FOR loop decrementing

[RyanBurnside]

Hello,

I've noticed that in the InterLisp Exec I can run something like

(FOR I FROM 3 TO 0 COLLECT I) 

giving (3 2 1 0)

While in a saved function it simply doesn't loop.
I know InterLisp features CLISP but do these things ONLY work in the Exec?

I tried defining the function with DEFINEQ as well in the Exec but even there, it was not corrected as it is outside the DEFINEQ definition.

It's a bit annoying that code can work in the Exec but then suddenly has totally different behavior when used in a function. BUT part of learning a new system is to of course accept the way it works.

Sincerely,
Ryan

[pamoroso]

It works for me by defining the function with SEdit:

But also when defining it with DEFINEQ from the Exec:

With DEFINEQ also in a fresh session:

[RyanBurnside]

Thanks for testing it.
I didn't realize that using a parameter from outside the for loop would cause the bad behavior for the starting value.

[nbriggs]

 If you expand (meta-x in sedit is one way) your function with a for loop, you can see what it's really doing. This is FOO above -- (LAMBDA NIL (* ; "Edited 18-Oct-2025 09:40 by briggs") (PROG (($$TEM3 0) $$VAL $$TEM2 $$TEM1 (I 3)) $$LP (COND ((LESSP I $$TEM3) (GO $$OUT))) (SETQ $$TEM1 I) (COND ($$TEM2 (FRPLACD $$TEM2 (SETQ $$TEM2 (LIST $$TEM1)))) (T (SETQ $$VAL (SETQ $$TEM2 (LIST $$TEM1))))) $$ITERATE (SETQ I (PLUS I -1)) (GO $$LP) $$OUT (RETURN $$VAL))) What I'm guessing is that you didn't have constants in the FOR loop in your function case, so it assumed a default positive increment, which will indeed not perform any iterations, being the equivalent of 2/20_ (FOR I FROM 3 TO 1 BY 1 COLLECT I) NIL 2/21_ (FOR I FROM 3 TO 1 BY -1 COLLECT I) (3 2 1)

…

-- Nick

On Oct 17, 2025, at 23:57, Ryan Burnside ***@***.***> wrote: Hello, I've noticed that in the InterLisp Exec I can run something like (FOR I FROM 3 TO 0 COLLECT I) giving (3 2 1 0) While in a saved function it simply doesn't loop. I know InterLisp features CLISP but do these things ONLY work in the Exec? I tried defining the function with DEFINEQ as well in the Exec but even there, it was not corrected as it is outside the DEFINEQ definition. It's a bit annoying that code can work in the Exec but then suddenly has totally different behavior when used in a function. BUT part of learning a new system is to of course accept the way it works. Sincerely, Ryan — Reply to this email directly, view it on GitHub <#2318>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB6DAWIM4TLNJNRMDVM77AL3YHQEZAVCNFSM6AAAAACJRJM43SVHI2DSMVQWIX3LMV43ERDJONRXK43TNFXW4OZZGAZTSOJQGI>. You are receiving this because you are subscribed to this thread.

[RyanBurnside]

Ah, thanks for the tip about macro expansion.
So it gets fixed in the Exec but I need the decrement in my code.
I'll just know to provide the decrement from now on.

Here is what I had prior.

[nbriggs]

This is, by the way, all documented in the TO clause of the FOR I.S.OPRS section in the manual (and DInfo) --

Note:  When both the operands to TO and FROM are numbers, and TO's operand
is less than FROM's operand, the i.v. is decremented by 1 after each iteration.
In this case, the i.s. terminates when the i.v. becomes less than the value of FORM.
For example, (from 10 to 1 do PRINT) prints the numbers from 10 down to 1.

[masinter]

When CLISP was being developed by Warren Teitelman, the desires for ease of use vs. consistent and predictable results were in some flux. Interlisp is more 'forgiving' in lots of ways, even when the forgiving behavior is a little too loose. We see this conflict in lots of places but it's rarely pointed out in the documentation. pathnames vs. symbols vs. strings for files, coercion of fonts and font names and other ways of indicating the desired font.... there are hundreds of cases. DWIM isn't a 'feature' as much as it is a philosophy. https://LarryMasinter.net https://interlisp.org

…

On Sat, Oct 18, 2025 at 11:34 AM Ryan Burnside ***@***.***> wrote: Thanks for testing it. I didn't realize that using a parameter from outside the for loop would cause the bad behavior for the starting value. — Reply to this email directly, view it on GitHub <#2318 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAIQTKYLESEQ363DT3ZRJLL3YKBZ5AVCNFSM6AAAAACJRJM43SVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTINZRHAZTOMY> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>