0.20 bug/regression Array Pattern issue #6641

About the development of the Part Design module/workbench. PLEASE DO NOT POST HELP REQUESTS HERE!
User avatar
uwestoehr
Veteran
Posts: 4251
Joined: Sun Jan 27, 2019 3:21 am
Location: Germany

Re: 0.20 bug/regression Array Pattern issue #6641

Post by uwestoehr »

jnxd wrote: Tue Apr 12, 2022 3:23 am [EDIT] Conveniently I just checked 15 seconds after posting this and the first file works with the following after a recompute:
It fails for me. This is what I get as result when I recompute the whole file:
FreeCAD_hfjDhan66W.png
FreeCAD_hfjDhan66W.png (35.93 KiB) Viewed 608 times

Ad as I wrote the issue is independent on polar patterns, I cannot do any transformations with lofts and pipes, no matter if patterning or mirroring. With FC 0.19 this works.

Code: Select all

OS: Windows 10 Version 2009
Word size of FreeCAD: 64-bit
Version: 0.20.28682 (Git)
Build type: Release
Branch: master
Hash: a13e251ad45c3562875e6bcc8e1c7e84882a4d52
Python 3.8.6+, Qt 5.15.2, Coin 4.0.1, OCC 7.5.3
Locale: English/United States (en_US) [ OS: German/Germany (de_DE) ]
jnxd
Posts: 577
Joined: Mon Mar 30, 2015 2:30 pm

Re: 0.20 bug/regression Array Pattern issue #6641

Post by jnxd »

uwestoehr wrote: Tue Apr 12, 2022 3:50 am It fails for me. This is what I get as result when I recompute the whole file: FreeCAD_hfjDhan66W.png

Code: Select all

OS: Windows 10 Version 2009
Word size of FreeCAD: 64-bit
Version: 0.20.28682 (Git)
Build type: Release
Branch: master
Hash: a13e251ad45c3562875e6bcc8e1c7e84882a4d52
Python 3.8.6+, Qt 5.15.2, Coin 4.0.1, OCC 7.5.3
Locale: English/United States (en_US) [ OS: German/Germany (de_DE) ]
Note that you are still using master. Please try https://github.com/wwmayer/FreeCAD/tree/issue_6641.
wmayer
Founder
Posts: 18785
Joined: Thu Feb 19, 2009 10:32 am

Re: 0.20 bug/regression Array Pattern issue #6641

Post by wmayer »

jnxd wrote: Tue Apr 12, 2022 3:23 am @wmayer are you waiting for something to merge this branch?
Well, I didn't want to merge it without other people having tested it. So, if it's OK I will remove the handling with the CompSolid and just always do the boolean operation at the very end.

I will work on it later today...
User avatar
uwestoehr
Veteran
Posts: 4251
Joined: Sun Jan 27, 2019 3:21 am
Location: Germany

Re: 0.20 bug/regression Array Pattern issue #6641

Post by uwestoehr »

jnxd wrote: Tue Apr 12, 2022 3:55 am Note that you are still using master. Please try https://github.com/wwmayer/FreeCAD/tree/issue_6641.
Ah, I wondered why GitHub desktop did not offer me this PR and first now I see it is not a PR in FC but in Werner's FC.
However, I cannot test today.

@Werner, for me it is easier to have a PR to FC directly. I check the pending PRs regularly and are this way I am informed and can test.
User avatar
uwestoehr
Veteran
Posts: 4251
Joined: Sun Jan 27, 2019 3:21 am
Location: Germany

Re: 0.20 bug/regression Array Pattern issue #6641

Post by uwestoehr »

wmayer wrote: Tue Apr 12, 2022 8:58 am Well, I didn't want to merge it without other people having tested it. So, if it's OK I will remove the handling with the CompSolid and just always do the boolean operation at the very end.
As I wrote yesterday, the regression is only in Loft and Sweep. These objects have been changed in 0.20 to support single vertices as section. This seems to cause the regression with the transformation.

In your PR I see you use CompSolid. Does this means only with a CompSolid, the single-vertex lofts can be handled?

---------------

I applied your PR now manually and will report back my testing results.
User avatar
uwestoehr
Veteran
Posts: 4251
Joined: Sun Jan 27, 2019 3:21 am
Location: Germany

Re: 0.20 bug/regression Array Pattern issue #6641

Post by uwestoehr »

uwestoehr wrote: Tue Apr 12, 2022 10:31 am I applied your PR now manually and will report back my testing results.
It works well - all these testcases work now: https://forum.freecadweb.org/viewtopic. ... 77#p587977

And also single-vertex loft sections work in transformations, here an example:
Kappe50-jeno-pattern-loft-vertex.FCStd
(619.82 KiB) Downloaded 11 times
wmayer
Founder
Posts: 18785
Joined: Thu Feb 19, 2009 10:32 am

Re: 0.20 bug/regression Array Pattern issue #6641

Post by wmayer »

In your PR I see you use CompSolid. Does this means only with a CompSolid, the single-vertex lofts can be handled?
No, I have copied the CompSolid stuff from jnxd's branch because using a plain compound doesn't work correctly. But as said above it's not needed and we can just do a single boolean operation at the end.
jnxd
Posts: 577
Joined: Mon Mar 30, 2015 2:30 pm

Re: 0.20 bug/regression Array Pattern issue #6641

Post by jnxd »

wmayer wrote: Tue Apr 12, 2022 8:58 am I will work on it later today...
I was planning on making a PR into your branch for supporting both fuse and cut tool shapes in the same pattern, similar to https://github.com/FreeCAD/FreeCAD/pull ... d8c66b3b4e. Can do that after you've made your changes.

[EDIT] Made the PR: https://github.com/wwmayer/FreeCAD/pull/4. It shouldn't be too difficult to adapt it to the overlay only version.
Last edited by jnxd on Tue Apr 12, 2022 2:34 pm, edited 1 time in total.
wmayer
Founder
Posts: 18785
Joined: Thu Feb 19, 2009 10:32 am

Re: 0.20 bug/regression Array Pattern issue #6641

Post by wmayer »

jnxd wrote: Tue Apr 12, 2022 12:52 pm
wmayer wrote: Tue Apr 12, 2022 8:58 am I will work on it later today...
I was planning on making a PR into your branch for supporting both fuse and cut tool shapes in the same pattern, similar to https://github.com/FreeCAD/FreeCAD/pull ... d8c66b3b4e. Can do that after you've made your changes.
git commit 5ab7d17f67
Now it's your turn.
jnxd
Posts: 577
Joined: Mon Mar 30, 2015 2:30 pm

Re: 0.20 bug/regression Array Pattern issue #6641

Post by jnxd »

wmayer wrote: Tue Apr 12, 2022 2:34 pm git commit 5ab7d17f67
Now it's your turn.
:lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol:

You pushed almost the same time I made a PR.
Post Reply