ed@contex.UUCP (Edward Bagdonas) (12/05/89)
In converting several old C programs to C++, I came across two
types of behavior from cfont (ATT 1.2) that I found perplexing.
Consider these in regard to the relationship between C and C++.
1) Const expressions may not contain floats or
expressions that evaluate to floats.
#define MACRO1 (i) ((int)(this * i) / that)
int array[ MACRO1( 1000 ) ];
I do see the merit of such restrictions.
Does this behavior provide an exception to the statement that
"C++ is that it is a superset of C"?
2) Cfront has imposed an internal limit of several thousand,
elments per aggregate initialization.
float table[ 6000 ] = { 0.0001, 0.0002, 0.0003, ... };
Is there a particular need or reason for this limit?
E. Bagdonas