see http://www.linuxcnc.org/docview/html//gcode_overview.html#sub:Numbered-Parameters
example have an array of values #200 to #300 and use #3 as a counter #[200+#3] Dave Caroline On Wed, Aug 26, 2009 at 7:59 AM, Erik Christiansen<dva...@internode.on.net> wrote: > On Wed, Aug 26, 2009 at 12:34:14AM +0200, Andy Pugh wrote: >> 2009/8/26 Alan Battersby <alan.batter...@ntlworld.com>: >> >> > So can someone suggest the best way of achieving my aim? >> >> Possibly Python? ie instead of having the cleverness in g-code, have >> it in a higher level language that produces g-code. > > There are almost too many text processing languages to choose from. When > I created a code generator which converted Structured English to a 'C' > implementation of large and numerous interacting state machines, Awk > (gawk) fitted the bill admirably. (Its C-like syntax, plus associative > arrays and regular expressions [1], made life a lot easier.) > > One convenience of gawk is that a variable is simultaneously a float and > a string. Thus a textual depth value can be read in, computations performed > without explicit type conversion, and the result merged into textual gcode > output, again without the need for explicit type conversion. > > With gawk now providing coprocess communications, it would be simple to > have it read (separate) surface profile and cutting data files, then > generate and feed gcode to emc2, so long as the latter will accept it. > (If it doesn't yet, then that would be very interesting to look at. > However, reading from a file is the same as reading from a pipe, so emc2 > shouldn't know that it is reading from gawk output instead of a file.) > > For those of us who were programmers, the approach also offers the > option of writing all our "gcode" in a higher level pseudocode, > possessing increased readability, then using another gawk filter convert > it to raw gcode. The gawk manpage is a little too terse for learning, > naturally. > > Should the language appeal, the Addison - Wesley book "The AWK > Programming Language" is a goldmine. Its mere 204 pages plus powerful > Keyword-in-context index speed learning by minimising verbiage, and > maximising routes to the information sought. > > I'd be pleased to assist the climbing of the inevitable learning > curve. And if a suitably difficult challenge can be found in your > generator, it would be interesting to help nut out a solution. > > [1] The same EREs as "grep -E" uses. (Not some odd dialect.) > I wouldn't tackle any text processing task without them. > > Erik > > ------------------------------------------------------------------------------ > Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day > trial. Simplify your report design, integration and deployment - and focus on > what you do best, core application coding. Discover what's new with > Crystal Reports now. http://p.sf.net/sfu/bobj-july > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users > ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
