Empathy List Archives

W

wmartin

Thu, Dec 4, 2014 11:29 PM

Hello,
Trying to do a simple "toggle" variable, 0-1-0-1... inside a for loop, using
the following code:
assign( k = (k==1) ? 0 : 1)
"k" is not changing as I expected.
ANY help or thoughts appreciated!

Bill Martin

--
View this message in context: http://forum.openscad.org/should-this-work-tp10351.html
Sent from the OpenSCAD mailing list archive at Nabble.com.

Hello, Trying to do a simple "toggle" variable, 0-1-0-1... inside a for loop, using the following code: assign( k = (k==1) ? 0 : 1) "k" is not changing as I expected. ANY help or thoughts appreciated! Bill Martin -- View this message in context: http://forum.openscad.org/should-this-work-tp10351.html Sent from the OpenSCAD mailing list archive at Nabble.com.

B

Bananapeel

Thu, Dec 4, 2014 11:34 PM

The documentation says Assign: "Set variables to a new value for a sub-tree."

for (i = [10:50])
{
assign (angle = i360/20, distance = i10, r = i*2)
{ // subtree starts here
rotate(angle, [1, 0, 0])
translate([0, distance, 0])
sphere(r = r);
} // subtree ends here
}

So you can't use it like that.

--
View this message in context: http://forum.openscad.org/should-this-work-tp10351p10352.html
Sent from the OpenSCAD mailing list archive at Nabble.com.

The documentation says Assign: "Set variables to a new value for a sub-tree." for (i = [10:50]) { assign (angle = i*360/20, distance = i*10, r = i*2) { // subtree starts here rotate(angle, [1, 0, 0]) translate([0, distance, 0]) sphere(r = r); } // subtree ends here } So you can't use it like that. -- View this message in context: http://forum.openscad.org/should-this-work-tp10351p10352.html Sent from the OpenSCAD mailing list archive at Nabble.com.

M

MichaelAtOz

Fri, Dec 5, 2014 12:36 AM

for(i=[0:10]) {
assign(k=i%2){
echo(i=i,k=k);
}
}

Unless specifically shown otherwise above, my contribution is in the Public Domain; To the extent possible under law, I have waived all copyright and related or neighbouring rights to this work. This work is published globally via the internet. :) Inclusion of works of previous authors is not included in the above.

The TPP is no simple “trade agreement.” Fight it! http://www.ourfairdeal.org/

View this message in context: http://forum.openscad.org/should-this-work-tp10351p10357.html
Sent from the OpenSCAD mailing list archive at Nabble.com.

for(i=[0:10]) { assign(k=i%2){ echo(i=i,k=k); } } ----- Unless specifically shown otherwise above, my contribution is in the Public Domain; To the extent possible under law, I have waived all copyright and related or neighbouring rights to this work. This work is published globally via the internet. :) Inclusion of works of previous authors is not included in the above. The TPP is no simple “trade agreement.” Fight it! http://www.ourfairdeal.org/ -- View this message in context: http://forum.openscad.org/should-this-work-tp10351p10357.html Sent from the OpenSCAD mailing list archive at Nabble.com.

EN

Ed Nisley

Fri, Dec 5, 2014 12:40 PM

for(i=[0:10]) {
assign(k=i%2){
echo(i=i,k=k);
}
}

It used to...

Last night I did a git pull / compile and discovered that Something
Changed in the last week or two, as a conditional test that formerly
worked no longer did.

The original test:
else if ((!(iy % 2) && !(ix % 2)))

What works now:
else if ((0 == (iy % 2)) && (0 == (ix % 2)))

The ix and iy integers iterate from 0 to maximum values, generating
coordinates at discrete points in the plane. At positions where the
indexes are both even, the program generates a particular type of chain
mail link.

It seems the ! operator recently stopped turning (integer) 0 into
(boolean) true, as the result remained obstinately false.

A similar test:
if ((iy % 2) && (ix % 2))

continues to function properly, generating a different link shape at the
(odd,odd) points.

The offending conditional is at the bottom of the listing at:

http://softsolder.com/2014/12/04/3d-printed-chain-mail-armor-cosplay-edition/

As the post observes, the code needs refactoring. I recompiled OpenSCAD
and started fiddling, only to discover half the chain mail links were
missing...

--
Ed
softsolder.com

> for(i=[0:10]) { > assign(k=i%2){ > echo(i=i,k=k); > } > } It used to... Last night I did a git pull / compile and discovered that Something Changed in the last week or two, as a conditional test that formerly worked no longer did. The original test: else if ((!(iy % 2) && !(ix % 2))) What works now: else if ((0 == (iy % 2)) && (0 == (ix % 2))) The ix and iy integers iterate from 0 to maximum values, generating coordinates at discrete points in the plane. At positions where the indexes are both even, the program generates a particular type of chain mail link. It seems the ! operator recently stopped turning (integer) 0 into (boolean) true, as the result remained obstinately false. A similar test: if ((iy % 2) && (ix % 2)) continues to function properly, generating a different link shape at the (odd,odd) points. The offending conditional is at the bottom of the listing at: http://softsolder.com/2014/12/04/3d-printed-chain-mail-armor-cosplay-edition/ As the post observes, the code needs refactoring. I recompiled OpenSCAD and started fiddling, only to discover half the chain mail links were missing... -- Ed softsolder.com

TP

Torsten Paul

Fri, Dec 5, 2014 5:40 PM

On 12/05/2014 01:40 PM, Ed Nisley wrote:

It used to...

Last night I did a git pull / compile and discovered that Something
Changed in the last week or two, as a conditional test that formerly
worked no longer did.

Yes, there was quite a big change to support the better recursion
handling. I guess that introduced the bug. I'll have a look.

ciao,
Torsten.

On 12/05/2014 01:40 PM, Ed Nisley wrote: > It used to... > > Last night I did a git pull / compile and discovered that Something > Changed in the last week or two, as a conditional test that formerly > worked no longer did. > Yes, there was quite a big change to support the better recursion handling. I guess that introduced the bug. I'll have a look. ciao, Torsten.

TP

Torsten Paul

Fri, Dec 5, 2014 8:09 PM

On 12/05/2014 01:40 PM, Ed Nisley wrote:

The original test:
else if ((!(iy % 2) && !(ix % 2)))

This should be fixed now.

ciao,
Torsten.

On 12/05/2014 01:40 PM, Ed Nisley wrote: > The original test: > else if ((!(iy % 2) && !(ix % 2))) > This should be fixed now. ciao, Torsten.

EN

Ed Nisley

Fri, Dec 5, 2014 8:37 PM

On 12/05/2014 03:09 PM, Torsten Paul wrote:

This should be fixed now.

And so it is!

Thanks...

--
Ed
softsolder.com

On 12/05/2014 03:09 PM, Torsten Paul wrote: > This should be fixed now. And so it is! Thanks... -- Ed softsolder.com

discuss@lists.openscad.org

should this work?

The TPP is no simple “trade agreement.” Fight it! http://www.ourfairdeal.org/