On 07.02.12 19:17, Stuart Stevenson wrote: > My comment was dual purpose. > > To encourage you to use the APT language and syntax when you develop the > language for running LinuxCNC.
I'll have to look for some APT language doco then. Maybe it does provide sufficient gcode equivalents for hand-writing of routines, in addition to automated gcode generation from higher-level CAM abstractions, such as circles and polygons. But is the following much more readable than raw gcode? INDIRV/0,1,0 $$ direction the tool initially moves I am a great fan of literate programming, and a previous tool translates full English sentences into C code, with function overloading, to implement populations of interacting state machines of multiple classes. e.g. The full Call Control code for a PABX, with hundreds of extensions, many trunk (central office) lines, and supplementary state machines to handle conference calls, allowing individual legs to be put on hold, then clearing down when all participants have hang up. That tool had the ability to compile the semantics from each sentence, while discarding extra phrases included only for commenting purposes. (Admittedly, some application specific table entries were required to achieve that.) Oops, I seem to have revealed where my heart is. > Using the APT language would encourage capable people to take Apt360 and > develop it further. Yeaaaah. I saw that it's primarily been auto-translated to C from an ancient fortran code base, so it's likely to still be highly unstructured. It looks like a scary beast. A big problem is that maintaining and updating ancient code involves a lot of effort to generate a mental image of the whole code base, so that it's safe to make structural improvements, or even understand all the consequences of a proposed tweak. If there was only ever one maintainer, and he's no longer on the scene, then we're Dr. Livingstone up the Nile. The biggest hurdle, though, is that the above is a lot of fuggly work, while exploring new grammars, and demonstrating that the "can't be done" grammar isn't too hard at all, is enjoyable. And that is what keeps a project going. It's a pity, because I applaud standardisation, both for its benefits, and to satisfy my taste for order. But I'm not finding any motivating enjoyment in the prospect of battling with that code base. > Yes - I can be a shameless promoter :) Advocacy is good. We need a broad base of ideas. I'd climb on board if I could see a way to keep interested long enough to do any good. Erik -- Good judgement comes from experience. Experience comes from bad judgement. - Jim Horning ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
