the fellow over at
http://www.linuxcnc.org/component/option,com_kunena/Itemid,20/func,view/catid,38/id,14365/limit,6/limitstart,6/lang,english/#14623
came up with the following gcode (101.ngc in the attachment; non-essential
lines deleted)
O<101> sub (tool change sub)
o<101> if [#5400 eq #9 and #9 ne 0]
..
o<101> endif
o<101> if [#5400 EQ #9 AND #5399 EQ -1]
...
...
o<101> endif
o<101> if [#5400 EQ #9 and #9 ne 0 AND #5399 NE -1]
o<101> endsub
o<101> endif
As the interpreter stands now, this isnt detected, and as long as there are no
nested block structures, it isnt necessarily an error. It will fail miserably
as soon as nested block structures are used.
however, I propose to firm up the usage of oword labels as follows:
1. labels marking a while/if/repeat shall not be reused
2. 'endwhile','break', and 'continue' require a 'while' with the same label
name and lower line number
3. 'elsif', 'else' and ' endif' require an 'if' with the same label name and
lower line number (yes, there's an undocumented 'elsif')
4. 'endrepeat' requires a preceding 'repeat' with same label and lower line
number
I've done it for 'if' to detect above example label reuse. It's about 8 lines
before making it a function.
I'll make it warning for now (with canon MESSAGE() or to stderr) and see how it
works out
-m
------------------------------------------------------------------------------
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
_______________________________________________
Emc-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-developers
