Add a step size to get rid of the warming. E.g.  for (i=[1 : 1 : n-1])

On Sat, 8 Aug 2020 at 07:24, Rogier Wolff R.E.Wolff@bitwizard.nl wrote:

On Sat, Aug 08, 2020 at 09:13:46AM +0100, nop head wrote:

OK! Thanks for the tip. Where in the code?

Actually, its relatively easy: The code is only 282 lines long.
And there are only three for loops:

240:  for (i=[0:nx-1]) {
241:    for (j=[0:ny-1]) {
257:  for (i=[0:180:180])

nx and ny are both "2" today.

Roger. 

P.S. I just tried adding the implicit :1 and it makes the whole thing
stop working. Without the :1 I get the 4 modules, with the added :1 I
get nothing.

Here is my code:

module m4 (nx=2, ny=2, model=0)
{
p=75;

for (i=[0:nx-1]) {  // this is line 240.
for (j=[0:ny-1]) {
translate ([ip,jp,0]) mymodule (model);
}
}
}
...

m4 (nx=2, ny=2, 0);

Note: The m4 module started out doing 2x2 or 4 modules as I'm
rendering/printing today, but since then I added the parameters but
have not changed the name.

--
** R.E.Wolff@BitWizard.nl ** https://www.BitWizard.nl/ ** +31-15-2049110 **
**    Delftechpark 11 2628 XJ  Delft, The Netherlands.  KVK: 27239233    **
The plan was simple, like my brother-in-law Phil. But unlike
Phil, this plan just might work.

On Sat, Aug 08, 2020 at 10:36:55AM +0200, Rogier Wolff wrote:

I just implemented a "full program" that shows this problem.
Below is the code, no warning, not working, remove the :1 and
it does what I expect of it.

module mymodule (model)
{
cylinder (d=65,h=10);
}

module m4 (nx=2, ny=2, model=0)
{
p=75;

for (i=[0:1:nx-1]) {  // this is line 240.
for (j=[0:ny-1]) {
translate ([ip,jp,0]) mymodule (model);
}
}
}

m4 (nx=2, ny=2, 0);

--
** R.E.Wolff@BitWizard.nl ** https://www.BitWizard.nl/ ** +31-15-2049110 **
**    Delftechpark 11 2628 XJ  Delft, The Netherlands.  KVK: 27239233    **
The plan was simple, like my brother-in-law Phil. But unlike
Phil, this plan just might work.

Sorry, in your case nx is zero so it is looping from 0 to -1, so a negative
step would get rid of the warning.

I think you meant to set model to 0 but it is actually setting nx to zeo.

On Sat, 8 Aug 2020 at 09:43, Rogier Wolff R.E.Wolff@bitwizard.nl wrote:

On Sat, Aug 08, 2020 at 09:59:01AM +0100, nop head wrote:

So me redundantly specifying the name of the first and second
parameter causes the third parameter (of the call) to overwrite the
just explicitly specified-by-name first parameter (of the called
module)?????? You're shitting me, right?

Now that's something what might warrant a warning if not a rewrite to
change that behaviour....

Roger.

...

--
** R.E.Wolff@BitWizard.nl ** https://www.BitWizard.nl/ ** +31-15-2049110 **
**    Delftechpark 11 2628 XJ  Delft, The Netherlands.  KVK: 27239233    **
The plan was simple, like my brother-in-law Phil. But unlike
Phil, this plan just might work.

Yes the first unnamed parameter is assigned to position 1 and so on. In
Python you get "SyntaxError: positional argument follows keyword argument"

On Sat, 8 Aug 2020 at 12:25, Rogier Wolff R.E.Wolff@bitwizard.nl wrote:

On 8/7/2020 11:23 PM, Rogier Wolff wrote:

That would be good.  But the historical behavior is that it just
reverses the two endpoints.  Pretty much everybody agrees that's not
desirable, hence the warning.

for (i=[3:1]) echo(i);

DEPRECATED: Using ranges of the form [begin:end] with begin value
greater than the end value is deprecated.

ECHO: 1

ECHO: 2

ECHO: 3

Yes, the new behaviour makes more sense as I can't remember a writing a
loop where I wanted to go in both directions. And now we don't need an if
to test for the empty case.

I.e.
if(n) for(i = [0 : n - 1]) ...

becomes

for(i = [0 : 1 : n - 1]) ...

and in Rogier's case the warning found another bug, because the loop was
actually running 0, -1 instead of 0, 1

On Sun, 9 Aug 2020 at 07:40, OzAtMichael oz.at.michael@gmail.com wrote: