Re: Are STOP and INIT good names? (was Re: [PATCH pod/perlmod.pod 5.005_63] Special Blocks For Less Than Gurus)

[Gurusamy Sarathy <gsar@ActiveState.com>]

On Wed, 26 Jan 2000 09:31:33 +1100, Damian Conway wrote:
>   > I like DONE (as in "done with compilation"), which hasn't been
>   > proposed yet, and which conveys the meaning that these things are
>   > a finishing point for compilation, not a starting point for
>   > execution (which is what INIT is).
>   > 
>   >     BEGIN
>   >     DONE
>   >     INIT
>   >     END
>
>Hmmmm. To me the obvious order of these is:
>
>        BEGIN
>        INIT
>        DONE
>        END

Sorry, I don't see why this is any more obvious than the other, or
something else like:

         INIT
	 BEGIN
	 END
	 DONE

But as you said, I probably know too much. ;-)

>Remember that most "normal" folks (;-) don't have the sophisticated
>understanding of the perlguts that is enjoyed by most in this forum. We
>are a simple people who think, not in terms of lexing, compilation, then
>execution, but in terms of a single flow of control through the program
>source.

I think the "single flow of control" idea is the problem.  It really is
in two broad, distinct phases.

   Phase I (compile):
      BEGIN
      DONE
   Phase II (run):
      INIT
      END

Now, it just so happens that you can ignore DONE and pretend that it
is a "single flow of control", but it sure ain't like that.  I think
it is counter-productive to perpetuate that myth by choosing names
that "fit in" with the myth.

I wish we could do PARSE::BEGIN, PARSE::END, RUN::BEGIN, and END::RUN
(or even {BEGIN,END}::{PARSE,RUN}).  Maybe that's what we oughtta
do.


Sarathy
gsar@ActiveState.com

---

Follow-Ups from:

Nathan Torkington <gnat@frii.com>

References to:

Damian Conway <damian@cs.monash.edu.au>

---

[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index][Thread Index][Top&Search][Original]

---