On Sat, 17 Nov 2018 at 10:26, Tage Johansson <frans.t...@gmail.com> wrote:
>
> So, if I understand everything correctly, there is a g71 in a
> development branch, but it doesn't support pockets.

Yes.

> How ever, you've
> made a python remap that does pockets, and the goal was to convert it to
> c and build it in to the interpreter.

Yes, that was the plan.

Just be aware that if all you want to do is make things on your lathe
then you can simply install the remap on your current LinuxCNC
installation and you will have G71 and G72 available (no coding, no
compiling)

> According to my understanding, the last thing that has to be done before
> g71 and g72 can be merged in to the master branch, is to make that
> pocket support. Right?

It _could_ be merged as it is now, without pocket support, but pockets
would be better.


> I've some experience in tool path programming, and also some experience
> in c. So I wonder if I can help you with this?

If you want to, that would be great. You would need to check with
mozmck if this would be acceptable for introduction in to 2.8 at this
point. (I think probably it would, a new G-code shouldn't affect
existing behaviour)

You can see the changes made, and the files changed, here:
https://github.com/LinuxCNC/linuxcnc/commits/BenPotter/G71
The first 4 commits in that list. (There is probably a way to see them
as a unified diff too)

The version-with-pockets is here, in Python:
https://github.com/LinuxCNC/linuxcnc/blob/andypugh/g71type2remap/configs/sim/axis/g71/python/remap.py

If you need me to explain anything about it to you, then I can try to
remember how it works.

The key to the algorithm, as I recall, is splitting arcs at the
"cardinal points" to get their maximum extents in X and Z. That makes
it easier to determine if a straight X or Z cut intercepts any given
arc segment.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916


_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to