4294967298)