Classicladder sample configs suggestions

Hey Chris

I see in the demo_sim_cl file in truck u committed changes to use the new features of classicladder. I'm not
sure if you intended to hide the classic ladder GUI . If you want to see the GUI, have the symbols window
display properly and have no problems saving ladder programs I would change the hal file a bit.

change:
loadusr -w classicladder --nogui demo_sim_cl.clp

to:
loadusr classicladder  demo_sim_cl.clp

I also see that demo_sim_cl shoud be changed from:

loadusr -w classicladder --nogui demo_step_cl.clp
loadusr classicladder

to:
loadusr  classicladder demo_step_cl.clp

If you already knew this please disregard.

I really liked the changes Jeff add to display the signal names instead of symbol names. I am thinking
classicladder should load the signal names into the symbol array so they are displayed in the symbol window.
 wow that's a mouth full!

regards Chris Morley

_________________________________________________________________

Re: Classicladder sample configs suggestions

On Sat, Sep 01, 2007 at 12:31:33AM +0000, Chris Morley wrote:
> 
> Hey Chris
> 
> I see in the demo_sim_cl file in truck u committed changes to use
> the new features of classicladder. I'm not sure if you intended to
> hide the classic ladder GUI . If you want to see the GUI, have the
> symbols window display properly and have no problems saving ladder
> programs I would change the hal file a bit.

In AXIS now you can start the classicladder gui from the File menu.  I
have no opinion about whether the demo_sim_cl should start it
initially or not...  If you'd like it to, please go ahead and change
it.

> I really liked the changes Jeff add to display the signal names
> instead of symbol names. I am thinking classicladder should load the
> signal names into the symbol array so they are displayed in the symbol
> window.  
> wow that's a mouth full!

Yes I really like this change too!  

I could not figure out the symbol window.  It was just empty for me.
If it would show the table of %I0 => signal name, I think that would
be very nice.  (I'm not sure if this is what you mean or not)

Thanks again for all your work on CL.  I think it's becoming a lot
more useful and friendly.

Re: Classicladder sample configs suggestions

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/

_______________________________________________
Emc-developers mailing list
Emc-developers@...
https://lists.sourceforge.net/lists/listinfo/emc-developers

Re: Emc-developers Digest, Vol 16, Issue 4

Dear Jeff:
Thanks for your help.
I have uploaded my prbs.c onto the wiki as you told me.

http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?ContributedComponents

Two further problems I am interested in are
1) how the user can compile the file after downloading the
user-contributed components
 When I develop this module, I modified
emc/src/Makefile,
emc2/src/Makefile.inc
so that it can be compiled.
As you wrote in the last email, is it that easy to use emc2-dev to
make the compilation an automatic process?

2) How can I modify the my file later if I may want to improve it? Can
the modifications be recorded as in CVS system?

Regards
Xuecheng

> From: Jeff Epler <jepler@...>
> Subject: Re: [Emc-developers] PRBS (Pseudo-Random Binary Sequence)
>         signal  generator
>
> I have created a page on the linuxcnc wiki for user-contributed
> components.
>     http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?ContributedComponents
>
> There are several advantages to this approach that I see:
>  * This is community driven; there is no need or requirement for
>    an "official" developer to approve the component, or coordinate with
>    the contributor whenever a new version is released
>  * New components can reach users before the next emc 2.x release cycle
>    is done
>  * As the number of components grow, and as they become useful to more
>    specific (but smaller) groups of users, the core emc2 distribution
>    does not grow to be unmanagably large
>
> For a properly written third-party component, installation should be as
> easy as one of these commands:
>     sudo comp --install example.comp
>     sudo comp --install example.c
>     chmod 755 example.py && sudo cp example.py /usr/local/bin
>     make && sudo make install
> once the 'emc2-dev' package has been installed.
>
> Jeff
>
>

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/

Re: PRBS (Pseudo-Random Binary Sequence) signal generator

On Sun, Sep 02, 2007 at 04:42:55PM +0800, Xuecheng Xi wrote:
> Dear Jeff:
> Thanks for your help.
> I have uploaded my prbs.c onto the wiki as you told me.
> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?ContributedComponents

Thank you.  I hope others will add to this page as well.

> Two further problems I am interested in are
> 1) how the user can compile the file after downloading the
> user-contributed components
>  When I develop this module, I modified
> emc/src/Makefile,
> emc2/src/Makefile.inc
> so that it can be compiled.
> As you wrote in the last email, is it that easy to use emc2-dev to
> make the compilation an automatic process?

Yes, any single-file component should be installable with comp:
    comp --install prbs.c
however, I found that several changes were needed before it compiled on
my system (CVS TRUNK version, simulator):

--- prbs.c.orig	2007-09-01 09:35:06.000000000 -0500
+++ prbs.c	2007-09-02 08:23:06.000000000 -0500
 <at>  <at>  -26,20 +26,15  <at>  <at> 
 #include <rtapi.h>		/* RTAPI realtime OS API */
 #include <rtapi_app.h>		/* RTAPI realtime module decls */
 #include <hal.h>		/* HAL public API decls */
-#include "hal_priv.h"	/* HAL private API decls */

 
-#ifdef MODULE
 /* module information */
 MODULE_AUTHOR("Xue-Cheng Xi");
 MODULE_DESCRIPTION("PRBS signal generator for EMC HAL");
-#ifdef MODULE_LICENSE
 MODULE_LICENSE("GPL");
-#endif /* MODULE_LICENSE */
 static int num_chan = 2;	/* number of channels - default = 2 */
 MODULE_PARM(num_chan, "i");
 MODULE_PARM_DESC(num_chan, "number of channels");
-#endif /* MODULE */

 
 /***********************************************************************
 <at>  <at>  -291,7 +286,7  <at>  <at> 
 		return retval;
     }
     rtapi_snprintf(buf, HAL_NAME_LEN, "prbs.%d.output", num);
-    retval = hal_pin_float_new(buf, HAL_RO, &(addr->output), comp_id);
+    retval = hal_pin_float_new(buf, HAL_OUT, &(addr->output), comp_id);
     if (retval != 0){
 		return retval;
     }

After these changes, the component installed with 'comp --install' and I
was able to successfully load it.

> 2) How can I modify the my file later if I may want to improve it? Can
> the modifications be recorded as in CVS system?

I believe you can simply replace the existing file using 'upload', and
note in the page text that the component was updated on that date.
There is no revision control for uploaded files on the wiki.

Jeff

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/

opinion on classicladder ideas

looking for opinion:

    In cvs Jeff made some great changes to import HAL signal names to use as symbols (eg ' lube out>' instead of '%Q3').
    To do this he cheated a little and circumvented displaying the HAL signal names in the symbol window,and the
signal names take presidence over symbol names.
    In my personal version I fixed the window problem  but  symbol names take presidence over HAL signal names.
(in other words HAL signal names are only used if there is not already a symbol name for the variable)
    One problem I see is that very long signal names look terrible in classsicladder. (I am wondering if there is
a way to change the font smaller when displaying text but can not find a way in GTK.-I'm sure there is a way)
     Also my version only checks for HAL signals names when classicladder is loaded-meaning signals created
after classicladder loads will not be displayed.

So the questions....

should classicladder always display HAL signal names if they are available or should the programmer
decide if he wants to use a different name.
and
should the symbol window be updated with new signal names (maybe with a "check for signals" button??) or is
loading then once enough (this is not a problem for a properly laid out HAL file but for a
new-to-integrating person ..

Thanks
 Chris Morley

_________________________________________________________________
News, entertainment and everything you care about at Live.com. Get it now!
http://www.live.com/getstarted.aspx
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/

Re: opinion on classicladder ideas

Hi Chris

Signal names are okay if they are less than 8 characters.  Get 'em much longer and they cause major display
issues.  I say okay because I started programming PLC's when they all used B, Q, and I and had no real names.  I
still print a cheat sheet.

One rub I do have with some stuff in Ubuntu and EMC is a failure to accurately  and consistently display state.
I'd much rather a system that always listed a coil as %Q4 than a system that sometimes uses that label and
sometimes tells me its ext-estp.

Rayh

On Mon, 3 Sep 2007 08:36:44 +0000, EMC developers
<emc-developers@...> said:
>
>looking for opinion:
>
>    In cvs Jeff made some great changes to import HAL signal names to use as symbols (eg ' lube out>' instead of '%Q3').
<s>

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/

Re: opinion on classicladder ideas

my opinion for what it's worth:

>    In cvs Jeff made some great changes to import HAL signal names to use 
> as symbols (eg ' lube out>' instead of '%Q3').
>    To do this he cheated a little and circumvented displaying the HAL 
> signal names in the symbol window,and the signal names take presidence 
> over symbol names.

was this done to prevent breaking some bit of code? why not just use the 
HAL signal name as the symbol name? having two names for the same object 
sounds like asking for trouble.

>    In my personal version I fixed the window problem but symbol names 
> take presidence over HAL signal names. (in other words HAL signal names 
> are only used if there is not already a symbol name for the variable)

i didn't even know you could change the names of classicladder variables.

>    One problem I see is that very long signal names look terrible in 
> classsicladder.

one way to deal with this is to truncate the name to N characters and 
display the full name when the user mouses over the element.

> (I am wondering if there is a way to change the font 
> smaller when displaying text but can not find a way in GTK.-I'm sure 
> there is a way)

after staring at this for a while I decided (again) that the whole 
classicladder GUI should just be rewritten from scratch. now that it is 
usable of course :)

>     Also my version only checks for HAL signals names when classicladder 
> is loaded-meaning signals created after classicladder loads will not be 
> displayed.

this should be fixed then

>
> should classicladder always display HAL signal names if they are 
> available or should the programmer decide if he wants to use a different 
> name.

classicladder should always use the HAL signal name. there's no reason to 
confuse things by having two names for the same variable.

> and should the symbol window be updated with new signal names 
> (maybe with a "check for signals" button??) or is loading then once 
> enough (this is not a problem for a properly laid out HAL file but for a 
> new-to-integrating person ..

an easy way to do this is to read the hal signal names when the window is 
redrawn

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/

Re: opinion on classicladder ideas

On Mon, Sep 03, 2007 at 06:56:25PM +0000, ben lipkowitz wrote:

> after staring at this for a while I decided (again) that the whole 
> classicladder GUI should just be rewritten from scratch. now that it is 
> usable of course :)

I disagree.

> classicladder should always use the HAL signal name. there's no reason to 
> confuse things by having two names for the same variable.

I agree.

Chris

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/

[ emc-Bugs-1788078 ] gcode produces "bug: function should not have been called"

Bugs item #1788078, was opened at 2007-09-04 16:56
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=106744&aid=1788078&group_id=6744

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: EMC2 G-code Interpreter
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Jeff Epler (jepler)
Assigned to: Kenneth Lerman (lerman)
Summary: gcode produces "bug: function should not have been called"

Initial Comment:
The following gcode:
 O102 if 0
 O102 endif
 M2
produces "bug function should not have been called".  Tested on CVS TRUNK of this date.  The gcode may or may
not be valid, but in any case it should not produce this diagnostic which indicates a software bug in emc.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=106744&aid=1788078&group_id=6744

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/