Re: [Emc-developers] Proposal For Named Parameters

2006-12-28 Thread Kenneth Lerman
Mario is correct that one COULD write a filter that converted named parameters to conventional parameter numbers. (I believe someone -- was it on IRC -- commented about having done this.) But you couldn't have the limited scoping of parameter names; at least not easily. You would have to essential

Re: [Emc-developers] Proposal For Named Parameters

2006-12-28 Thread Stephen Wille Padnos
Mario. wrote: >So we could all this orient towards a programmable plugin? >I think it might be more versatile and easier that to rewrite whole >g-code parser. > >If this powerful editing tool is *READILY* available in AXIS, why not use it? > > It's not an editing tool, it's a filter that's appli

Re: [Emc-developers] Proposal For Named Parameters

2006-12-28 Thread Mario .
So we could all this orient towards a programmable plugin? I think it might be more versatile and easier that to rewrite whole g-code parser. If this powerful editing tool is *READILY* available in AXIS, why not use it? On 12/28/06, Chris Radek <[EMAIL PROTECTED]> wrote: > On Thu, Dec 28, 2006 at

Re: [Emc-developers] Proposal For Named Parameters

2006-12-28 Thread Chris Radek
On Thu, Dec 28, 2006 at 03:02:26PM -0500, Kenneth Lerman wrote: > > But... I think cradek will point out that axis already can use python > programs to produce g-code output. or any other language using the AXIS "filter" feature. -

Re: [Emc-developers] Proposal For Named Parameters

2006-12-28 Thread Kenneth Lerman
Possible? Probably. By me? No. But... I think cradek will point out that axis already can use python programs to produce g-code output. Ken [EMAIL PROTECTED] Mark Kenny Products Company, LLC 55 Main Street Voice: (203)426-7166 Newtown, CT 06470Fax: (203)4

Re: [Emc-developers] Proposal For Named Parameters

2006-12-28 Thread Kenneth Lerman
Ray, Thanks for your comments. Someone else pointed out a problem with the brackets -- there is an ambiguous case of #[sqrt[1.0]]. More precisely, it would require more lookahead than I'd like. I really don't want to rewrite the entire parser. There is a need for some delimeter though. Consider

Re: [Emc-developers] Proposal For Named Parameters

2006-12-28 Thread John Prentice
> > I see the wiki page to be read-only for me, so post I can not post it > there. > > On 12/28/06, Ray Henry <[EMAIL PROTECTED]> wrote: Mario Hmm, I had same difficulty at first. You need to "Login". See http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?BasicSteps for details on "wiki work" Jo

Re: [Emc-developers] Proposal For Named Parameters

2006-12-28 Thread Stephen Wille Padnos
Ray Henry wrote: >Hi Ken > >Thank you again for your systematic approach to making changes to the >interpreter and how we enter commands. I do like the idea of named >variables -- I've use a lot of #1000 = 12 or whatever. I don't see any >problem with replacing the number part of the parameter c

Re: [Emc-developers] Proposal For Named Parameters

2006-12-28 Thread Mario .
Would it be possible to add commands like these // or behaviour as a plugin? I mean, it would be read and processed on EMC startup only, so it would not be hardcoded. some 2.5 years ago I planned an interpreter of G-codes, so thta it would speak in BASIC-like, English-like language. With programm

Re: [Emc-developers] Proposal For Named Parameters

2006-12-28 Thread Ray Henry
Hi Ken Thank you again for your systematic approach to making changes to the interpreter and how we enter commands. I do like the idea of named variables -- I've use a lot of #1000 = 12 or whatever. I don't see any problem with replacing the number part of the parameter call with a name. #my