Back
[00:03:58] <jepler> there is a problem with python components exiting on the signal...
[00:51:28] <danex> * danex is back.
[01:27:34] <danex> * danex is away: Away at the moment
[12:36:15] <alex_joni> jepler: I think I spotted a bug with axis..
[12:36:55] <alex_joni> I made a small program with only a G0 and a G38.2 (probe move) in it, and it gets confused if I want to view it in perspective..
[12:37:53] <alex_joni> I think it doesn't properly know the G38.2 is a linear move, and doesn't calc the workpiece extents
[13:02:40] <jepler> alex_joni: I see
[13:06:40] <alex_joni> jepler: I think the scaling is way off, and I can't find the origin or the tool :D
[13:11:20] <jepler> the extents are also messed up for this program:
[13:11:20] <jepler> G0 X0Y0Z1 / G1 Z0 F8 / M2
[13:12:43] <alex_joni> I didn't have much more than that in my program ;)
[13:12:59] <alex_joni> actually I sent it as a reply to Goettler's email about probing
[13:13:12] <alex_joni> so if you need to use it for testing you'll find it there
[13:15:03] <SWPadnos> hi there
[13:16:02] <alex_joni> hi SWamPy ;)
[13:16:10] <SWPadnos> heh
[13:16:21] <alex_joni> kidding... how are you?
[13:16:23] <SWPadnos> hey - I'll be closer to your neck of the woods this week
[13:16:26] <alex_joni> enjoying the weekend?
[13:16:28] <SWPadnos> I'm okl
[13:16:30] <SWPadnos> ok
[13:16:31] <alex_joni> oh.. poland?
[13:16:48] <SWPadnos> yeah - we went Kayaking yesterday, before the rain started
[13:16:59] <SWPadnos> yep - Krakow on Wednesday (Thursday morning)
[13:17:34] <SWPadnos> I really should learn how to pronounce a few words (bathroom, coffee, please, thank you ...)
[13:20:38] <jepler> you forgot: "yes, I am an american, but I do not support the president's policies"
[13:21:01] <alex_joni> jepler: the phrase "yes, I am an american" usually excuses you of anything in europe
[13:21:05] <SWPadnos> I already know that one ;)
[13:22:02] <SWPadnos> you get further if you rearrange that: 'I thinkthe president is an idiot, even though I'm an American"
[13:24:46] <alex_joni> jepler: thanks, a lot better..
[13:24:50] <alex_joni> but still not 100% right
[13:25:19] <alex_joni> for the program I posted, it only shows the X extent, no Y extent
[13:26:33] <alex_joni> and G38.2 appears in magenta on the preview plot, not like G1.. I don't mind, but found it odd that it doesn't correspond to your commit message :)
[13:34:11] <jepler> it seems to stop loading the program after the first probe
[13:34:48] <jepler> if you change the first 'G38.2' in your program to 'G1', then it also shows something at Y1 .. but nothing at Y2
[13:36:38] <alex_joni> hmm.. think it has something to do with the nature of the code ?
[13:41:59] <jepler> I think I'm not handling INTERP_EXECUTE_FINISH properly
[13:45:01] <alex_joni> bbl
[14:23:34] <jepler> alex_joni: I checked in another fix
[14:42:13] <jepler> hi ray!
[14:50:18] <rayh> Hi Jeff.
[14:50:46] <jepler> what's new?
[14:50:48] <rayh> I see you got an auto make of the docs in html.
[14:50:56] <jepler> yes, I did
[14:51:05] <rayh> I'll download latex2html and try it.
[14:51:13] <jepler> you can take a look at the autoconverted files here:
http://unpy.net/emc2-docs/
[14:51:23] <jepler> not everything is perfect -- for some reason, some of the diagrams are missing
[14:51:32] <jepler> also, the organization is a bit different from the pdf manuals right now
[14:51:33] <rayh> Ah.
[14:51:41] <rayh> That seems to happen.
[14:52:22] <jepler> what have you been up to lately?
[14:52:54] <rayh> Starting a new wood production line for a company.
[14:53:11] <SWPadnos> hi there.
[14:53:22] <SWPadnos> is that the company you bought recently?
[14:54:19] <rayh> Yes
[14:54:24] <SWPadnos> cool
[14:54:28] <rayh> No emc there yet but linux is.
[14:55:29] <SWPadnos> all things in good time
[14:56:09] <jepler> anybody know a good description of the method to extend a count (like encoder counts) from a small width to a large one (e.g., when the 16-bit count rolls over from 65535 to 0, the 32 bit count goes to 65536 instead)?
[14:56:35] <SWPadnos> like the USC driver, or a general term?
[14:57:01] <jepler> I mean in general, though I do have encoders in mind
[14:57:28] <SWPadnos> for what purpose? (end-user documentation, code notes, ...)
[14:57:46] <jepler> so I can write some code that behaves in the same way
[14:57:53] <jepler> I guess I could look at USC
[14:57:55] <SWPadnos> ah, ok.
[14:58:28] <SWPadnos> the method is a little weird, since it looks at the top two bits as bits instead of their numerical value
[14:59:00] <rayh> I haven't used latex2html in a while. tth was my choice but there may be ways to pass args to latex2html
[14:59:07] <rayh> that will help it work.
[14:59:28] <SWPadnos> I think it's easier than that though - if the sign is different from one reading to the next, then add or subtract (depending on which way the change went) MAXVAL for the smaller bit width number
[15:00:00] <SWPadnos> essentially if (rollover) unroll();
[15:00:13] <jepler> rayh: the Makefile has lyx export .tex, then processes it with latex2html with full control over the commandline args.
[15:00:44] <jepler> rayh: the same could be done with tth
[15:00:51] <rayh> Okay. When I get a chance I'll play with the output.
[15:01:06] <jepler> rayh: you have some experience with tth?
[15:01:20] <rayh> I did a couple years ago.
[15:01:35] <rayh> been using pdf since.
[15:01:40] <jepler> rayh: feel free to change it all around, especially if the output is better.
[15:01:53] <rayh> You bet.
[15:02:04] <rayh> It will take me a while to get back into html output.
[15:02:50] <rayh> I have gotten a couple requests for the manuals online in html.
[15:03:19] <rayh> and one for google local search on linuxcnc.org
[15:04:19] <SWPadnos> hmmm - that may be as simple as having a search button that prepends "site:linuxcnc.org" to a search term
[15:04:23] <jepler> 'joomla' doesn't have a built-in search, or it's just not good enough?
[15:05:01] <SWPadnos> but I'm not sure how it would work across the various subdomains (dreamhost, SF, Jeff's basement)
[15:05:41] <rayh> There is a search but the guy couldn't find what he was looking for by using it.
[15:06:07] <SWPadnos> I think the joomla search will only search joomla content, not the whole site
[15:06:27] <SWPadnos> there's not much more on the site ATM, so it isn't a big deal yet
[15:07:42] <rayh> It may well be the content he wanted just wasn't there yet.
[15:08:32] <jepler> http://www.joomla.org/content/view/99/37/
[15:15:19] <SWPadnos> cool
[15:15:48] <SWPadnos> I'm not sure how joomla and mambo modules work - it may be a better question for Alex
[15:16:13] <SWPadnos> I suspect it's similar to the minor issues you sometimes get when using extensions with firefox/mozilla
[15:16:24] <rayh> http://www.wwwsearchsolutions.com/google.php
[15:22:42] <rayh> Catch you guys later. Got a family thing going on.
[15:45:29] <alex_joni> hello
[15:45:38] <alex_joni> * alex_joni is back
[15:51:46] <SWPadnos> yay
[15:54:02] <alex_joni> seen the google search thingie.. I'll try to put it up now
[15:54:07] <SWPadnos> ok
[15:57:49] <alex_joni> jepler: fwiw AXIS is now properly opening programs with G38.2 in it
[15:57:51] <alex_joni> looks great
[15:57:54] <jepler> good
[15:58:05] <alex_joni> even colours and all :)
[15:58:37] <jepler> yes, the backplot gets its own color .. but in the preview, it's white like a regular feed
[15:58:57] <alex_joni> I was wondering.. shouldn't we write a default program for a probing example?
[15:59:04] <jepler> sure we could
[15:59:09] <alex_joni> I'd write one, but I'm not that experienced with O words
[15:59:25] <alex_joni> something like having a few parameters for grid probing
[15:59:48] <alex_joni> x-distance, x-number, y-distance, y-number (distance between probes, and number of probes)
[16:00:01] <alex_joni> then maybe safe Z-height, and Z-probe height
[16:00:06] <alex_joni> then a simple loop
[16:04:40] <alex_joni> ok, I replaced the old search, and tweaked the google site search..
[16:04:44] <alex_joni> how does it look guys?
[16:05:21] <SWPadnos> looks good to me
[16:05:33] <alex_joni> heh... it even searches the wiki
[16:05:36] <alex_joni> which is GREAT
[16:05:39] <SWPadnos> I noticed that
[16:05:44] <SWPadnos> and I agree :)
[16:08:06] <alex_joni> the only thing which bugs me is that it doesn't set the default search to the Domain
[16:08:12] <alex_joni> and I checked that option :(
[16:08:36] <SWPadnos> it doesn't seem to set the default search at all
[16:08:53] <alex_joni> hmm.. actually we got lucky ;) it only has 3 types of possible backgrounds (white, black and gray)
[16:09:00] <SWPadnos> heh
[16:09:10] <alex_joni> which is a moron thing to do..
[16:09:20] <alex_joni> the part there already has a CSS background..
[16:09:31] <alex_joni> luckily the colour matches, or it would have looked awfull
[16:10:02] <SWPadnos> hmmm - should I go with the A64 3500+ for $91 (67 watts), or the A64 3700+ for $102 (89W)
[16:10:21] <SWPadnos> I think for a home server, the 3500 will be plenty :)
[16:10:25] <alex_joni> right..
[16:10:33] <alex_joni> and you'll save 20$ / year in electricity
[16:10:36] <SWPadnos> yep
[16:11:18] <alex_joni> ROFL.. I really love some explanations..
[16:11:24] <alex_joni> check this out: "For example, you may have an Item on 'Breeding Parrots' and you may have another on 'Hand Raising Black Cockatoos'. If you include the keyword 'parrot' in both Items, then the Related Items module will list the 'Breeding Parrots' Item when viewing 'Hand Raising Black Cockatoos' and vice-versa."
[16:11:35] <SWPadnos> heh
[16:15:15] <alex_joni> think it's too much like this?
[16:15:25] <alex_joni> with syndication, and users?
[16:15:45] <SWPadnos> hmmm. it probably is
[16:15:51] <SWPadnos> what are the different RSS feeds?
[16:16:09] <alex_joni> I think different formats for the same thing
[16:16:12] <alex_joni> I'll leave only 2
[16:16:18] <SWPadnos> ok - those are RSS version numbers?
[16:16:43] <alex_joni> yeah
[16:16:59] <alex_joni> better like this?
[16:17:22] <SWPadnos> interesting - those links don't work in Mozilla 1.7.12
[16:17:50] <SWPadnos> yes, though I'm not sure the word "Syndicate" will be meaningful to users
[16:18:09] <SWPadnos> maybe "News Feeds" or something would be better
[16:19:09] <alex_joni> I changed the name, and assigned it only to the About page
[16:20:01] <SWPadnos> ok - that's cool.
[16:20:19] <jepler> wheel hal_vcp is neat
[16:20:20] <alex_joni> likewise the login page is only on Documentation
[16:20:29] <alex_joni> jepler: wheel?
[16:20:38] <SWPadnos> wheee! ??
[16:20:50] <jepler> whee
[16:20:52] <jepler> !
[16:21:04] <alex_joni> SWPadnos: shush.. I was hoping jepler implemented a wheel widget :D
[16:21:10] <SWPadnos> yeah - damn
[16:21:16] <SWPadnos> sorry - my fault
[16:21:24] <alex_joni> jepler: it really is :D
[16:21:27] <jepler> no, I was just using it to create a virtual probe
[16:21:28] <alex_joni> I built a GUI with it..
[16:21:38] <jepler> as I'm sure alex_joni did as well
[16:22:11] <alex_joni> jepler: actually I was just using halcmd -kf :D
[16:22:18] <alex_joni> I'm too lazy to write those 2 lines
[16:22:27] <alex_joni> http://wiki.linuxcnc.org/uploads/haluihalvcp.png
[16:22:37] <alex_joni> it's almost as nice as AXIS don't you think?
[16:22:39] <alex_joni> (lol)
[16:22:45] <jepler> ummm
[16:22:59] <SWPadnos> much more functional than axis
[16:23:03] <alex_joni> don't answer that if you don't agree :D
[16:23:05] <SWPadnos> (at least for those functions ;) )
[16:23:35] <alex_joni> jepler: you (hopefully) know I am kidding
[16:23:45] <SWPadnos> does VCP have the ability to do absolute positioning of elements?
[16:24:09] <alex_joni> SWPadnos: yes, using invisible labels of a certain amount of " "
[16:24:18] <SWPadnos> ie, no
[16:24:21] <alex_joni> right
[16:24:22] <SWPadnos> but you can fake it
[16:24:51] <SWPadnos> ok. do you know if it's because there's no easy way of doing that in gtk, or because "that's the way jmk did it"?
[16:25:01] <alex_joni> I think the later
[16:26:14] <SWPadnos> ok. a graphical UI builder would be much easier to implement if absolute positioning were available (or relative to the parent container)
[16:26:35] <alex_joni> SWPadnos: tell me about it :D
[16:26:42] <SWPadnos> heh
[18:39:56] <alex_joni> any idea why this would show up?
[18:40:05] <alex_joni> EMC2 - 2.0.3
[18:40:05] <alex_joni> Machine configuration directory is '/etc/emc2/sample-configs/stepper/'
[18:40:05] <alex_joni> Machine configuration file is 'stepper_mm.ini'
[18:40:05] <alex_joni> Starting emc...
[18:40:05] <alex_joni> insmod: error inserting '/usr/realtime-2.6.15-magma/modules/emc2/blocks.ko': -1 Unknown symbol in module
[18:42:27] <jepler> alex_joni: what's dmesg say the missing symbols are?
[18:43:17] <jepler> ./bin/halcmd loadrt blocks bogus_flag=1
[18:43:18] <alex_joni> jepler: got it off a forum
[18:43:18] <jepler> insmod: error inserting '/home/jepler/src/emc2/rtlib/blocks.ko': -1 Unknown symbol in module
[18:43:21] <jepler> could be as simple as this ^^^
[18:43:37] <alex_joni> I suspect that might be it.. thanks
[18:43:53] <alex_joni> loadrt blocks ddt=6 hypot=2 wcomp=1
[18:43:54] <jepler> the distributed .hal file for stepper_mm don't seem to use 'blocks' at all
[18:44:04] <alex_joni> no, the guy needed a boost signal
[18:44:14] <alex_joni> to boost the power on the steppers during accel & decel
[18:44:21] <jepler> I see
[18:45:49] <alex_joni> so I suggested soemthing like this:
[18:45:53] <alex_joni> http://pastebin.ca/159955
[18:49:16] <alex_joni> I probably forgot some stuff.. like adding the hypot functions to a thread
[18:49:46] <jepler> I would have to think about it before I understood it
[18:50:27] <alex_joni> I thought about it this way: we already have the XYZspeed in some sim
[18:50:38] <alex_joni> if I ddt it once, I'll get the XYZaccel
[18:50:51] <alex_joni> now only compare that to 0, and you'll have a boost signal
[19:02:38] <danex> * danex is back.
[19:31:26] <danex> * danex is away: Away at the moment
