Sketcher: Can't join two arcs

[dancesWithMachines]

Is there an existing issue for this?

• I have searched the existing issues

Problem description

Scenario:

1. Create one half of the arc and constrain it.
2. Add second arc using same point as center and join the "arms".

Expected results:
Arcs are joined together, sketch is constrained.

Actual result:
Cannot join the arcs.

Visual:

Full version info

[code]
OS: macOS 14.4
Word size of FreeCAD: 64-bit
Version: 0.21.2.33771 (Git)
Build type: Release
Branch: (HEAD detached at 0.21.2)
Hash: b9bfa5c5507506e4515816414cd27f4851d00489
Python 3.10.13, Qt 5.15.8, Coin 4.0.0, Vtk 9.2.6, OCC 7.6.3
Locale: C/Default (C)
[/code]

Subproject(s) affected?

Sketcher

Anything else?

I believe this is a regression.
Two arcs cannot be joined in FreeCad 21.0+.
I verified this on 20.2 and it would be over-constrained but work.

Code of Conduct

• I agree to follow this project's Code of Conduct

[maxwxyz]

can you check if you have enabled the setting auto remove constraints and play with it on/off?
Could you please also test this in the current dev version: https://github.com/FreeCAD/FreeCAD-Bundle/releases/tag/weekly-builds

[dancesWithMachines]

can you check if you have enabled the setting auto remove constraints and play with it on/off?

I had auto remove redundants enabled by default. I disabled it and now it behaves like this.
On 21.2 I get over-constrain error like below.

On 20.02 I initially get same error, but if I try to move the sketch it resolves.

Could you please also test this in the current dev version: https://github.com/FreeCAD/FreeCAD-Bundle/releases/tag/weekly-builds

On newest dev-build (36958-2024-04-23 ) it just crashes and exits.

Crash log from devbuild: freecad_crash.zip

[maxwxyz]

@PaddleStroke FYI

[PaddleStroke]

I know what is happening here. I have seen this kind of issue before. The problem is that coincidence is taking 2 dof and arcs have only 5 dof. So you make a coincidence first with the centers, that's 2 dof. Then try to make coincidence with both end point. So that's total 6 dof. So the arc is over constrained. So the auto remove removes the last one. So I don't know what to say. I agree that from a user perspective it's annoying and seems to be a bug. But it's actually not. Perhaps we could accommodate the code to handle this situations. But I don't know how hard that would be

…

On Thu, Apr 25, 2024, 20:49 Max Wilfinger ***@***.***> wrote: @PaddleStroke <https://github.com/PaddleStroke> FYI — Reply to this email directly, view it on GitHub <#13645 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEYO6MLQR6L62DPFZAK6PADY7FF5LAVCNFSM6AAAAABGZHGM2OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANZXHE2TMMZWGU> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[dancesWithMachines]

In this particular scenario equal can be used to achieve same result, but this feels like workaround'ing. From logical point of view I don't see why joining two arcs would not work, it isn't intuitional.

[dancesWithMachines]

The scenario from this bug is very simple, but I originally come up from this.

Here the outcome is different, depending on what I do first.

[PaddleStroke]

I totally agree that this feels very much like a bug. I have argued that in the past.
I'm just not sure how to solve it. Anyone know how other Cads are handling this situation?

Maybe we need to add an exceptions in the solver for coincidence constraints for cases where arcs have 3 coincidences. Something like :

if (arc has 3 coincidence)
      ignore one dof of center coincidence