>>>>> "r" == rogers-pm5 <[email protected]> writes:
r> I have used languages (PL/1 and Pascal come to mind) that allow you to r> declare both upper and lower bounds to each array dimension. But for r> both languages, I have found such arrays more trouble than they are r> worth. A Pascal example I saw created a histogram of something grouped r> by year (declared as "array [1970..1984] of integer", or some such). r> But since the bounds must be fixed at compile time, you cannot write a r> practical *general* histogram generator this way. r> Of course, a large part of the problem with these languages is r> compile-time binding of array dimensions, so maybe there are better use r> cases. But note that even Common Lisp, which supports multidimensional r> arrays with dynamic upper bounds, still fixes all lower bounds to 0. in one view it is is a pain but for speed, compile time array knowledge is key. what pl1 and fortran can do with array access optmization is amazing and blazing. most decent compilers will even factor out of loops common deep access code. they have decades of experience in squeezing out the most from array handling. perl5 can't ever do that (perl 6 will have some abilities in this area). uri -- Uri Guttman ------ [email protected] -------- http://www.sysarch.com -- ----- Perl Code Review , Architecture, Development, Training, Support ------ --------- Gourmet Hot Cocoa Mix ---- http://bestfriendscocoa.com --------- _______________________________________________ Boston-pm mailing list [email protected] http://mail.pm.org/mailman/listinfo/boston-pm
