On Monday, September 2, 2024 at 01:37:08 PM EDT, Sanjeev Prabhakar via Discuss discuss@lists.openscad.org wrote:

The perceived rudeness was one of omission and a failure to consider the overall culture and bandwidth of other folks here.

Some folks are only interested in OpenSCAD as OpenSCAD, and wish to avoid the complications of 3rd party tools or additional programming tools.

Identifying such discussions in the subject line allows folks to either:

 - delete and ignore if not interested
 - read avidly and focus on such discussions intently if interested

Hopefully we can all be considerate of each other and remember that these messages are being inserted into folk's e-mail inboxes.

William

--
Sphinx of black quartz, judge my vow
https://designinto3d.com/

On Monday, September 2, 2024 at 01:32:32 PM EDT, Leonard Martin Struttmann via Discuss discuss@lists.openscad.org wrote:

Nice!

It is definitely a good example of OpenSCAD being suited to things amenable to being described mathematically and bounded by one's facility/fluency with mathematics.

On Mon, Sep 2, 2024 at 12:06 PM Sanjeev Prabhakar via Discuss discuss@lists.openscad.org wrote:

For folks who aren't familiar with this, there is a page on it at:

https://pythonscad.org/

I've been using it in an OpenSCAD module:

https://github.com/WillAdams/gcodepreview

and am currently reading through John C. Ousterhout's A Philosophy of Software Design and applying those principles which has been interesting and productive (though it has spurred a couple of bug reports, and more importantly, fixes).

OpenSCAD being the focused and limited tool which it is has aided it immensely in being implemented, and also making it approachable and accessible and easing distribution of code (no need to worry about malware) to the benefit of a wide variety of users.

OpenSCAD lacking mutable variables or the ability to write out files has been a marked limitation in the ability of some folks to make use of it some ways (using a 3rd party tool to parse Log files so as to write out a BOM comes to mind).

Perhaps a new path forward will be Python as an optional mode in OpenSCAD where the current status quo is the default, but where folks who wish it may enable the added language and use it either natively, or by calling it from OpenSCAD code.

William

--
Sphinx of black quartz, judge my vow.
https://designinto3d.com/

yes, i have also tested cadquery (actually build123d) and it's very
powerful.

openscad is not better or worse than cadquery, it's just 2 different
programs with different focuses and different strengths .
I like both of them!

Most value of cadquery is in its powerful opc kernel and you just " import
" into your python to make use of it.
Some time back I successfully imported the kernel into pythonscad and it
was working fine in linux.

see here:
[image: image.png]

build123d is just a cadquery derivative, it works like a charm.
why dont use both world in one model ?

On Mon, Sep 2, 2024 at 7:43 PM Kevin Cole via Discuss <
discuss@lists.openscad.org> wrote:

Thanks Leonard
I know it's quite complex for anyone to understand.
I have hard time explaining concepts maybe due to my not so great English
skills. Explaining through pictures and videos take a long time.
Never mind

On Mon, 2 Sept, 2024, 11:02 pm Leonard Martin Struttmann via Discuss, <
discuss@lists.openscad.org> wrote:

Isn't it the case that Sanjeev's approach is to simply use OpenSCAD to do
union or intersection of separate objects where the objects are all
specified in openscad using polyhedron() modules and their coordinates are
computed in Python?  Basically Sanjeev is simply building his own geometry
engine in Python to perform the geometrical calculations and hence produce
the polyhedra---basically building a Python library for geometry.  As such,
very little actual OpenSCAD is involved and the scad files are just huge
point lists defining the result.

The main limitation here is that if Sanjeev is going to write a library, he
also needs to write documentation.  It's never the programmer's favorite
task.  Without a reference manual, nobody else can use it.

On Mon, Sep 2, 2024 at 9:55 PM Sanjeev Prabhakar via Discuss <
discuss@lists.openscad.org> wrote:

On Monday, September 2, 2024 at 09:55:05 PM EDT, Sanjeev Prabhakar via Discuss discuss@lists.openscad.org wrote:

Far better than any of languages I've been taught over the years.

and it is much appreciated --- I've learned a great deal from your efforts and hope to learn more.

Please do mind --- I and many other folks I'm sure are very appreciative of your generosity in sharing your work.

If you want some other venue to share your designs, there is always Reddit:

https://old.reddit.com/r/openscad/

and there's also a subreddit for PythonSCAD:

https://old.reddit.com/r/pythonscad/

(and the newer releases have a mechanism for sharing design files)

William

Thanks William for your appreciation
Regards

On Tue, 3 Sept, 2024, 7:52 am William F. Adams via Discuss, <
discuss@lists.openscad.org> wrote:

Hi Adrian,

Whenever I try to write documentation, I find it too hard and very soon
give up.

But I am sure that this approach of defining objects in python works and
can create very complex objects and almost anything.

Someone like you can do far better than me in making this useful for
others.

Regards

On Tue, 3 Sept, 2024, 7:42 am Adrian Mariano via Discuss, <
discuss@lists.openscad.org> wrote:

On 9/2/24 21:55, Sanjeev Prabhakar via Discuss wrote:

Don't apologize for your English skills Sanjeev, they are better than
mine. But you likely have quite a few more years learning it than I. I
quit school shortly after the first 8 grades, and went to work fixing
the new fangled things called a tv's, in 1948. Eventually switched to
broadcasting, I kept a series of tv stations on the air from '63, to the
middle of 2002 as the CE & often the only tech in the building. I enjoy
someone else thinking outside that famous box they try to put us in.  So
please do continue.

OpenSCAD, FWIW, has leaks. What it has taught me about a list of
polygons was leaked over into linuxcnc which I also have used for the
last 25+ years since before I retired. Thank you /all/ for the
inspiration and the ability to be productive as I approach my 90th.
Pretty good for a now old & diabetic Iowa farm kid.

Cheers, Gene Heskett, CET.

"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.

• Louis D. Brandeis

As an side -- generally, it isn't useful to check computer-generated
data files into git. It's equivalent to checking assembler or object
files into git. It bloats a repo while serving no purpose since they
don't need version control. Instead, provide a Makefile or script that
generates the output files on demand. Check in only the source code (in
this case the Python) and the script.

On 9/2/24 09:40, Sanjeev Prabhakar via Discuss wrote: