Hi,
I hope to have this ready now.
Am 06.01.23 um 11:52 schrieb andy pugh:
On Fri, 6 Jan 2023 at 09:52, <l...@pibf.de> wrote:
I found values with negative exponent, but not with positive.
How to proceed?
- 1/CART_FUZZ (#define CART_FUZZ (1.0e-8))
Probably 1/CART_FUZZ. The compiler will replace that with a constant
anyway, so it isn't any more computation.
done.
It's probably better to match the surrounding code than to match the
coding style.
I will do this. It takes some time.
I copy the changed code to a new file, run
indent -kr -i8
as suggested here: https://slurm.schedmd.com/coding_style.pdf
and copy it back.
done.
No, I can't say that I like that particularly. Why not use an extra
(optional) letter in the G6.2 command?
G6.2 uses now K, P, R.
The G code interpreter is a black box for me.
Suggestion Q?
Q sounds like a good choice. You would need to add the G-codes to here:
https://github.com/LinuxCNC/linuxcnc/blob/master/src/emc/rs274ngc/interp_check.cc#L342
(G5 is already in there)
done with Q.
gcode files in
https://github.com/jjjjjf/linuxcnc/tree/nurbs_G5_G6_xyz/nc_files/nurbs/G6
are modified with Q instead of #1.
/src/emc/rs274ngc/interp_convert.cc line 142.
" setup_pointer settings) //!< pointer to machine settings" ?
Persistent no. They are used only for calculations
between start of G6.2 and the last line with G6.2 Kx.
I was looking in the wrong branch.
See the examples:
https://github.com/jjjjjf/linuxcnc/tree/nurbs_G5_G6_xyz/nc_files/nurbs/G6
If in a g code file a additional nurbs curve with 6.2 starts,
they habe to be reset.
I think that they should be reset on m2 (ie the end of a G-code
program) like everything else is.
ie. in convert_stop:
https://github.com/jjjjjf/linuxcnc/blob/nurbs_G5_G6_xyz/src/emc/rs274ngc/interp_convert.cc#L4930
done.
There is a file
https://github.com/jjjjjf/linuxcnc/blob/nurbs_G5_G6_xyz/nc_files/nurbs/G6/Rhino_Convert_NURBS.rvb
which is a Rhino script. I modified it to match for the Q number.
But because I have no Rhino, I can not test it.
Joachim
_______________________________________________
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers
