View Issue Details
|ID||Project||Category||View Status||Date Submitted||Last Update|
|0003298||Sketcher||Bug||public||2018-01-03 18:12||2018-01-13 12:06|
|Target Version||0.18||Fixed in Version||0.17|
|Summary||0003298: Validate sketch tool reports false positives on tangent constraints|
|Description||The Tool Menu->PartDesign->Validate sketch reports missing coincidences if the tangent constraint is used on the endpoints only.|
It should be the other way round: If points are coincident and have a tangent constraint the redundant and often problematic coincidence should be reported.
I tried to reproduce the issue with the most simple sketch of a line plus attached arc, but in that case the missing coincidence wasn't reported.
|Steps To Reproduce||See forum discussion|
|Additional Information||OS: Mac OS X|
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.17.12915 (Git)
Build type: Release
Branch: (HEAD detached at 90a71f3)
Python version: 2.7.14
Qt version: 5.6.2
Coin version: 4.0.0a
OCC version: 7.1.0
Locale: German/Germany (de_DE)
|Tags||No tags attached.|
validateSketchFalsePositive.FCStd (3,933 bytes)
This reports 2 non-coincidents
The apparently problematic one is the one on the ellipse arc endpoint to the right, where the coincident and tangency are.
What FC is trying to notify the end-user is that at that point there are more vertices than the ones that the number of vertex-constraining constraints can fix in place. Let's check:
There is a coincidence arc endpoint to line endpoint, which removes the potential coincidence between the arc and the line.
However, at that same point, there is the endpoint of the major axis of the elliptical arc. This point is geometrically coincident with those other vertices. However, there is no (external) constraint forcing a coincidence. So what FC is notifying is: "hey! take care, you have points apparently on the same spot but I can not detect anything forcing them to stay there".
Now, you can argue: "Ok, but this major axis issue is going to drive everybody crazy". I think you are right, but my question would be: "Do we have a situation where we would like this to be detected?". Maybe the one attached is one far-fetched example. There I intentionally did not make a coincidence constraint between the two construction lines.
I think it is not a bug, but it is indeed annoying. I think we need a solution for it. Maybe marking non-coincidence of construction geometry with a different colour?
You will come with something better...
|2018-01-03 18:12||chrisb||New Issue|
|2018-01-03 18:12||chrisb||Status||new => assigned|
|2018-01-03 18:12||chrisb||Assigned To||=> abdullah|
|2018-01-03 18:12||chrisb||File Added: validateSketchFalsePositive.FCStd|
|2018-01-03 20:34||Kunda1||Relationship added||related to 0003299|
|2018-01-08 07:29||abdullah||File Added: val2-detection.png|
|2018-01-08 07:29||abdullah||File Added: val2.png|
|2018-01-08 07:29||abdullah||Note Added: 0010728|
|2018-01-11 09:55||abdullah||Note Added: 0010736|
|2018-01-13 12:06||wmayer||Status||assigned => closed|
|2018-01-13 12:06||wmayer||Resolution||open => fixed|
|2018-01-13 12:06||wmayer||Fixed in Version||=> 0.17|
|2018-01-13 12:06||wmayer||Note Added: 0010752|