Circular Hole Tool

[chrisb]

Sorry in advance for ranting, I wasn't able to put it shorter.

We should separate the code duplication problem from the usability issues.

Code duplication should almost always be avoided; less errors, better maintenance. Technically it can always be done by creating common classes or functions and using them in different places. However, there are rare cases where too many parameters are involved or other things have to be prepared to use the common code. Then these techniques may clutter the code more than they help.
I would leave it to the programmers to decide which way to go. They are all experienced enough to see, where the additional effort to refactor the code is to be preferred against a short sighted gain of time by copy/paste.

The user's view is something different. Different operations can share the same code or vice versa a checked box in an operation can lead to the usage of completely different code.

To me the question is, if the basic concepts are the same or not, and I am not yet sure about it in case of the holes.
If I consider the model itself - which resembles the final product of executing the path operations - I don't care if the hole is drilled or milled or both. It's just a hole, and I know that different operations might have been necessary to create it.
If I take a machinists view it is rather the operations I have in mind and there might be little difference whether I put a drill in the machine and it creates a hole or if I take an endmill and it creates a hole. But if I look at the whole process there are more differences: center drill + drill in one case, and in the other possibly drilling + helix from inside out or from outside in + pattern to be used.
I cannot follow the argument of possibly taking the wrong tools, because that is always a problem. The final question to me is, if the helical hole is rather a special pocket or still a hole.

[GeneFC]

As long as it works consistently and reliably, and as long as the needed functions are available, I am happy with task panels, property panes, or whatever. Grouped, dedicated, complex, simple, cluttered, uncluttered; doesn't really matter.

Beginners will be daunted by almost anything (just check the Help forum), and experienced users will learn whatever is needed. Sure, it is great if the interface is *exactly* what best suits *me*, but that seems a bit unrealistic.

Gene

[cox]

If the combo operation is used it would be nice for Icon and operation name in tree view to indicate witch operation is selected.

Another point I may see as problematic is that there is, as far as I can think of no president for "all in one" operations in FreeCAD, so it might diverge some from the existing ux. That is not to say that we should not go that route, but it is worth contemplating.

[chrisb]

As long as it works consistently and reliably, and as long as the needed functions are available, I am happy with task panels, property panes, or whatever. Grouped, dedicated, complex, simple, cluttered, uncluttered; doesn't really matter.

Beginners will be daunted by almost anything (just check the Help forum), and experienced users will learn whatever is needed. Sure, it is great if the interface is *exactly* what best suits *me*, but that seems a bit unrealistic.

Gene

+1

[Konstantin]

So? Have you decided what to do with this tools?

[sliptonic]

So? Have you decided what to do with this tools?

Mlampert is doing some refactoring that will help eliminate duplicated code in a lot of operations. It doesn't make sense for me to change anything until that is done or ready to apply to drilling and helix.

I know people have personal preferences about the GUI but Genefc's comment is right on. Reliability and features are more important.

I'm still leaning heavily toward combining them because it makes cloning operations a natural extension of the work flow.

If the combo operation is used it would be nice for Icon and operation name in tree view to indicate witch operation is selected.

Another point I may see as problematic is that there is, as far as I can think of no president for "all in one" operations in FreeCAD, so it might diverge some from the existing ux. That is not to say that we should not go that route, but it is worth contemplating.

I like the idea of changing the icon/label in the tree but it's hard to do in a way that feels natural. I'll look at it though.

I really don't think this is the unholy abomination that people people fear. It's worth a try and if it sucks we can undo.