I would just have my subroutines all call a separate subroutine that initializes global named parameter. So: o<gene:Init> sub #<_gene:height> = 62 #<_gene:chestSize> = 38 o<gene:Init> endsub
And: o<gene:millFront> sub o<gene:Init> call (do all sorts of stuff using global parameters that have been setup) o<gene:millFront> endsub o<gene:millBack> sub o<gene:Init> call (do stuff for the back) o<gene:millBack> sub Regards, Ken [No nasty comments about a height of five-two and a chest size of 38.] KL On Mon, Aug 17, 2015 at 10:43 AM, Gene Heskett <ghesk...@wdtv.com> wrote: > On Monday 17 August 2015 07:21:33 andy pugh wrote: > > > On 16 August 2015 at 18:52, Gene Heskett <ghesk...@wdtv.com> wrote: > > > But as I contemplate adding another section to the code by way of > > > copy/paste, with an intervening paragraph of setting new values for > > > the 2nd tool, it strike me that either this code should source a > > > config file, or it could become a callable SUB for the configuration > > > controlling file. > > > > > > So, one, is it possible, and two, whats the prefered order and > > > syntax? > > > > I think that it is optionally possible to access _the_ INI file from > > G-code, but not a generic file. > > http://www.linuxcnc.org/docs/html/remap/structure.html#_optional_inter > >preter_features_ini_file_configuration_a_id_sub_ini_features_a > > > > So, if your INI contains [RS274NGC]FEATURES = 4 then you can use (for > > example) #1 = #<_ini[AXIS_0]PGAIN> in the G-code. > > Which probably doesn't achieve what you want. Note that "4" is a > > bit-field, You would typically use 12 to enable HAL-pin reading too. > > > > I suspect that what you really want to do is use a subroutine. In fact > > I alterd my Lathe GUI to use subs and params rather than HAL-pin > > access, partly because of queuing behaviour. > > > > example > > > > O<cut_joint> CALL [6.35] [13] > > > > ... > > > > O<cut_joint> SUB > > (DEBUG, I will cut #2 fingers with a #1 diameter cutter) > > O<cut_joint> ENDSUB > > In my thoughts and since 90% of the variables are globally defined in the > #<_name_of_var> style, subroutines have access to them once set without > being passed. They'll survive, in my tests, everything but a powerdown > reboot! > > What I have in mind is making one nominally 300 line subroutine out of > what I have, and control it by a small header file that sets these > variables for various jobs, and once its configured to do that job, call > the main program as a subroutine to do that version of the job. This > main subroutine also calls its own subroutines defined in this big file > before use. > > The idea is to have a config file for carving the ends of an upper part > of the box board that is the long side. > > And one to carve the mirror of that for the upper parts of the ends of > the box that is the short side. > > One to carve the side/long boards of its base skirt > One to carve the mirror in the end/short boards of the base skirt > > Thats 4. > > one, after a tool change, to do the 1/8" roundover of the top side of the > long/side upper boards > One, after a tool change, to do the roundover of end/short upper boards > One, after the tool change, to do the roundover of the side/long base > boards > One, after the tool change, to do the inside, was bottom faces, again > with 3 more variations. > > So 8 config files, all calling the same subroutine. > > Getting a precise cut with the bearingless point cutting roundover, on > both faces will be a trick. It needs to cut a full 90 degree arc. 2 thou > too deep in either x or z and it will be a lot of sandpaper to correct. > 10 thou too shallow and the joints won't bottom well. > > Sequenced to do as much as can be done without removing the board from > the jig, but it will have to be removed and turned over to do the > roundovers on the back sides. > > This code should be able to do all that, simply by using the globally set > vars in the calling file. > > And because the boards might vary in both thickness and width as they > come S4S, but may have dried further before being shipped to me, I might > have to do a different autozero than I am doing now by buying a > reneshaw, and probing the top, and both sides (x direction) to ascertain > the boards true width. The code, if that is adjusted, self expands or > contracts in the x axis to maintain the pattern it carves to exactly fit > the board. But because that would effect the joint fits when the mirrors > mate, I an inclined not to do that, but cut as if the board was 20 thou > narrower, oriented in the jig so the resultant short cut is on the same > edge of the board on both ends, and jointer plane it off to get > instantly identical width boards (until the humidity changes of > course :). By then I hope to have them assembled, squared and glued up. > > There is also a separate operation, a thru hole for the assembly screw in > the center of each finger, with a 5/16" deep circular counterbore for > the screw head, and a 1/8" deep square pocket a few thou wider than the > counterbore that a contoured top Ebony button will be glued into. > Screws to be pilot drilled thru the predrilled holes and installed while > the glue is still green if possible. > > At least thats the plan at this stage. I have one chest I built last > winter on the small mill which needed a lot of manual help and used > every last mm of its x axis, and 4 boys left, so 3 more to do yet. > > I won't have the roundover bits till the end of the week, so at some > point, I am going to chisel off the left edge stopper of my jig as its > too short, and glue on a much longer wheelbase strip of white ash and > machine it square as I seem to be getting some rocking on the shorter > version. How much longer depends on the max Y of the table, the ball > screw conversion cost me some motion because the stem of the ball screws > nut holder hits the end of the slot in the base coming forward, and the > swarf shielding rubber gets in the way of full away from me movement. I > have some bellows material that may make that less of a problem, but > haven't investigated installing it instead of that huge and thick rubber > sheet furnished that "tents up" leaving the Y screw somewhat exposed. > Built by lowest bidder crap. :( > > The lid is 2 ea 2x12's edge glued, using biscuits as close together as > practical, and to prevent warpage, will have "breadboard" ends made from > (believe it or not) Gibson guitar mahogany neck blanks that weren't > quite up to snuff due to the lay of the grain & thrown out by the shop > that makes them for Gibson right here in these here hills of WV. I have > a friend and ex employee who lives over in that neck of the woods who > checks their burn pile and scarfs them up about weekly for his own > projects. Same exact stuff Grizzly has in their catalog at $75-$175 a > stick! I can usually make both breadboard ends from one neck blank. It > seems almost criminal to burn it, but if we don't get it, that is > exactly what that shop will do when the pile gets big enough. > > Bottom is mahogany faced plywood, with a 5/16" wide groove to hold the > cedar planks for the lining, and the box edges capped with mahogany > strips with a matching 5/16" groove for the tops of the cedar planks. > > Thanks Andy. I was pretty sure it could be done, but you've confirmed it. > > Cheers, Gene Heskett > -- > "There are four boxes to be used in defense of liberty: > soap, ballot, jury, and ammo. Please use in that order." > -Ed Howdershelt (Author) > Genes Web page <http://geneslinuxbox.net:6309/gene> > > > ------------------------------------------------------------------------------ > _______________________________________________ > Emc-developers mailing list > Emc-developers@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-developers > -- Kenneth Lerman 55 Main Street Newtown, CT 06470 ------------------------------------------------------------------------------ _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
