>From: "Rich, Dave" <Dave_Rich@mentor.com> >A "const" does not affect the RHS of the declaration. Whatever is >allowed for a non-cost variable declaration is allowed for a const. In >fact, a const variable declared inside an automatic block is initialized >every time the block is entered to whatever is the current value of the >RHS expression. It's just that a const variable becomes read-only after >initialization. Dave, The LRM actually does specify different rules for the RHS (initializer) on a const variable. This difference is silly, and serves no useful purpose. If it were restricted to constant expressions, then it would make the initializer value well-defined. But as soon as it allows any variable, including a const variable, it is susceptible to initialization ordering differences. I pointed this out in a Mantis item a while ago, but nothing has been done about it. I assume that various implementors either did not notice this rule, or realized that it was pointless and ignored it. Steven Sharp sharp@cadence.com -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.Received on Tue Sep 23 09:36:24 2008
This archive was generated by hypermail 2.1.8 : Tue Sep 23 2008 - 09:37:04 PDT