How does FreeCAD choose the direction of angle constraints?

[openBrain]

Just a quick reply ATM, I'll carefully read your post later.

3) if one or both lines start at the outer end and selecting both lines and the center yields the complementary angle.

This isn't what I'm observing.
If (only) one line starts at the outer end, I indeed have the complementary angle when center is selected.
But if both line start at the outer ends, selecting center gives me the supplementary angle, i.e. the opposite angle with same value as the inner angle.

[thomas-neemann]

apparently you can mix gtk and qt. it also seems to be possible to run gtk on ms windows and mac os.

then does it make sense to write new functions in freecad in gtk?

if something is revised it will change in gtk?

then qt would run out and in the end the rest could still be changed.

@wmayer @yorik @realthunder

what's your opinion?

[chrisb]

Just a quick reply ATM, I'll carefully read your post later.

3) if one or both lines start at the outer end and selecting both lines and the center yields the complementary angle.

This isn't what I'm observing.
If (only) one line starts at the outer end, I indeed have the complementary angle when center is selected.
But if both line start at the outer ends, selecting center gives me the supplementary angle, i.e. the opposite angle with same value as the inner angle.

You are right, but I wouldn't distinguish this case, because you can drag the angle's display always to the supplementary angle.

[openBrain]

You are right, but I wouldn't distinguish this case, because you can drag the angle's display always to the supplementary angle.

Fair point.
So we have only 2 cases currently :
* If lines starts or ends at the same (common) point, user is stuck with inner angle, being the center selected or not
* If one starts where the other ends, user can vary between inner or complementary angle depending on whether center is selected or not
Am I correct ?

[openBrain]

Sorry chrisb, very fragmented conversation.

If case two could be changed to work with the complementary angle too, we would have a sound concept and for all of these cases a possibility to per user's choice the inner or the complementary angle. Is it worth writing a feature request?

Please give me just some time to analyse the code. If it's a matter of 2 code lines, I'll do it directly. Otherwise a feature request will be a good idea.

Things are different when the angle between lines is specified which don't have a common center: Then the angle is drawn between the start of one and the end of the other line. It is not possible to specify the complementary angle.

Hmmm... In case of 2 separate lines, I can get the complementary angle by selecting any line endpoint in addition to the lines themselves.

[chrisb]

Sorry no time left. I tested before and think I got the complementary angle if both lines start at their outer ends, but I cannot reproduce now. So You are most probably correct. I will have some time tonight.
And thanks for looking at it.

[openBrain]

OK, after having a look at the code, it looks like the "ability" to get complementary angle is more a side effect than a real feature.
After some testing, I also admit I hardly understand what is the behavior when selecting 2 lines and a vertex that are not already linked to each other by some way...
Probably the best is to re-think it as a whole, define the behavior we want as a new one, then open a feature request.

[chrisb]

Thanks for looking at it! I will ponder about the issue.

[abdullah]

OK, after having a look at the code, it looks like the "ability" to get complementary angle is more a side effect than a real feature.
After some testing, I also admit I hardly understand what is the behavior when selecting 2 lines and a vertex that are not already linked to each other by some way...
Probably the best is to re-think it as a whole, define the behavior we want as a new one, then open a feature request.

I do not know which combinations produce which angle. If a feature request is done, I may eventually be able to look at the code and see what is exactly happening behind the scenes. Of course, if you have time and want to undertake it, you are more than welcome.

A word on the line, line, vertex. This is the AngleViaPoint kind of Angle constraint (DeepSOIC created it). That very same constraint is also used in many tangency/normal constraints (not all). It has a locking mechanism to prevent flipping (or at least reduce it, I do not remember by heart the internals).

[openBrain]

Of course, if you have time and want to undertake it, you are more than welcome.

Will greatly depends on my spare time at the moment the feature request will be created.