0.20 bug/regression Array Pattern issue #6641

[user1234]

Hello!

There is a bug/regression, source: https://forum.freecadweb.org/viewtopic.php?t=67116

where: git commit 03da9d3501c6b596ad1828b99cb8d923cea56485

Open the attachment and recompute. The pattern (also on mirrors) will not performed correctly (pattern add). Note there is already a bug fixed, which set always the occurrences to 3, i do not mean this bug).

before:

Code: Select all

OS: Debian GNU/Linux 11 (bullseye) (X-Cinnamon/lightdm-xsession)
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.20.24415 (Git)
Build type: Unknown
Branch: (no branch, bisect started on master)
Hash: 89b8e81d5c3b11572a0288720d0b4145d2ec2a7e
Python version: 3.9.2
Qt version: 5.15.2
Coin version: 4.0.0
OCC version: 7.4.0
Locale: English/United States (en_US)

before.png (254.87 KiB) Viewed 2026 times

after:

Code: Select all

OS: Debian GNU/Linux 11 (bullseye) (X-Cinnamon/lightdm-xsession)
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.20.24416 (Git)
Build type: Unknown
Branch: (no branch, bisect started on master)
Hash: 03da9d3501c6b596ad1828b99cb8d923cea56485
Python version: 3.9.2
Qt version: 5.15.2
Coin version: 4.0.0
OCC version: 7.4.0
Locale: English/United States (en_US)

after.png (170.84 KiB) Viewed 2026 times

Greetings
user1234

[chrisb]

For testing it may be convenient to use this simplified version (linked from the other discussion). It is freed from all know coincidence stuff.



Kappe50_cb.FCStd

[user1234]

For testing it may be convenient to use this simplified version (linked from the other discussion).

It is not big and has also radius on it. I had a case, where it does sometimes work. So i think that is really a good/better sanity check test case.

Greetings
user1234

[adrianinsaval]

Interestingly, if you change the Overlap property to Overlap mode the second instance works but the third (and following if set to a higher number) doesn't. My example is based on the minimal one posted by chrisb but the same happens with the other

[uwestoehr]

Thanks for the report. Here is the issue: https://github.com/FreeCAD/FreeCAD/issues/6641

[user1234]

Thanks!

Greetings
user1234

[jnxd]

The pattern (also on mirrors) will not performed correctly (pattern add). Note there is already a bug fixed, which set always the occurrences to 3, i do not mean this bug).

Where is this fixed?

[user1234]

Note there is already a bug fixed, which set always the occurrences to 3, i do not mean this bug).

Where is this fixed?

When you mean the bug with the 3 occurrences? Here is the link to the bug + bugfixes (i am a little bit in a hurry, if you need something tested, i can a test it, but not asap).

https://forum.freecadweb.org/viewtopic. ... 1&start=10


Greetings
user1234

[jnxd]

Just looking at the behavior and the text of the commit:

- Use fuzzy fuse/cut method with tolerance Precision::Confusion()

This one change seems to be the culprit.

EDIT: I should just clarify it's purely my speculation.

[jnxd]

When you mean the bug with the 3 occurrences? Here is the link to the bug + bugfixes (i am a little bit in a hurry, if you need something tested, i can a test it, but not asap).

https://forum.freecadweb.org/viewtopic. ... 1&start=10

Thanks.