In the file the scale factor is 0.5. See image.PheasantCoucal wrote: ↑Fri Feb 11, 2022 12:39 pm I note that the file you attached was still unscaled
Scaling of Arcs and Elipses
Forum rules
Be nice to others! Respect the FreeCAD code of conduct!
Be nice to others! Respect the FreeCAD code of conduct!
Re: Scaling of Arcs and Elipses
- Attachments
-
- scaled.png (4 KiB) Viewed 376 times
-
PheasantCoucal
- Posts: 18
- Joined: Tue Feb 08, 2022 2:40 am
Re: Scaling of Arcs and Elipses
Again you are correct re the scaling.
As imported, dimensions are in metres. I observed the dimensions still to be in metres and presumed it to be unscaled.
I wasn't entirely successful in reproducing your result. The problem was with Arcs in the Sketch -> Draft step. None of my Sketch arcs were returned as Draft arcs. Some returned as "Circles" but were indeed part circles with incorrect start and end points(angles). And yes I know that an arc is a part circle, but Draft labelled them as circles even though they displayed as arcs. Other Sketch arcs were just not converted to Draft arcs.
In the source, none of the arcs are contiguous with one another. When scaled X 0.5, they are still non-contiguous in Sketcher (There are no Coincident points). To convert metres to millimetres however requires a scale of X 0.001, following which, some of the arcs do become contiguous with one another and Sketch -> Draft returns Shapes to represent them. I suspect that is why you found that you had to Compound Lines and Arcs separately. (The circles are not contiguous with or intersect one another, or any other objects, so they behave as desired.)
Your approach required a priori knowledge of the geometry of each Shape in order to Compound like types. I suspect that it may be enough to Compound any non-contiguous Shapes. In the case of more complex architectural drawings this may be difficult to achieve in practice. Fortunately arcs are less common than straight lines in such drawings and it may be easier just to redraw them when their absence is observed.
You have given me some great insights into the tools I can use. I will now go back and see if I can improve my import workflow.
As imported, dimensions are in metres. I observed the dimensions still to be in metres and presumed it to be unscaled.
I wasn't entirely successful in reproducing your result. The problem was with Arcs in the Sketch -> Draft step. None of my Sketch arcs were returned as Draft arcs. Some returned as "Circles" but were indeed part circles with incorrect start and end points(angles). And yes I know that an arc is a part circle, but Draft labelled them as circles even though they displayed as arcs. Other Sketch arcs were just not converted to Draft arcs.
In the source, none of the arcs are contiguous with one another. When scaled X 0.5, they are still non-contiguous in Sketcher (There are no Coincident points). To convert metres to millimetres however requires a scale of X 0.001, following which, some of the arcs do become contiguous with one another and Sketch -> Draft returns Shapes to represent them. I suspect that is why you found that you had to Compound Lines and Arcs separately. (The circles are not contiguous with or intersect one another, or any other objects, so they behave as desired.)
Your approach required a priori knowledge of the geometry of each Shape in order to Compound like types. I suspect that it may be enough to Compound any non-contiguous Shapes. In the case of more complex architectural drawings this may be difficult to achieve in practice. Fortunately arcs are less common than straight lines in such drawings and it may be easier just to redraw them when their absence is observed.
You have given me some great insights into the tools I can use. I will now go back and see if I can improve my import workflow.
-
PheasantCoucal
- Posts: 18
- Joined: Tue Feb 08, 2022 2:40 am
[SOLVED] Re: Scaling of Arcs and Elipses
My previous post refers to the outcome when using vers 0.19 (as my 0.20 install was broken when I undertook those tests).
I can now confirm that when using vers 0.20 arcs in the source are now returned correctly as arcs, and that I have been able to reproduce roy_043's result.
I can also confirm that in my experience the geometric shape of the source objects is unimportant. All that is important is that they are non-contiguous.
It is thus possible to implement a workflow whereby a selection of apparently non-contiguous elements is processed into draft primitive elements and saved into one or more hidden Groups. The process is then repeated with another selection until there is nothing left. It is tedious because one can't select a block of elements. But it does work.
This thread started with my frustration with Scaling. It turned out to have little to do with scaling but with the way that I was attempting to extract primitive geometric elements from Shapes. Roy_043 has showed me a different way, and I am very grateful for that.
I can now confirm that when using vers 0.20 arcs in the source are now returned correctly as arcs, and that I have been able to reproduce roy_043's result.
I can also confirm that in my experience the geometric shape of the source objects is unimportant. All that is important is that they are non-contiguous.
It is thus possible to implement a workflow whereby a selection of apparently non-contiguous elements is processed into draft primitive elements and saved into one or more hidden Groups. The process is then repeated with another selection until there is nothing left. It is tedious because one can't select a block of elements. But it does work.
This thread started with my frustration with Scaling. It turned out to have little to do with scaling but with the way that I was attempting to extract primitive geometric elements from Shapes. Roy_043 has showed me a different way, and I am very grateful for that.