HW
Harvey white
Fri, Jan 23, 2026 5:08 PM
I'm looking for the latest, but most stable, version of OpenSCAD. Going
through the snapshots, it looks as if 2026.01.19 is it.
Do I have it right? The website says that the latest stable version
2021.01, which I have.
Thanks.
Harvey
I'm looking for the latest, but most stable, version of OpenSCAD. Going
through the snapshots, it looks as if 2026.01.19 is it.
Do I have it right? The website says that the latest stable version
2021.01, which I have.
Thanks.
Harvey
JB
Jordan Brown
Fri, Jan 23, 2026 5:16 PM
I'm looking for the latest, but most stable, version of OpenSCAD. Going through the snapshots, it looks as if 2026.01.19 is it.
Do I have it right? The website says that the latest stable version 2021.01, which I have.
My recent experience is that any nightly build (like 2026.01.19) is better in pretty much every metric than 2021.01. I think that 2021.01 has suffered some bit rot; I find that it never exits properly.
Some will complain about a couple of niche incompatible changes, but those are not instability per se.
Make sure that you have Manifold selected as the rendering engine in Preferences/Advanced; it is far faster than CGAL.
> I'm looking for the latest, but most stable, version of OpenSCAD. Going through the snapshots, it looks as if 2026.01.19 is it.
>
> Do I have it right? The website says that the latest stable version 2021.01, which I have.
My recent experience is that any nightly build (like 2026.01.19) is better in pretty much every metric than 2021.01. I think that 2021.01 has suffered some bit rot; I find that it never exits properly.
Some will complain about a couple of niche incompatible changes, but those are not instability per se.
Make sure that you have Manifold selected as the rendering engine in Preferences/Advanced; it is far faster than CGAL.
NH
nop head
Fri, Jan 23, 2026 5:22 PM
2021.01 still works fine for me on two Win11 PCs.
On Fri, 23 Jan 2026 at 17:17, Jordan Brown via Discuss <
discuss@lists.openscad.org> wrote:
I'm looking for the latest, but most stable, version of OpenSCAD.
Going through the snapshots, it looks as if 2026.01.19 is it.
Do I have it right? The website says that the latest stable version
2021.01, which I have.
My recent experience is that any nightly build (like 2026.01.19) is better
in pretty much every metric than 2021.01. I think that 2021.01 has suffered
some bit rot; I find that it never exits properly.
Some will complain about a couple of niche incompatible changes, but those
are not instability per se.
Make sure that you have Manifold selected as the rendering engine in
Preferences/Advanced; it is far faster than CGAL.
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
2021.01 still works fine for me on two Win11 PCs.
On Fri, 23 Jan 2026 at 17:17, Jordan Brown via Discuss <
discuss@lists.openscad.org> wrote:
>
>
> > I'm looking for the latest, but most stable, version of OpenSCAD.
> Going through the snapshots, it looks as if 2026.01.19 is it.
> >
> > Do I have it right? The website says that the latest stable version
> 2021.01, which I have.
>
> My recent experience is that any nightly build (like 2026.01.19) is better
> in pretty much every metric than 2021.01. I think that 2021.01 has suffered
> some bit rot; I find that it never exits properly.
>
> Some will complain about a couple of niche incompatible changes, but those
> are not instability per se.
>
> Make sure that you have Manifold selected as the rendering engine in
> Preferences/Advanced; it is far faster than CGAL.
> _______________________________________________
> OpenSCAD mailing list
> To unsubscribe send an email to discuss-leave@lists.openscad.org
HW
Harvey white
Fri, Jan 23, 2026 7:44 PM
I have a relatively complex model (for me) which is a pick and place
frame with controllable positions. I need to model various parts that
are 3D printed, and also see how things fit. It takes a bit of time.
I'm looking for a speed increase, and I may find enhancements that are
useful. I think I'll try 2026.01.19 to see what it does.
Thanks.
Harvey
On 1/23/2026 12:22 PM, nop head via Discuss wrote:
2021.01 still works fine for me on two Win11 PCs.
On Fri, 23 Jan 2026 at 17:17, Jordan Brown via Discuss
discuss@lists.openscad.org wrote:
I'm looking for the latest, but most stable, version of
OpenSCAD. Going through the snapshots, it looks as if 2026.01.19
is it.
Do I have it right? The website says that the latest stable
version 2021.01, which I have.
My recent experience is that any nightly build (like 2026.01.19)
is better in pretty much every metric than 2021.01. I think that
2021.01 has suffered some bit rot; I find that it never exits
properly.
Some will complain about a couple of niche incompatible changes,
but those are not instability per se.
Make sure that you have Manifold selected as the rendering engine
in Preferences/Advanced; it is far faster than CGAL.
_______________________________________________
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
I have a relatively complex model (for me) which is a pick and place
frame with controllable positions. I need to model various parts that
are 3D printed, and also see how things fit. It takes a bit of time.
I'm looking for a speed increase, and I may find enhancements that are
useful. I think I'll try 2026.01.19 to see what it does.
Thanks.
Harvey
On 1/23/2026 12:22 PM, nop head via Discuss wrote:
> 2021.01 still works fine for me on two Win11 PCs.
>
> On Fri, 23 Jan 2026 at 17:17, Jordan Brown via Discuss
> <discuss@lists.openscad.org> wrote:
>
>
>
> > I'm looking for the latest, but most stable, version of
> OpenSCAD. Going through the snapshots, it looks as if 2026.01.19
> is it.
> >
> > Do I have it right? The website says that the latest stable
> version 2021.01, which I have.
>
> My recent experience is that any nightly build (like 2026.01.19)
> is better in pretty much every metric than 2021.01. I think that
> 2021.01 has suffered some bit rot; I find that it never exits
> properly.
>
> Some will complain about a couple of niche incompatible changes,
> but those are not instability per se.
>
> Make sure that you have Manifold selected as the rendering engine
> in Preferences/Advanced; it is far faster than CGAL.
> _______________________________________________
> OpenSCAD mailing list
> To unsubscribe send an email to discuss-leave@lists.openscad.org
>
>
> _______________________________________________
> OpenSCAD mailing list
> To unsubscribe send an email to discuss-leave@lists.openscad.org
HW
Harvey white
Fri, Jan 23, 2026 8:05 PM
Now that was fun:
lots of errors with deprecated variables starting with numbers. Most of
that seems to come from the nopSCADlib. I seem to remember that it was
a subject of discussion a bit ago. Was there a solution?
Also
When I get an error such as:
WARNING: Can't find include file 'customSCADlib/lib.scad'. in file , line 15
I suppose I should ask "what file" line 15.
Harvey
On 1/23/2026 2:44 PM, Harvey white via Discuss wrote:
I have a relatively complex model (for me) which is a pick and place
frame with controllable positions. I need to model various parts that
are 3D printed, and also see how things fit. It takes a bit of time.
I'm looking for a speed increase, and I may find enhancements that are
useful. I think I'll try 2026.01.19 to see what it does.
Thanks.
Harvey
On 1/23/2026 12:22 PM, nop head via Discuss wrote:
2021.01 still works fine for me on two Win11 PCs.
On Fri, 23 Jan 2026 at 17:17, Jordan Brown via Discuss
discuss@lists.openscad.org wrote:
> I'm looking for the latest, but most stable, version of
OpenSCAD. Going through the snapshots, it looks as if 2026.01.19
is it.
>
> Do I have it right? The website says that the latest stable
version 2021.01, which I have.
My recent experience is that any nightly build (like 2026.01.19)
is better in pretty much every metric than 2021.01. I think that
2021.01 has suffered some bit rot; I find that it never exits
properly.
Some will complain about a couple of niche incompatible changes,
but those are not instability per se.
Make sure that you have Manifold selected as the rendering engine
in Preferences/Advanced; it is far faster than CGAL.
_______________________________________________
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
Now that was fun:
lots of errors with deprecated variables starting with numbers. Most of
that seems to come from the nopSCADlib. I seem to remember that it was
a subject of discussion a bit ago. Was there a solution?
Also
When I get an error such as:
WARNING: Can't find include file 'customSCADlib/lib.scad'. in file , line 15
I suppose I should ask "what file" line 15.
Harvey
On 1/23/2026 2:44 PM, Harvey white via Discuss wrote:
> I have a relatively complex model (for me) which is a pick and place
> frame with controllable positions. I need to model various parts that
> are 3D printed, and also see how things fit. It takes a bit of time.
> I'm looking for a speed increase, and I may find enhancements that are
> useful. I think I'll try 2026.01.19 to see what it does.
>
> Thanks.
>
> Harvey
>
>
> On 1/23/2026 12:22 PM, nop head via Discuss wrote:
>> 2021.01 still works fine for me on two Win11 PCs.
>>
>> On Fri, 23 Jan 2026 at 17:17, Jordan Brown via Discuss
>> <discuss@lists.openscad.org> wrote:
>>
>>
>>
>> > I'm looking for the latest, but most stable, version of
>> OpenSCAD. Going through the snapshots, it looks as if 2026.01.19
>> is it.
>> >
>> > Do I have it right? The website says that the latest stable
>> version 2021.01, which I have.
>>
>> My recent experience is that any nightly build (like 2026.01.19)
>> is better in pretty much every metric than 2021.01. I think that
>> 2021.01 has suffered some bit rot; I find that it never exits
>> properly.
>>
>> Some will complain about a couple of niche incompatible changes,
>> but those are not instability per se.
>>
>> Make sure that you have Manifold selected as the rendering engine
>> in Preferences/Advanced; it is far faster than CGAL.
>> _______________________________________________
>> OpenSCAD mailing list
>> To unsubscribe send an email to discuss-leave@lists.openscad.org
>>
>>
>> _______________________________________________
>> OpenSCAD mailing list
>> To unsubscribe send an email to discuss-leave@lists.openscad.org
> _______________________________________________
> OpenSCAD mailing list
> To unsubscribe send an email to discuss-leave@lists.openscad.org
NH
nop head
Fri, Jan 23, 2026 8:09 PM
My solution was to stick to an older snapshot in the GUI and the last
release for the command line to get correct STLs.
On Fri, 23 Jan 2026 at 20:06, Harvey white via Discuss <
discuss@lists.openscad.org> wrote:
Now that was fun:
lots of errors with deprecated variables starting with numbers. Most of
that seems to come from the nopSCADlib. I seem to remember that it was
a subject of discussion a bit ago. Was there a solution?
Also
When I get an error such as:
WARNING: Can't find include file 'customSCADlib/lib.scad'. in file , line
15
I suppose I should ask "what file" line 15.
Harvey
On 1/23/2026 2:44 PM, Harvey white via Discuss wrote:
I have a relatively complex model (for me) which is a pick and place
frame with controllable positions. I need to model various parts that
are 3D printed, and also see how things fit. It takes a bit of time.
I'm looking for a speed increase, and I may find enhancements that are
useful. I think I'll try 2026.01.19 to see what it does.
Thanks.
Harvey
On 1/23/2026 12:22 PM, nop head via Discuss wrote:
2021.01 still works fine for me on two Win11 PCs.
On Fri, 23 Jan 2026 at 17:17, Jordan Brown via Discuss
discuss@lists.openscad.org wrote:
I'm looking for the latest, but most stable, version of
OpenSCAD. Going through the snapshots, it looks as if 2026.01.19
is it.
Do I have it right? The website says that the latest stable
version 2021.01, which I have.
My recent experience is that any nightly build (like 2026.01.19)
is better in pretty much every metric than 2021.01. I think that
2021.01 has suffered some bit rot; I find that it never exits
properly.
Some will complain about a couple of niche incompatible changes,
but those are not instability per se.
Make sure that you have Manifold selected as the rendering engine
in Preferences/Advanced; it is far faster than CGAL.
_______________________________________________
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
My solution was to stick to an older snapshot in the GUI and the last
release for the command line to get correct STLs.
On Fri, 23 Jan 2026 at 20:06, Harvey white via Discuss <
discuss@lists.openscad.org> wrote:
> Now that was fun:
>
> lots of errors with deprecated variables starting with numbers. Most of
> that seems to come from the nopSCADlib. I seem to remember that it was
> a subject of discussion a bit ago. Was there a solution?
>
> Also
>
> When I get an error such as:
>
> WARNING: Can't find include file 'customSCADlib/lib.scad'. in file , line
> 15
>
> I suppose I should ask "what file" line 15.
>
> Harvey
>
>
>
> On 1/23/2026 2:44 PM, Harvey white via Discuss wrote:
> > I have a relatively complex model (for me) which is a pick and place
> > frame with controllable positions. I need to model various parts that
> > are 3D printed, and also see how things fit. It takes a bit of time.
> > I'm looking for a speed increase, and I may find enhancements that are
> > useful. I think I'll try 2026.01.19 to see what it does.
> >
> > Thanks.
> >
> > Harvey
> >
> >
> > On 1/23/2026 12:22 PM, nop head via Discuss wrote:
> >> 2021.01 still works fine for me on two Win11 PCs.
> >>
> >> On Fri, 23 Jan 2026 at 17:17, Jordan Brown via Discuss
> >> <discuss@lists.openscad.org> wrote:
> >>
> >>
> >>
> >> > I'm looking for the latest, but most stable, version of
> >> OpenSCAD. Going through the snapshots, it looks as if 2026.01.19
> >> is it.
> >> >
> >> > Do I have it right? The website says that the latest stable
> >> version 2021.01, which I have.
> >>
> >> My recent experience is that any nightly build (like 2026.01.19)
> >> is better in pretty much every metric than 2021.01. I think that
> >> 2021.01 has suffered some bit rot; I find that it never exits
> >> properly.
> >>
> >> Some will complain about a couple of niche incompatible changes,
> >> but those are not instability per se.
> >>
> >> Make sure that you have Manifold selected as the rendering engine
> >> in Preferences/Advanced; it is far faster than CGAL.
> >> _______________________________________________
> >> OpenSCAD mailing list
> >> To unsubscribe send an email to discuss-leave@lists.openscad.org
> >>
> >>
> >> _______________________________________________
> >> OpenSCAD mailing list
> >> To unsubscribe send an email to discuss-leave@lists.openscad.org
> > _______________________________________________
> > OpenSCAD mailing list
> > To unsubscribe send an email to discuss-leave@lists.openscad.org
> _______________________________________________
> OpenSCAD mailing list
> To unsubscribe send an email to discuss-leave@lists.openscad.org
HW
Harvey white
Fri, Jan 23, 2026 8:26 PM
Oh, I'm not using command line interfaces. Will there be a more
permanent solution in the library or might variables starting with
numbers be permitted?
Harvey
On 1/23/2026 3:09 PM, nop head via Discuss wrote:
My solution was to stick to an older snapshot in the GUI and the last
release for the command line to get correct STLs.
On Fri, 23 Jan 2026 at 20:06, Harvey white via Discuss
discuss@lists.openscad.org wrote:
Now that was fun:
lots of errors with deprecated variables starting with numbers.
Most of
that seems to come from the nopSCADlib. I seem to remember that
it was
a subject of discussion a bit ago. Was there a solution?
Also
When I get an error such as:
WARNING: Can't find include file 'customSCADlib/lib.scad'. in file
, line 15
I suppose I should ask "what file" line 15.
Harvey
On 1/23/2026 2:44 PM, Harvey white via Discuss wrote:
I have a relatively complex model (for me) which is a pick and
frame with controllable positions. I need to model various
are 3D printed, and also see how things fit. It takes a bit of
I'm looking for a speed increase, and I may find enhancements
useful. I think I'll try 2026.01.19 to see what it does.
Thanks.
Harvey
On 1/23/2026 12:22 PM, nop head via Discuss wrote:
2021.01 still works fine for me on two Win11 PCs.
On Fri, 23 Jan 2026 at 17:17, Jordan Brown via Discuss
discuss@lists.openscad.org wrote:
> I'm looking for the latest, but most stable, version of
OpenSCAD. Going through the snapshots, it looks as if
is it.
>
> Do I have it right? The website says that the latest stable
version 2021.01, which I have.
My recent experience is that any nightly build (like
is better in pretty much every metric than 2021.01. I think
2021.01 has suffered some bit rot; I find that it never exits
properly.
Some will complain about a couple of niche incompatible
but those are not instability per se.
Make sure that you have Manifold selected as the rendering
in Preferences/Advanced; it is far faster than CGAL.
_______________________________________________
OpenSCAD mailing list
To unsubscribe send an email to
discuss-leave@lists.openscad.org
_______________________________________________
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
Oh, I'm not using command line interfaces. Will there be a more
permanent solution in the library or might variables starting with
numbers be permitted?
Harvey
On 1/23/2026 3:09 PM, nop head via Discuss wrote:
> My solution was to stick to an older snapshot in the GUI and the last
> release for the command line to get correct STLs.
>
> On Fri, 23 Jan 2026 at 20:06, Harvey white via Discuss
> <discuss@lists.openscad.org> wrote:
>
> Now that was fun:
>
> lots of errors with deprecated variables starting with numbers.
> Most of
> that seems to come from the nopSCADlib. I seem to remember that
> it was
> a subject of discussion a bit ago. Was there a solution?
>
> Also
>
> When I get an error such as:
>
> WARNING: Can't find include file 'customSCADlib/lib.scad'. in file
> , line 15
>
> I suppose I should ask "what file" line 15.
>
> Harvey
>
>
>
> On 1/23/2026 2:44 PM, Harvey white via Discuss wrote:
> > I have a relatively complex model (for me) which is a pick and
> place
> > frame with controllable positions. I need to model various
> parts that
> > are 3D printed, and also see how things fit. It takes a bit of
> time.
> > I'm looking for a speed increase, and I may find enhancements
> that are
> > useful. I think I'll try 2026.01.19 to see what it does.
> >
> > Thanks.
> >
> > Harvey
> >
> >
> > On 1/23/2026 12:22 PM, nop head via Discuss wrote:
> >> 2021.01 still works fine for me on two Win11 PCs.
> >>
> >> On Fri, 23 Jan 2026 at 17:17, Jordan Brown via Discuss
> >> <discuss@lists.openscad.org> wrote:
> >>
> >>
> >>
> >> > I'm looking for the latest, but most stable, version of
> >> OpenSCAD. Going through the snapshots, it looks as if
> 2026.01.19
> >> is it.
> >> >
> >> > Do I have it right? The website says that the latest stable
> >> version 2021.01, which I have.
> >>
> >> My recent experience is that any nightly build (like
> 2026.01.19)
> >> is better in pretty much every metric than 2021.01. I think
> that
> >> 2021.01 has suffered some bit rot; I find that it never exits
> >> properly.
> >>
> >> Some will complain about a couple of niche incompatible
> changes,
> >> but those are not instability per se.
> >>
> >> Make sure that you have Manifold selected as the rendering
> engine
> >> in Preferences/Advanced; it is far faster than CGAL.
> >> _______________________________________________
> >> OpenSCAD mailing list
> >> To unsubscribe send an email to
> discuss-leave@lists.openscad.org
> >>
> >>
> >> _______________________________________________
> >> OpenSCAD mailing list
> >> To unsubscribe send an email to discuss-leave@lists.openscad.org
> > _______________________________________________
> > OpenSCAD mailing list
> > To unsubscribe send an email to discuss-leave@lists.openscad.org
> _______________________________________________
> OpenSCAD mailing list
> To unsubscribe send an email to discuss-leave@lists.openscad.org
>
>
> _______________________________________________
> OpenSCAD mailing list
> To unsubscribe send an email to discuss-leave@lists.openscad.org
JB
Jordan Brown
Fri, Jan 23, 2026 10:36 PM
On 1/23/2026 12:05 PM, Harvey white via Discuss wrote:
lots of errors with deprecated variables starting with numbers. Most
of that seems to come from the nopSCADlib. I seem to remember that it
was a subject of discussion a bit ago. Was there a solution?
See my note about some complaining about niche incompatible changes.
Right now the answer is either:
a) Ignore the warnings and risk that future changes might break those
names, or
b) Change the names.
Note: in the general case, they never worked, because 1e2 was always
a number, not an identifier. When I added hexadecimal-constant 0xHHHH
support and so made 0x<anything> not work either, the advice was to warn
that any such identifiers were on thin ice.
If you're using nopSCADlib you might run into other problems with $
variables... see again the note about niche incompatible changes.
WARNING: Can't find include file 'customSCADlib/lib.scad'. in file ,
line 15
I suppose I should ask "what file" line 15.
Usually that means an unnamed unsaved file - the file in front of you in
the editor.
On 1/23/2026 12:05 PM, Harvey white via Discuss wrote:
> lots of errors with deprecated variables starting with numbers. Most
> of that seems to come from the nopSCADlib. I seem to remember that it
> was a subject of discussion a bit ago. Was there a solution?
See my note about some complaining about niche incompatible changes.
Right now the answer is either:
a) Ignore the warnings and risk that future changes might break those
names, or
b) Change the names.
Note: in the general case, they *never* worked, because 1e2 was always
a number, not an identifier. When I added hexadecimal-constant 0xHHHH
support and so made 0x<anything> not work either, the advice was to warn
that any such identifiers were on thin ice.
If you're using nopSCADlib you might run into other problems with $
variables... see again the note about niche incompatible changes.
> WARNING: Can't find include file 'customSCADlib/lib.scad'. in file ,
> line 15
>
> I suppose I should ask "what file" line 15.
Usually that means an unnamed unsaved file - the file in front of you in
the editor.
NH
nop head
Fri, Jan 23, 2026 11:40 PM
I haven't got any identifiers that look like hex constants. All identifiers
that start with numbers now warn.
On Fri, 23 Jan 2026, 22:37 Jordan Brown via Discuss, <
discuss@lists.openscad.org> wrote:
On 1/23/2026 12:05 PM, Harvey white via Discuss wrote:
lots of errors with deprecated variables starting with numbers. Most of
that seems to come from the nopSCADlib. I seem to remember that it was a
subject of discussion a bit ago. Was there a solution?
See my note about some complaining about niche incompatible changes.
Right now the answer is either:
a) Ignore the warnings and risk that future changes might break those
names, or
b) Change the names.
Note: in the general case, they never worked, because 1e2 was always a
number, not an identifier. When I added hexadecimal-constant 0xHHHH
support and so made 0x<anything> not work either, the advice was to warn
that any such identifiers were on thin ice.
If you're using nopSCADlib you might run into other problems with $
variables... see again the note about niche incompatible changes.
WARNING: Can't find include file 'customSCADlib/lib.scad'. in file , line
15
I suppose I should ask "what file" line 15.
Usually that means an unnamed unsaved file - the file in front of you in
the editor.
OpenSCAD mailing list
To unsubscribe send an email to discuss-leave@lists.openscad.org
I haven't got any identifiers that look like hex constants. All identifiers
that start with numbers now warn.
On Fri, 23 Jan 2026, 22:37 Jordan Brown via Discuss, <
discuss@lists.openscad.org> wrote:
> On 1/23/2026 12:05 PM, Harvey white via Discuss wrote:
>
> lots of errors with deprecated variables starting with numbers. Most of
> that seems to come from the nopSCADlib. I seem to remember that it was a
> subject of discussion a bit ago. Was there a solution?
>
>
> See my note about some complaining about niche incompatible changes.
>
> Right now the answer is either:
> a) Ignore the warnings and risk that future changes might break those
> names, or
> b) Change the names.
>
> Note: in the general case, they *never* worked, because 1e2 was always a
> number, not an identifier. When I added hexadecimal-constant 0xHHHH
> support and so made 0x<anything> not work either, the advice was to warn
> that any such identifiers were on thin ice.
>
> If you're using nopSCADlib you might run into other problems with $
> variables... see again the note about niche incompatible changes.
>
> WARNING: Can't find include file 'customSCADlib/lib.scad'. in file , line
> 15
>
> I suppose I should ask "what file" line 15.
>
>
> Usually that means an unnamed unsaved file - the file in front of you in
> the editor.
> _______________________________________________
> OpenSCAD mailing list
> To unsubscribe send an email to discuss-leave@lists.openscad.org
HW
Harvey white
Fri, Jan 23, 2026 11:41 PM
interleaved:
On 1/23/2026 5:36 PM, Jordan Brown via Discuss wrote:
On 1/23/2026 12:05 PM, Harvey white via Discuss wrote:
lots of errors with deprecated variables starting with numbers. Most
of that seems to come from the nopSCADlib. I seem to remember that
it was a subject of discussion a bit ago. Was there a solution?
See my note about some complaining about niche incompatible changes.
Right now the answer is either:
a) Ignore the warnings and risk that future changes might break those
names, or
b) Change the names.
a) in my experience, ignoring that kind of thing will eventually bite
you... (or is that byte?)
b) I have one or two things in my own library that I can cheerfully
change. Since these are not in my library, I could go through the
exercise of changing them all. I am loath to do that because it will
force me to do the same thing every time the evolving library changes.
I suspect that I'll have to back down to a version that does not have
code about variable names. Likely it's back to 2021.
Note: in the general case, they never worked, because 1e2 was
always a number, not an identifier. When I added hexadecimal-constant
0xHHHH support and so made 0x<anything> not work either, the advice
was to warn that any such identifiers were on thin ice.
If you're using nopSCADlib you might run into other problems with $
variables... see again the note about niche incompatible changes.
yeah, I can imagine that. I've had similar problems in the C++
environment, not this problem, but something involving a similar amount
of bother. The C++ was a structural change, though.
WARNING: Can't find include file 'customSCADlib/lib.scad'. in file ,
line 15
I suppose I should ask "what file" line 15.
Usually that means an unnamed unsaved file - the file in front of you
in the editor.
Ah, ok, classic use of "." in a path, gotcha
Thanks,
Harvey
interleaved:
On 1/23/2026 5:36 PM, Jordan Brown via Discuss wrote:
> On 1/23/2026 12:05 PM, Harvey white via Discuss wrote:
>> lots of errors with deprecated variables starting with numbers. Most
>> of that seems to come from the nopSCADlib. I seem to remember that
>> it was a subject of discussion a bit ago. Was there a solution?
>
> See my note about some complaining about niche incompatible changes.
>
> Right now the answer is either:
> a) Ignore the warnings and risk that future changes might break those
> names, or
> b) Change the names.
a) in my experience, ignoring that kind of thing will eventually bite
you... (or is that byte?)
b) I have one or two things in my own library that I can cheerfully
change. Since these are *not* in my library, I could go through the
exercise of changing them all. I am loath to do that because it will
force me to do the same thing every time the evolving library changes.
I suspect that I'll have to back down to a version that does not have
code about variable names. Likely it's back to 2021.
>
> Note: in the general case, they *never* worked, because 1e2 was
> always a number, not an identifier. When I added hexadecimal-constant
> 0xHHHH support and so made 0x<anything> not work either, the advice
> was to warn that any such identifiers were on thin ice.
>
> If you're using nopSCADlib you might run into other problems with $
> variables... see again the note about niche incompatible changes.
yeah, I can imagine that. I've had similar problems in the C++
environment, not this problem, but something involving a similar amount
of bother. The C++ was a structural change, though.
>
>> WARNING: Can't find include file 'customSCADlib/lib.scad'. in file ,
>> line 15
>>
>> I suppose I should ask "what file" line 15.
>
> Usually that means an unnamed unsaved file - the file in front of you
> in the editor.
Ah, ok, classic use of "." in a path, gotcha
Thanks,
Harvey
> _______________________________________________
> OpenSCAD mailing list
> To unsubscribe send an email to discuss-leave@lists.openscad.org
