On Sun, 12 Jun 2005 18:40:48 +0100, <[EMAIL PROTECTED]> wrote:
> GP lisper <[EMAIL PROTECTED]> writes:
>> On Sun, 12 Jun 2005 12:30:01 +0300 (EEST), <[EMAIL PROTECTED]> wrote:
>>> On Sun, 12 Jun 2005, Nicolas Neuss wrote:
>>>
>>>>>>> (loop with m = 5
>>>>>>> initially (setq m 3)
>>>>>>> for i below m do (princ i))
>>>>>>>
>>>>>>> should print only "012", no?
>>
>> After reading 6.1.1.6, I say "no" still. The progn for the prologue is:
>> (setq m 3)
>>
>> and then 'm' is set to 5 every pass thru the loop.
>>
>> On the other hand, I keep wondering what you see that I miss...
>
> From the page which you said you read:
>
> A with clause introduces a variable binding and an optional initial
> value. The initial values are calculated in the order in which the
> with clauses occur.
>
> Things which "with" clauses do not do include any kind of stepping or
> setting at each iteration round the loop.
(loop with m = 5
initially (setq m 0)
for i below m do (princ i))
==> 01234
But
(loop with m = 5
initially (setq m 0)
for i below m do (princ m))
==> 00000
Which matches your reply.
So the clause "for i below m do (princ m)" is translated to "for i
below 5 do (princ m)" before the loop prologue is run. That follows
"All variables are initialized first, regardless of where the
establishing clauses appear in the source." (JonL CLtL2 26.3.1)
Thanks! I think...depends if I get run over in the next c.l.l. LOOP
thread. Or more likely, how bad I get run over....
--
The LOOP construct is really neat, it's got lots of knobs to turn!
Don't push the yellow one on the bottom.
