ED
Ethan Dicks
Mon, Mar 9, 2015 8:15 PM
Hi, All,
I've been working with OpenSCAD for a few years, and as such, I can
get blinded to better ways to do things because I already have some
way to do them.
Case in point. I want to make a square cover plate and put holes in
each of the corners. There's a brute force way to just difference()
the plate and the four holes, but is there a "pretty" way to do this?
(crude) code example:
difference() {
cube([100,100,1], center=true);
translate([45, 45, -1])cylinder(r=2, h=2);
translate([45, -45, -1])cylinder(r=2, h=2);
translate([-45, 45, -1])cylinder(r=2, h=2);
translate([-45, -45, -1])cylinder(r=2, h=2);
}
What I'm after is some sort of more elegant structure to replace 4
nearly identical lines of code making one hole each. For this sort of
thing, it's not hard to do it, but perhaps each "hole" is more
elaborate, requiring more than a simple "cylinder()" call
(counter-sunk, threaded, etc) so I'd rather find a nice way to
describe a hole and iterate over some structure that results in ([X,
Y], [-X, Y], [X, -Y], [-X, -Y]) somehow.
I can think of several ways to tackle this. What I'm asking for is,
is this a solved problem and is there a "best" way to do this in the
existing framework?
Thanks!
-ethan
Hi, All,
I've been working with OpenSCAD for a few years, and as such, I can
get blinded to better ways to do things because I already have _some_
way to do them.
Case in point. I want to make a square cover plate and put holes in
each of the corners. There's a brute force way to just difference()
the plate and the four holes, but is there a "pretty" way to do this?
(crude) code example:
difference() {
cube([100,100,1], center=true);
translate([45, 45, -1])cylinder(r=2, h=2);
translate([45, -45, -1])cylinder(r=2, h=2);
translate([-45, 45, -1])cylinder(r=2, h=2);
translate([-45, -45, -1])cylinder(r=2, h=2);
}
What I'm after is some sort of more elegant structure to replace 4
nearly identical lines of code making one hole each. For this sort of
thing, it's not hard to do it, but perhaps each "hole" is more
elaborate, requiring more than a simple "cylinder()" call
(counter-sunk, threaded, etc) so I'd rather find a nice way to
describe a hole and iterate over some structure that results in ([X,
Y], [-X, Y], [X, -Y], [-X, -Y]) somehow.
I can think of several ways to tackle this. What I'm asking for is,
is this a solved problem and is there a "best" way to do this in the
existing framework?
Thanks!
-ethan
W
whosawhatsis
Mon, Mar 9, 2015 8:19 PM
for(x = [45, -45], y = [45, -45]) translate([x, y, -1]) cylinder(r=2, h=2);
I also often do it like this, to make it a bit easier to adjust the values symmetrically later:
for(x = [1, -1], y = [1, -1]) translate([x * 45, y * 45, -1]) cylinder(r=2, h=2);
On Monday, March 9, 2015 at 13:15, Ethan Dicks wrote:
Hi, All,
I've been working with OpenSCAD for a few years, and as such, I can
get blinded to better ways to do things because I already have some
way to do them.
Case in point. I want to make a square cover plate and put holes in
each of the corners. There's a brute force way to just difference()
the plate and the four holes, but is there a "pretty" way to do this?
(crude) code example:
difference() {
cube([100,100,1], center=true);
translate([45, 45, -1])cylinder(r=2, h=2);
translate([45, -45, -1])cylinder(r=2, h=2);
translate([-45, 45, -1])cylinder(r=2, h=2);
translate([-45, -45, -1])cylinder(r=2, h=2);
}
What I'm after is some sort of more elegant structure to replace 4
nearly identical lines of code making one hole each. For this sort of
thing, it's not hard to do it, but perhaps each "hole" is more
elaborate, requiring more than a simple "cylinder()" call
(counter-sunk, threaded, etc) so I'd rather find a nice way to
describe a hole and iterate over some structure that results in ([X,
Y], [-X, Y], [X, -Y], [-X, -Y]) somehow.
I can think of several ways to tackle this. What I'm asking for is,
is this a solved problem and is there a "best" way to do this in the
existing framework?
Thanks!
-ethan
OpenSCAD mailing list
Discuss@lists.openscad.org (mailto:Discuss@lists.openscad.org)
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
for(x = [45, -45], y = [45, -45]) translate([x, y, -1]) cylinder(r=2, h=2);
I also often do it like this, to make it a bit easier to adjust the values symmetrically later:
for(x = [1, -1], y = [1, -1]) translate([x * 45, y * 45, -1]) cylinder(r=2, h=2);
On Monday, March 9, 2015 at 13:15, Ethan Dicks wrote:
> Hi, All,
>
> I've been working with OpenSCAD for a few years, and as such, I can
> get blinded to better ways to do things because I already have _some_
> way to do them.
>
> Case in point. I want to make a square cover plate and put holes in
> each of the corners. There's a brute force way to just difference()
> the plate and the four holes, but is there a "pretty" way to do this?
>
> (crude) code example:
>
> difference() {
> cube([100,100,1], center=true);
> translate([45, 45, -1])cylinder(r=2, h=2);
> translate([45, -45, -1])cylinder(r=2, h=2);
> translate([-45, 45, -1])cylinder(r=2, h=2);
> translate([-45, -45, -1])cylinder(r=2, h=2);
> }
>
> What I'm after is some sort of more elegant structure to replace 4
> nearly identical lines of code making one hole each. For this sort of
> thing, it's not hard to do it, but perhaps each "hole" is more
> elaborate, requiring more than a simple "cylinder()" call
> (counter-sunk, threaded, etc) so I'd rather find a nice way to
> describe a hole and iterate over some structure that results in ([X,
> Y], [-X, Y], [X, -Y], [-X, -Y]) somehow.
>
> I can think of several ways to tackle this. What I'm asking for is,
> is this a solved problem and is there a "best" way to do this in the
> existing framework?
>
> Thanks!
>
> -ethan
>
> _______________________________________________
> OpenSCAD mailing list
> Discuss@lists.openscad.org (mailto:Discuss@lists.openscad.org)
> http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
>
>
AP
Andrew Plumb
Mon, Mar 9, 2015 8:21 PM
Assuming symmetry about the origin, mirror() should do the trick:
http://en.wikibooks.org/wiki/OpenSCAD_User_Manual/The_OpenSCAD_Language#mirror http://en.wikibooks.org/wiki/OpenSCAD_User_Manual/The_OpenSCAD_Language#mirror
difference() {
cube([100,100,1], center=true);
for(mx=[0:1],my=[0:1]) mirror([mx,my,0]) translate([45, 45, -1]) cylinder(r=2, h=2);
}
Andrew.
On Mar 9, 2015, at 4:15 PM, Ethan Dicks ethan.dicks@gmail.com wrote:
Hi, All,
I've been working with OpenSCAD for a few years, and as such, I can
get blinded to better ways to do things because I already have some
way to do them.
Case in point. I want to make a square cover plate and put holes in
each of the corners. There's a brute force way to just difference()
the plate and the four holes, but is there a "pretty" way to do this?
(crude) code example:
difference() {
cube([100,100,1], center=true);
translate([45, 45, -1])cylinder(r=2, h=2);
translate([45, -45, -1])cylinder(r=2, h=2);
translate([-45, 45, -1])cylinder(r=2, h=2);
translate([-45, -45, -1])cylinder(r=2, h=2);
}
What I'm after is some sort of more elegant structure to replace 4
nearly identical lines of code making one hole each. For this sort of
thing, it's not hard to do it, but perhaps each "hole" is more
elaborate, requiring more than a simple "cylinder()" call
(counter-sunk, threaded, etc) so I'd rather find a nice way to
describe a hole and iterate over some structure that results in ([X,
Y], [-X, Y], [X, -Y], [-X, -Y]) somehow.
I can think of several ways to tackle this. What I'm asking for is,
is this a solved problem and is there a "best" way to do this in the
existing framework?
Thanks!
-ethan
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Assuming symmetry about the origin, mirror() should do the trick:
http://en.wikibooks.org/wiki/OpenSCAD_User_Manual/The_OpenSCAD_Language#mirror <http://en.wikibooks.org/wiki/OpenSCAD_User_Manual/The_OpenSCAD_Language#mirror>
difference() {
cube([100,100,1], center=true);
for(mx=[0:1],my=[0:1]) mirror([mx,my,0]) translate([45, 45, -1]) cylinder(r=2, h=2);
}
Andrew.
> On Mar 9, 2015, at 4:15 PM, Ethan Dicks <ethan.dicks@gmail.com> wrote:
>
> Hi, All,
>
> I've been working with OpenSCAD for a few years, and as such, I can
> get blinded to better ways to do things because I already have _some_
> way to do them.
>
> Case in point. I want to make a square cover plate and put holes in
> each of the corners. There's a brute force way to just difference()
> the plate and the four holes, but is there a "pretty" way to do this?
>
> (crude) code example:
>
> difference() {
> cube([100,100,1], center=true);
> translate([45, 45, -1])cylinder(r=2, h=2);
> translate([45, -45, -1])cylinder(r=2, h=2);
> translate([-45, 45, -1])cylinder(r=2, h=2);
> translate([-45, -45, -1])cylinder(r=2, h=2);
> }
>
> What I'm after is some sort of more elegant structure to replace 4
> nearly identical lines of code making one hole each. For this sort of
> thing, it's not hard to do it, but perhaps each "hole" is more
> elaborate, requiring more than a simple "cylinder()" call
> (counter-sunk, threaded, etc) so I'd rather find a nice way to
> describe a hole and iterate over some structure that results in ([X,
> Y], [-X, Y], [X, -Y], [-X, -Y]) somehow.
>
> I can think of several ways to tackle this. What I'm asking for is,
> is this a solved problem and is there a "best" way to do this in the
> existing framework?
>
> Thanks!
>
> -ethan
>
> _______________________________________________
> OpenSCAD mailing list
> Discuss@lists.openscad.org
> http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
AP
Andrew Plumb
Mon, Mar 9, 2015 8:25 PM
…and of course it goes (almost) without saying that you don’t want to use mirror for threaded structures. :-)
For more complex structures and placements, consider wrapping them in a “child placer” module definition, to decouple the shape from the placement.
Andrew.
On Mar 9, 2015, at 4:21 PM, Andrew Plumb andrew@plumb.org wrote:
Assuming symmetry about the origin, mirror() should do the trick:
http://en.wikibooks.org/wiki/OpenSCAD_User_Manual/The_OpenSCAD_Language#mirror http://en.wikibooks.org/wiki/OpenSCAD_User_Manual/The_OpenSCAD_Language#mirror
difference() {
cube([100,100,1], center=true);
for(mx=[0:1],my=[0:1]) mirror([mx,my,0]) translate([45, 45, -1]) cylinder(r=2, h=2);
}
Andrew.
On Mar 9, 2015, at 4:15 PM, Ethan Dicks <ethan.dicks@gmail.com mailto:ethan.dicks@gmail.com> wrote:
Hi, All,
I've been working with OpenSCAD for a few years, and as such, I can
get blinded to better ways to do things because I already have some
way to do them.
Case in point. I want to make a square cover plate and put holes in
each of the corners. There's a brute force way to just difference()
the plate and the four holes, but is there a "pretty" way to do this?
(crude) code example:
difference() {
cube([100,100,1], center=true);
translate([45, 45, -1])cylinder(r=2, h=2);
translate([45, -45, -1])cylinder(r=2, h=2);
translate([-45, 45, -1])cylinder(r=2, h=2);
translate([-45, -45, -1])cylinder(r=2, h=2);
}
What I'm after is some sort of more elegant structure to replace 4
nearly identical lines of code making one hole each. For this sort of
thing, it's not hard to do it, but perhaps each "hole" is more
elaborate, requiring more than a simple "cylinder()" call
(counter-sunk, threaded, etc) so I'd rather find a nice way to
describe a hole and iterate over some structure that results in ([X,
Y], [-X, Y], [X, -Y], [-X, -Y]) somehow.
I can think of several ways to tackle this. What I'm asking for is,
is this a solved problem and is there a "best" way to do this in the
existing framework?
Thanks!
-ethan
OpenSCAD mailing list
Discuss@lists.openscad.org mailto:Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
…and of course it goes (almost) without saying that you don’t want to use mirror for threaded structures. :-)
For more complex structures and placements, consider wrapping them in a “child placer” module definition, to decouple the shape from the placement.
Andrew.
> On Mar 9, 2015, at 4:21 PM, Andrew Plumb <andrew@plumb.org> wrote:
>
> Assuming symmetry about the origin, mirror() should do the trick:
>
> http://en.wikibooks.org/wiki/OpenSCAD_User_Manual/The_OpenSCAD_Language#mirror <http://en.wikibooks.org/wiki/OpenSCAD_User_Manual/The_OpenSCAD_Language#mirror>
>
> difference() {
> cube([100,100,1], center=true);
> for(mx=[0:1],my=[0:1]) mirror([mx,my,0]) translate([45, 45, -1]) cylinder(r=2, h=2);
> }
>
> Andrew.
>
>> On Mar 9, 2015, at 4:15 PM, Ethan Dicks <ethan.dicks@gmail.com <mailto:ethan.dicks@gmail.com>> wrote:
>>
>> Hi, All,
>>
>> I've been working with OpenSCAD for a few years, and as such, I can
>> get blinded to better ways to do things because I already have _some_
>> way to do them.
>>
>> Case in point. I want to make a square cover plate and put holes in
>> each of the corners. There's a brute force way to just difference()
>> the plate and the four holes, but is there a "pretty" way to do this?
>>
>> (crude) code example:
>>
>> difference() {
>> cube([100,100,1], center=true);
>> translate([45, 45, -1])cylinder(r=2, h=2);
>> translate([45, -45, -1])cylinder(r=2, h=2);
>> translate([-45, 45, -1])cylinder(r=2, h=2);
>> translate([-45, -45, -1])cylinder(r=2, h=2);
>> }
>>
>> What I'm after is some sort of more elegant structure to replace 4
>> nearly identical lines of code making one hole each. For this sort of
>> thing, it's not hard to do it, but perhaps each "hole" is more
>> elaborate, requiring more than a simple "cylinder()" call
>> (counter-sunk, threaded, etc) so I'd rather find a nice way to
>> describe a hole and iterate over some structure that results in ([X,
>> Y], [-X, Y], [X, -Y], [-X, -Y]) somehow.
>>
>> I can think of several ways to tackle this. What I'm asking for is,
>> is this a solved problem and is there a "best" way to do this in the
>> existing framework?
>>
>> Thanks!
>>
>> -ethan
>>
>> _______________________________________________
>> OpenSCAD mailing list
>> Discuss@lists.openscad.org <mailto:Discuss@lists.openscad.org>
>> http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
>
> _______________________________________________
> OpenSCAD mailing list
> Discuss@lists.openscad.org
> http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
ED
Ethan Dicks
Mon, Mar 9, 2015 8:49 PM
for(x = [45, -45], y = [45, -45]) translate([x, y, -1]) cylinder(r=2, h=2);
Now that you shared this, I see it everywhere (including several times
in library code I was already calling!)
I also often do it like this, to make it a bit easier to adjust the values
symmetrically later:
for(x = [1, -1], y = [1, -1]) translate([x * 45, y * 45, -1]) cylinder(r=2,
h=2);
I'm now trying to figure out which form is "more elegant" to me.
The first example puts the complexity in the "for" expression, and the
second example pushes it down a level.
I appreciate all the suggestions. Once I have the part modeled (for
eventual fabrication and casting in aluminum) and the results are
unveiled to the recipient, I want the code to be pretty enough to
share, so the quality matters to me.
Cheers,
-ethan
On Mon, Mar 9, 2015 at 4:19 PM, whosawhatsis <whosawhatsis@gmail.com> wrote:
> for(x = [45, -45], y = [45, -45]) translate([x, y, -1]) cylinder(r=2, h=2);
Now that you shared this, I see it everywhere (including several times
in library code I was already calling!)
> I also often do it like this, to make it a bit easier to adjust the values
> symmetrically later:
>
> for(x = [1, -1], y = [1, -1]) translate([x * 45, y * 45, -1]) cylinder(r=2,
> h=2);
I'm now trying to figure out which form is "more elegant" to me.
The first example puts the complexity in the "for" expression, and the
second example pushes it down a level.
I appreciate all the suggestions. Once I have the part modeled (for
eventual fabrication and casting in aluminum) and the results are
unveiled to the recipient, I want the code to be pretty enough to
share, so the quality matters to me.
Cheers,
-ethan
EN
Ed Nisley
Mon, Mar 9, 2015 10:21 PM
On 03/09/2015 04:49 PM, Ethan Dicks wrote:
The first example puts the complexity in the "for" expression,
and the second example pushes it down a level.
Even though it marks me as old school, I generally separate iteration
and calculation:
- loop constructs use integer operands and simple logic
- operations in the loop calculate coordinates / rotations
That way, even I can figure out what the loop will do and, when I
inevitably change the coordinates, I don't screw up the loop logic.
Which is why they don't let me use list comprehension... [grin]
--
Ed
softsolder.com
On 03/09/2015 04:49 PM, Ethan Dicks wrote:
> The first example puts the complexity in the "for" expression,
> and the second example pushes it down a level.
Even though it marks me as old school, I generally separate iteration
and calculation:
- loop constructs use integer operands and simple logic
- operations in the loop calculate coordinates / rotations
That way, even I can figure out what the loop will do and, when I
inevitably change the coordinates, I don't screw up the loop logic.
Which is why they don't let me use list comprehension... [grin]
--
Ed
softsolder.com
H
hagen
Tue, Mar 10, 2015 9:34 PM
the child placer code (without for loop) could be something like:
module corners(o)
{
translate([o, o, -1]) children(0);
translate([o, -o, -1]) children(0);
translate([-o, o, -1]) children(0);
translate([-o, -o, -1]) children(0);
}
difference() {
cube([100,100,1], center=true);
corners(45) cylinder(r=2, h=2);
}
--
View this message in context: http://forum.openscad.org/looking-for-symmetry-shortcuts-tp11950p11960.html
Sent from the OpenSCAD mailing list archive at Nabble.com.
the child placer code (without for loop) could be something like:
module corners(o)
{
translate([o, o, -1]) children(0);
translate([o, -o, -1]) children(0);
translate([-o, o, -1]) children(0);
translate([-o, -o, -1]) children(0);
}
difference() {
cube([100,100,1], center=true);
corners(45) cylinder(r=2, h=2);
}
--
View this message in context: http://forum.openscad.org/looking-for-symmetry-shortcuts-tp11950p11960.html
Sent from the OpenSCAD mailing list archive at Nabble.com.
JC
JinHwan Choi
Wed, Mar 11, 2015 9:53 AM
Relativity library provides neat functions wrapping for loop and translations.
below codes do the same thing with relativity.scad included.
difference() {
cube([100,100,1], center=true);
rotated(90*z,[0:3]) translate([45,45,-1]) cylinder(r=2,h=2);
}
difference() {
cube([100,100,1], center=true);
mirrored(x) mirrored(y) translate([45,45,-1]) cylinder(r=2,h=2);
}
More informations here.
https://github.com/davidson16807/relativity.scad
2015-03-11 6:34 GMT+09:00 hagen temp4pieter@hotmail.com:
the child placer code (without for loop) could be something like:
module corners(o)
{
translate([o, o, -1]) children(0);
translate([o, -o, -1]) children(0);
translate([-o, o, -1]) children(0);
translate([-o, -o, -1]) children(0);
}
difference() {
cube([100,100,1], center=true);
corners(45) cylinder(r=2, h=2);
}
--
View this message in context: http://forum.openscad.org/looking-for-symmetry-shortcuts-tp11950p11960.html
Sent from the OpenSCAD mailing list archive at Nabble.com.
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Relativity library provides neat functions wrapping for loop and translations.
below codes do the same thing with relativity.scad included.
difference() {
cube([100,100,1], center=true);
rotated(90*z,[0:3]) translate([45,45,-1]) cylinder(r=2,h=2);
}
difference() {
cube([100,100,1], center=true);
mirrored(x) mirrored(y) translate([45,45,-1]) cylinder(r=2,h=2);
}
More informations here.
https://github.com/davidson16807/relativity.scad
2015-03-11 6:34 GMT+09:00 hagen <temp4pieter@hotmail.com>:
> the child placer code (without for loop) could be something like:
>
> module corners(o)
> {
> translate([o, o, -1]) children(0);
> translate([o, -o, -1]) children(0);
> translate([-o, o, -1]) children(0);
> translate([-o, -o, -1]) children(0);
> }
>
> difference() {
> cube([100,100,1], center=true);
> corners(45) cylinder(r=2, h=2);
> }
>
>
>
> --
> View this message in context: http://forum.openscad.org/looking-for-symmetry-shortcuts-tp11950p11960.html
> Sent from the OpenSCAD mailing list archive at Nabble.com.
>
> _______________________________________________
> OpenSCAD mailing list
> Discuss@lists.openscad.org
> http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
