Re: Wondering about compile time error checking

[Marcus G. Daniels]

>>>>> "B" == Benedikt Stefansson <address@hidden> writes:

B> I believe that the Java implementation would also allow this error
B> to go unnoticed until at runtime. 

One advantage of a model compiler on top of Swarm (say, based on XML),
would be that we develop techniques for higher-level validation.  The
notion of a message disassociated from a type is foreign to Java --
thus the Selector class.  And Objective C folks *like* doing things at
runtime.  

It ought to be possible to enhance GCC to report M(undeclaredMethod)
in the same way that [obj undeclaredMethod] would be reported, but
that's a fair amount of work for a fairly small payoff, and it
wouldn't generalize to Java.  Keep in mind that a warning about
-undeclaredMethod would just say that there wasn't some header file
that declared it in *some* context...


                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.