OpenSCAD release candidate: 2020.12-RC3
Hi!
Release candidate 3 is now available.
Fixes relative to RC2:
- Handle case of flattening during iteration over VectorType. (#3541)
- Fix log output to separate preview/render executions. (#3533)
- Handle SCN_URIDROPPED introduced in QScintilla 2.11. (#3536)
- Add window menu to allow keyboard access to docked windows. (#3509)
- Calling tabifyDockWidget breaks window state save with Qt5.15/Linux.
- Handle empty first geometry in 2D Minkowski. (#3527)
- Abort Nef conversion early on invalid input, to avoid segfault. (#2847)
- Disable context menu editor switch if editor hidden. (#3519)
- Ensure experimental features are disabled in release. (#3517)
Big thanks go to everyone helping with testing and fixing those
issues. There's some more issues to check, so I expect at least
one more RC.
Download (Windows & Linux only for now):
https://www.openscad.org/downloads.html#rc
ciao,
Torsten.
The first time I ran this I got a heap corruption crash but I haven't been
able to reproduce it. I did F5 followed by F6 on a shape with intersecting
cylinders and it gave a CGAL error due to the grid snap. So I cleared the
cache to do another F6, which normally works. I think perhaps I managed to
clear the cache before the first F6 had finished because it doesn't stop on
the CGAL error. However I can't do it again, so it might be a race
condition.
[image: image.png]
This is F6 after F5, with the offending object missing.
[image: image.png]
The is flush cache followed by F6. It doesn't give the CGAL error but the
STL has self intersections.
Cylinders of different diameters meeting at an angle seems to be impossible
in OpenSCAD because it is hard to avoid close but not touching vertices
somewhere around the intersection.
[image: image.png]
On Fri, 18 Dec 2020 at 22:15, Torsten Paul Torsten.Paul@gmx.de wrote:
Hi!
Release candidate 3 is now available.
Fixes relative to RC2:
- Handle case of flattening during iteration over VectorType. (#3541)
- Fix log output to separate preview/render executions. (#3533)
- Handle SCN_URIDROPPED introduced in QScintilla 2.11. (#3536)
- Add window menu to allow keyboard access to docked windows. (#3509)
- Calling tabifyDockWidget breaks window state save with Qt5.15/Linux.
- Handle empty first geometry in 2D Minkowski. (#3527)
- Abort Nef conversion early on invalid input, to avoid segfault. (#2847)
- Disable context menu editor switch if editor hidden. (#3519)
- Ensure experimental features are disabled in release. (#3517)
Big thanks go to everyone helping with testing and fixing those
issues. There's some more issues to check, so I expect at least
one more RC.
Download (Windows & Linux only for now):
https://www.openscad.org/downloads.html#rc
ciao,
Torsten.
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Also RC3 still has the mouse click popup menu, even with the editor hidden.
On Sat, 19 Dec 2020 at 10:55, nop head nop.head@gmail.com wrote:
The first time I ran this I got a heap corruption crash but I haven't been
able to reproduce it. I did F5 followed by F6 on a shape with intersecting
cylinders and it gave a CGAL error due to the grid snap. So I cleared the
cache to do another F6, which normally works. I think perhaps I managed to
clear the cache before the first F6 had finished because it doesn't stop on
the CGAL error. However I can't do it again, so it might be a race
condition.
[image: image.png]
This is F6 after F5, with the offending object missing.
[image: image.png]
The is flush cache followed by F6. It doesn't give the CGAL error but the
STL has self intersections.
Cylinders of different diameters meeting at an angle seems to be
impossible in OpenSCAD because it is hard to avoid close but not touching
vertices somewhere around the intersection.
[image: image.png]
On Fri, 18 Dec 2020 at 22:15, Torsten Paul Torsten.Paul@gmx.de wrote:
Hi!
Release candidate 3 is now available.
Fixes relative to RC2:
- Handle case of flattening during iteration over VectorType. (#3541)
- Fix log output to separate preview/render executions. (#3533)
- Handle SCN_URIDROPPED introduced in QScintilla 2.11. (#3536)
- Add window menu to allow keyboard access to docked windows. (#3509)
- Calling tabifyDockWidget breaks window state save with Qt5.15/Linux.
- Handle empty first geometry in 2D Minkowski. (#3527)
- Abort Nef conversion early on invalid input, to avoid segfault. (#2847)
- Disable context menu editor switch if editor hidden. (#3519)
- Ensure experimental features are disabled in release. (#3517)
Big thanks go to everyone helping with testing and fixing those
issues. There's some more issues to check, so I expect at least
one more RC.
Download (Windows & Linux only for now):
https://www.openscad.org/downloads.html#rc
ciao,
Torsten.
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
I tried out this example posted on another thread in RC3:
https://github.com/rugyoga/catalan_solids/blob/main/disdyakis_triacontahedron.scad
I adjusted the code to call full_skeleton() instead of flattened_diamond()
and the model previews very quickly, but when I tried to render the
resulting model OpenSCAD ran for a long time (half an hour?) and then
crashed. The model is a bad model (It has 20 spheres in the same place
multiple times, and every cylinder has two copies in the same place). I
made similar model without any duplicated objects and it rendered in 24
minutes. But even the bad model presumably shouldn't crash OpenSCAD.
--
Sent from: http://forum.openscad.org/
For amusement value, try the following code in RC3, then try removing any of the superfluous translates, mirrors, or arguments.
Note that 2019.05 handles this correctly, but RC3 doesn’t. This was trimmed from a CSG tree of a more complex object that exposed this bug.
hull() {
translate([-4,-1,1]) cube(2, center=true);
translate([3,0,0]) {
intersection() {
rotate([90,0,0]) {
linear_extrude(height=4, center=true) {
circle(r=2, $fn=4);
}
}
translate([0,0,0]) {
translate([0,0,0]) {
mirror([0,0,1]) {
translate([0,0,0]) {
linear_extrude(height=4, center=true, convexity=2, $fn=4) {
circle(r=2, $fn=4);
}
}
}
}
}
}
}
}
- Revar
On Dec 18, 2020, at 2:14 PM, Torsten Paul Torsten.Paul@gmx.de wrote:
Hi!
Release candidate 3 is now available.
Fixes relative to RC2:
- Handle case of flattening during iteration over VectorType. (#3541)
- Fix log output to separate preview/render executions. (#3533)
- Handle SCN_URIDROPPED introduced in QScintilla 2.11. (#3536)
- Add window menu to allow keyboard access to docked windows. (#3509)
- Calling tabifyDockWidget breaks window state save with Qt5.15/Linux.
- Handle empty first geometry in 2D Minkowski. (#3527)
- Abort Nef conversion early on invalid input, to avoid segfault. (#2847)
- Disable context menu editor switch if editor hidden. (#3519)
- Ensure experimental features are disabled in release. (#3517)
Big thanks go to everyone helping with testing and fixing those
issues. There's some more issues to check, so I expect at least
one more RC.
Download (Windows & Linux only for now):
https://www.openscad.org/downloads.html#rc
ciao,
Torsten.
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Have I inadvertently enabled something that causes my preview to look like
this? I have never had the lines before.
$fn=30;
cylinder(h=10,d=3);
http://forum.openscad.org/file/t2121/Lines.png
--
Sent from: http://forum.openscad.org/
lar3ry wrote
Have I inadvertently enabled something that causes my preview to look like
this? I have never had the lines before.
View/Show-edges?
OpenSCAD Admin - email* me if you need anything, or if I've done something stupid...
- on the Forum, click on my MichaelAtOz label, there is a link to email me.
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. Obviously inclusion of works of previous authors is not included in the above.
--
Sent from: http://forum.openscad.org/
RevarBat wrote
For amusement value, try the following code in RC3, then try removing any
of the superfluous translates, mirrors, or arguments.
Doesn't do anything differently for me, what happens?
OpenSCAD Admin - email* me if you need anything, or if I've done something stupid...
- on the Forum, click on my MichaelAtOz label, there is a link to email me.
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. Obviously inclusion of works of previous authors is not included in the above.
--
Sent from: http://forum.openscad.org/
MichaelAtOz wrote
lar3ry wrote
Have I inadvertently enabled something that causes my preview to look
like
this? I have never had the lines before.
View/Show-edges?
Thanks. That was it.
--
Sent from: http://forum.openscad.org/