Re: [Emc-users] Macro Language

[Sven M. <em...@in...>]

Alex Joni schrieb:
>> Ask user to define minimum and maximum X, Y and Z positions by moving
>> the tooltip to the relevant positions manually or by entering the values
>> numerically. Then use a procedural (possibly recursive) approach to scan
>> the work pieces surface with a touch probe.
> 
> Not the most advanced waz to do this, but it does a good job.
> http://cvs.linuxcnc.org/cvs/emc2/nc_files/gridprobe.ngc?rev=1.5;content-type=text%2Fplain

The point is:

This is neither the most efficient nor the easiest to read way to achive
this. What would really be nice would be a way to tell emc:

move along X axis to 10(user units), stopping if probe is activated
report position back

Your example above leaves a few things to desire if you want to present
the result to people:
1) a more readable program, with symbolic variable names for starters
2) a more flexible way for the user to specify the borders
3) a way to - for example - use several different sized probes to first
   get a quick rough image of what the piece looks like to then do the
   fine checks with smaller movements and this way hopefully also
   quicker and doing it in a single pass. Just like when you do a
   roughing and a finishing pass when milling.

Also, as I mentioned in my previous mail, it's not clear how you need to
"wire" the probe in hal: which input triggers the G38.2 to stop and
print the position? I couldn't find that mentioned in the docs.

Finally: I'm not necessarily talking about scripting in the place where
currently Gcode is executed. Well documented script (perl, lua, ruby,
python) language bindings to the movement and IO parts of emc2 would
work just as well for me.

Actually, what I would like to see is some relatively optimized version
of surface scanning implemented with a GUI (perhaps in tkEMC or one of
the other GUIs?). Since the existing GUIs already allow manual
movements, I would think that all I would like to see (enter a position
directly or by doing a manual movement, multiple scan passes with
differently sized probe tips,...) should be possible using the same
techniques currently used in the GUIs. Though I'm relatively good at C
programming as well as Perl programming, I don't know how to build an
interface between the two or how to use tcl.

regards,
Sven