Le 2014-10-20 11:49, Martin.J Thompson a écrit : >> so it would be a compile-time error ? > No, runtime. If you have a variable (say) containing the maximum > value for its range, then incrementing it will cause a runtime error. > Similarly shifting it left as described in this proposal would error > through the same mechanism. but that would only be used on the "arithmetic" operations, namely sla and sra, and I can't remember ever using sla, which is identical to sll. If I want arithmetic accuracy, I use a standard multiply and divide. srl and sra do not overflow so no error needs to be checked. It makes even less sense for and/or/xor/not. Can you explain your logic for the errors ? and would it trap, or return a magic/meta-value ? > Martin yg -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.Received on Mon Oct 20 15:22:25 2014
This archive was generated by hypermail 2.1.8 : Mon Oct 20 2014 - 15:23:10 PDT