#linuxcnc-devel | Logs for 2012-03-21

[00:08:22] <JT-Shop> yes it does sound fair, but I assume somewhere down the line a programmer thought it should
[00:11:23] -!- bedah has quit [Quit: Ex-Chat]
[00:13:20] -!- raynerd has quit [Ping timeout: 244 seconds]
[01:02:44] -!- PCW has quit [Ping timeout: 260 seconds]
[01:02:55] PCW_ is now known as PCW
[01:03:05] -!- rob__H has quit [Ping timeout: 260 seconds]
[01:03:25] -!- mhaberler [mhaberler!~mhaberler@] has joined #linuxcnc-devel
[01:06:09] -!- JT-Shop has quit [Quit: ChatZilla [Firefox 10.0.2/20120215223356]]
[01:22:56] <skunkKandT> how can you not have a file loaded?
[01:25:41] -!- skunkKandT has quit [Remote host closed the connection]
[02:02:49] -!- skunkworks__ [skunkworks__!~chatzilla@str-bb-cable-south-3-102.dsl.airstreamcomm.net] has joined #linuxcnc-devel
[03:01:46] -!- skunkworks__ has quit [Ping timeout: 244 seconds]
[03:05:02] -!- mhaberler has quit [Quit: mhaberler]
[03:23:04] -!- demacus has quit [Ping timeout: 245 seconds]
[03:31:18] -!- ries has quit [Quit: ries]
[03:44:29] -!- koax_ [koax_!~kelb@p508D4A9F.dip.t-dialin.net] has joined #linuxcnc-devel
[03:48:04] -!- koax has quit [Ping timeout: 265 seconds]
[04:00:55] -!- SolarNRG has quit []
[04:14:10] -!- phantoxe has quit []
[04:25:01] -!- phillc54 has quit [Quit: ChatZilla [Firefox 11.0/20120312181643]]
[04:28:31] -!- ve7it has quit [Remote host closed the connection]
[04:34:09] <seb_kuzminsky> ok i'm back in the land of the living
[04:35:04] <seb_kuzminsky> so is the consensus that nontrivkins in 2.5 is so broken that it's not worth fixing the strangeness?
[04:35:40] <seb_kuzminsky> I mean with with Axis reading [TRAJ]JOINTS, but the docs not mentioning it and no one ever defining it, and instead defining [TRAJ]AXES
[04:36:15] <seb_kuzminsky> on this (broken) gantry machine I'm working on, in Free mode (Axis "Joint" mode), I can jog the first three joints but not the fourth
[04:36:26] <seb_kuzminsky> if i add [TRAJ]JOINTS = 4 i can jog all four
[04:36:35] <seb_kuzminsky> no big deal i guess...
[04:36:37] <seb_kuzminsky> as you were
[04:54:21] -!- mhaberler [mhaberler!~mhaberler@] has joined #linuxcnc-devel
[04:57:07] -!- AitalMAC has quit [Ping timeout: 245 seconds]
[05:12:20] -!- pcw_ has quit [Ping timeout: 246 seconds]
[05:12:24] -!- pcw__ has quit [Ping timeout: 252 seconds]
[05:22:07] -!- AitalMAC has quit [Ping timeout: 245 seconds]
[05:41:34] <CIA-51> 03cmorley 07v2.5_branch * re2469d4f8dcb 10/src/emc/usr_intf/pncconf/pncconf.py: pncconf -fix live tests with sserial I/O
[05:49:58] -!- mhaberler has quit [Quit: mhaberler]
[06:21:21] -!- vladimirek has quit [Remote host closed the connection]
[06:40:40] -!- maximilian_h [maximilian_h!~bonsai@] has joined #linuxcnc-devel
[06:51:33] -!- psha[work] [psha[work]!~psha@] has joined #linuxcnc-devel
[07:01:52] -!- Valen has quit [Quit: Leaving.]
[07:46:38] <alex_joni> seb_kuzminsky: a lot more things are broken for nontrivkins
[07:46:50] <alex_joni> but with a gantry you won't run into them (luckily for you)
[07:47:09] <alex_joni> like: - vel and accel limits are set for axes (but should be for joints)
[07:47:13] <alex_joni> etc
[07:58:17] -!- maximilian_h has quit [Quit: Leaving.]
[08:15:42] -!- Valen has quit [Quit: Leaving.]
[08:41:17] -!- DJ9DJ has quit [Quit: bye]
[08:45:20] -!- rob_h [rob_h!~rob_h@5ace705c.bb.sky.com] has joined #linuxcnc-devel
[08:58:59] -!- Valen has quit [Quit: Leaving.]
[09:39:56] -!- elmo40 has quit [Ping timeout: 255 seconds]
[09:41:52] -!- logger[mah] has quit [Ping timeout: 265 seconds]
[09:43:09] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[10:15:39] -!- mhaberler [mhaberler!~mhaberler@] has joined #linuxcnc-devel
[10:48:36] -!- logger[mah] [logger[mah]!~loggermah@burgi.mah.priv.at] has joined #linuxcnc-devel
[11:19:26] -!- mhaberler has quit [Quit: mhaberler]
[11:32:52] -!- DJ9DJ has quit [Changing host]
[12:12:28] -!- vladimirek has quit [Remote host closed the connection]
[12:43:07] <jepler> seb_kuzminsky: to the contrary, we're eagerly looking forward to a whole slew of gantry-related fixes from you
[12:44:15] <alex_joni> heh
[12:51:29] -!- bedah has quit [Quit: quit]
[13:02:38] <cradek> seb_kuzminsky: I can't see any problem with adding that to the 2.5 documentation. or are you suggesting something else?
[13:24:38] -!- psha[work] has quit [Quit: Lost terminal]
[13:26:17] <alex_joni> cradek: hi, sent you a question ;)
[13:26:48] -!- skunkworks [skunkworks!~skunkwork@68-115-41-210.static.eucl.wi.charter.com] has joined #linuxcnc-devel
[13:27:29] -!- JT-Shop [JT-Shop!~chatzilla@] has joined #linuxcnc-devel
[13:37:49] -!- Valen has quit [Quit: Leaving.]
[13:45:53] -!- JT-Shop has quit [Quit: ChatZilla [Firefox 10.0.2/20120215223356]]
[14:02:08] -!- SWPadnos_ has quit [Quit: ChatZilla [Firefox 10.0.2/20120215223356]]
[14:02:14] -!- vin321 has quit [Ping timeout: 246 seconds]
[14:04:58] -!- Loetmichel has quit [Disconnected by services]
[14:05:00] Cylly is now known as Loetmichel
[15:07:46] -!- psha [psha!~psha@] has joined #linuxcnc-devel
[15:24:17] -!- mhaberler [mhaberler!~mhaberler@] has joined #linuxcnc-devel
[15:38:17] -!- SolarNRG has quit []
[15:59:57] <CIA-51> 03seb 07v2.5_branch * r06a5cd518008 10/docs/src/config/ (5 files): docs: document Axis' [TRAJ]JOINTS ini config variable
[15:59:58] <CIA-51> 03seb 07v2.5_branch * r4aec4f9c16c9 10/docs/src/motion/ (5 files): docs: fix some markup, and remove some trailing white space
[16:09:17] -!- vin321 has quit [Ping timeout: 246 seconds]
[16:10:19] <CIA-51> 03seb 07v2.5_branch * r873725da6b65 10/docs/src/hal/comp.txt: docs: expand the list of forbidden symbold names
[16:10:19] <CIA-51> 03seb 07v2.5_branch * r23e1e92c618a 10/ (docs/src/hal/comp.txt src/hal/utils/comp.g): comp: forbid use of reserved variable names
[16:10:19] <CIA-51> 03seb 07v2.5_branch * r8cf401cb5a3d 10/ (docs/src/hal/comp.txt src/hal/utils/comp.g): comp: hide the 'struct state' type behind a '__comp_' prefix
[16:10:20] <CIA-51> 03seb 07v2.5_branch * ra7a7fd39ac97 10/ (docs/src/hal/comp.txt src/hal/utils/comp.g): comp: hide the internal variables inst, first_inst, and last_inst behind the __comp_ prefix
[16:10:20] <CIA-51> 03seb 07v2.5_branch * rd0c9b139a620 10/ (docs/src/hal/comp.txt src/hal/utils/comp.g): comp: hide get_data_size() behind the __comp_ prefix
[16:10:21] <CIA-51> 03seb 07v2.5_branch * r34c36e26d953 10/docs/src/hal/comp.txt: docs: warn about comp's private __comp_ namespace
[16:10:32] <CIA-51> 03seb 07v2.5_branch * r8ce878245543 10/ (docs/src/hal/comp.txt src/hal/utils/comp.g): comp: don't let the user override comp_id
[16:14:50] -!- ve7it [ve7it!~LawrenceG@S0106001c10b7770f.pk.shawcable.net] has joined #linuxcnc-devel
[16:26:45] -!- durer has quit [Client Quit]
[17:24:22] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[17:51:22] -!- JT-Shop [JT-Shop!~chatzilla@] has joined #linuxcnc-devel
[18:01:35] -!- sumpfralle has quit [Ping timeout: 260 seconds]
[18:09:42] -!- isssy has quit [Quit: Bye Bye]
[18:09:49] <JT-Shop> Is there a reason that PWM mode 1 gives a negative output with M4? It has a direction pin so should the PWM be positive for both M3 and M4?
[18:25:06] -!- mhaberler has quit [Ping timeout: 272 seconds]
[18:32:56] -!- JT-Shop has quit [Ping timeout: 255 seconds]
[18:37:45] -!- jv4779 has quit []
[18:43:47] -!- mhaberler [mhaberler!~mhaberler@] has joined #linuxcnc-devel
[18:45:12] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[18:47:02] -!- erasmo [erasmo!~erasmo@] has joined #linuxcnc-devel
[18:48:16] -!- IchGuckLive has quit [Read error: Operation timed out]
[18:51:19] -!- jthornton [jthornton!~chatzilla@] has joined #linuxcnc-devel
[18:52:17] -!- isssy has quit [Quit: Bye Bye]
[18:56:55] -!- Gast200 has quit [Quit: Bye Bye]
[18:58:23] -!- pcw__ has quit [Ping timeout: 245 seconds]
[18:58:45] -!- pcw_ has quit [Ping timeout: 244 seconds]
[19:00:04] <cradek> mhaberler: > Another glaring defect of line numbers in 'MDI mode' is that they are not idempotent, that is, reexecution of a statement does *not* yield identical line numbers for the same source location
[19:00:20] <mhaberler> yes?
[19:00:27] <cradek> are you saying that if the user does the same mdi command twice, you want matching slocs?
[19:01:01] <mhaberler> slocs, which dont necessarily mean 'execution context'
[19:01:05] <jepler> jthornton: pwmgen output_type=1 is behaving roughly like I expect
[19:01:12] -!- jthornton has quit [Ping timeout: 272 seconds]
[19:01:19] <jepler> jthornton: if I set .value = .1 then .pwm is true about 10% of the time and false about 90% of the time
[19:01:28] <jepler> and dir is false
[19:01:38] <jepler> if I set .value = -.1 then .pwm is still true about 10% of the time, and dir is true
[19:01:41] <mhaberler> I'd be interested where this 'increasing pseudoMdiLineNumber' stuff comes from (why)
[19:01:57] <cradek> I don't understand any of what you are saying about rfl in mdi etc
[19:02:22] <GoSebGo> jthornton: are you using sw pwmgen? Or mesa, or something else?
[19:02:24] <mhaberler> image you had a bit more than you currently have, namely:
[19:02:49] <cradek> my understanding is that currently the 'line number' in mdi isn't used for anything but it's important that each one is different
[19:03:10] <mhaberler> you execute 'o<foo>call' but tell it to start from line x, not from the very call
[19:03:30] <mhaberler> you currently cant do that in axis for instance, but it is conceivable
[19:03:55] <cradek> you mean something like run from line x of the file containing foo?
[19:04:02] <mhaberler> yes
[19:04:08] <mhaberler> do you know *why* they need to be different between MDI executions?
[19:04:41] <cradek> step needs them to be different to work; I know you can pause and resume mdi, not sure if you can step it currently
[19:04:46] <mhaberler> its not an important use case (at least for me) but the current impl would break it
[19:05:08] <mhaberler> ok, step line.
[19:05:09] <cradek> yes rfl from the middle of an O-call is pretty esoteric but I do see where it could be possible
[19:06:15] <cradek> but say foo is recursive and you want to rf the 10th line after you get to the 7th level
[19:06:36] <cradek> so I wonder how esoteric is too esoteric to worry about
[19:07:28] <mhaberler> well, drop rs274ngc and plug in python, and you have lots of granularity and debugger support for conditional breakpoints (which is another way of looking at RFL)
[19:08:00] <mhaberler> RFL is breakpoints101;)
[19:08:04] <jepler> all my archaeology has told me about pseudoMdiLineNumber is that it's always been in emc2
[19:08:21] <mhaberler> so its been broken a looong time
[19:08:38] <cradek> might I gently suggest that you add your examples to the "The issue" part: it kind of reads like a rant and I didn't understand exactly what the rant was about
[19:09:02] <mhaberler> well, not broken - just not suited to several source contexts
[19:09:14] <mhaberler> ok, I'll have a look - it might benefit from examples
[19:10:41] <cradek> that part was designed when code execution was linear - it was entirely reasonable in that situation - I feel defensive for the folks who designed it when I read your description which would be better if it were more emotionally neutral and technical
[19:10:47] <mhaberler> right
[19:10:52] <jepler> it was added in "emc1" is 2001 without much explanation http://git.unpy.net/view?p=emc.git;a=patch;h=d93eabe9b05fca5ed3f760a742f699ecdc84185e
[19:11:49] <mhaberler> well, semi. The negative MDI linenumber was the initial kludge
[19:12:13] <mhaberler> that introduced a data dependency
[19:12:46] <cradek> I don't know what that means
[19:14:00] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.87 [Firefox 10.0.2/20120216080748]]
[19:14:25] <mhaberler> what should have been done: dont touch the line number, separare out mode into a separate value; currently 'mode' is implied by the sign which is the data dependency
[19:14:47] <mhaberler> mode really is 'source id'
[19:14:54] <mhaberler> (string, file)
[19:15:41] <cradek> there was only one file, the open one
[19:16:01] <mhaberler> and the string passed as MDI which was the second one, hence the sign
[19:16:14] <cradek> yes
[19:16:49] <cradek> half the integers were used up by possible file lines, so they used the other half for the other source
[19:17:29] <mhaberler> I really havent drilled down yet to as why pseudoMdiLineNumbers need to be increasing
[19:17:44] <mhaberler> ambigous canon/motion entries?
[19:18:31] <cradek> I suggested a reason earlier - but they don't have to be increasing, they have to be different
[19:19:01] <cradek> increasing is just the easiest way to make them all different
[19:19:10] <mhaberler> ok, thought experiment
[19:19:40] <mhaberler> lets assume pseudoMdiLineNumber is reset before each MDI command
[19:19:45] <mhaberler> what would break?
[19:20:23] <cradek> like I said earlier, stepping through them (IF that currently works)
[19:20:42] <cradek> there might be other things? I dunno.
[19:20:55] <mhaberler> Hm, I need to try stepping MDI
[19:21:18] <cradek> I'm not sure AXIS supports pause/resume during mdi, but touchy sure does
[19:21:28] <cradek> I'm not sure touchy supports stepping during mdi
[19:22:30] <mhaberler> one scenario I could see: you queue several MDI commands, so you fill the canon queue with repeating sequences of line numbers - maybe that causes confusion downstream; but then you could execute a file several times and that shouldnt break either
[19:24:49] <jepler> I'm pretty sure that in emcTaskPlan when in EMC_TASK_MODE_MDI, EMC_TASK_PLAN_STEP_TYPE is handled by the default case, so that it's not even permitted to issue a step nml message during mdi
[19:24:53] -!- JT-Shop [JT-Shop!~chatzilla@] has joined #linuxcnc-devel
[19:25:11] <cradek> ok, then it'd be nice if that worked :-)
[19:25:18] <mhaberler> uh
[19:26:19] <mhaberler> well, my take on MDI is - I an not sure why its there in the first place - all I see is a different way of passing input to the interp (string vs file) and I dont quite see how that needs to be executed differently in task
[19:26:22] <jepler> is my reading wrong? there are a lot of lines there so it's quite possible I lost my place
[19:26:56] <mhaberler> if you can step a file, why cant you step a multi-line string? beats me.
[19:26:57] <cradek> I don't know and the only way I'd venture a guess is to try it
[19:33:15] <mhaberler> is stepping a canon/interplist affair, or motion queue stepping?
[19:33:35] <cradek> there are both types unfortunately
[19:33:39] <mhaberler> duh
[19:34:02] <mhaberler> but the arrow button in axis is canonq stepping I guess?
[19:34:35] <cradek> AXIS doesn't know one way or the other - which you get depends how you got paused
[19:34:55] <cradek> you can pause with the motion queue empty or full
[19:35:00] <mhaberler> oh.. so its interpreted in task
[19:35:02] <cradek> if it's full it uses the line numbers to step
[19:35:04] <cradek> yes
[19:35:09] <mhaberler> thanks
[19:35:53] <mhaberler> sorry to be an inquisitive idiot, but what's the point of having two schemes? I see some stuff which doesnt reach motion, but then?
[19:36:36] <cradek> step should ideally step forward one line of gcode
[19:37:04] <cradek> but if you've queued up a bunch of motion and then pause, many of those lines are gone
[19:37:29] <mhaberler> oh.. because of the two queues, the first might be drained
[19:37:57] -!- erasmo has quit [Ping timeout: 248 seconds]
[19:38:11] <mhaberler> so by the time youd want to break on a canon lineno, the canonq is empty
[19:38:19] <mhaberler> (in some cases)
[19:38:19] -!- erasmo [erasmo!~erasmo@] has joined #linuxcnc-devel
[19:38:34] <JT-Shop> while the brain trust is here is PWM mode 1 correct or should it output positive for both directions?
[19:40:46] <mhaberler> http://ircamera.as.arizona.edu/NatSci102/NatSci102/images/extbrain.htm
[19:41:37] <mhaberler> without any success on this question, I have to add
[19:46:05] <jepler> JT-Shop: I looked at the behavior and it made sense to me. I guess you were gone when I answered.
[19:46:08] <jepler> logger[mah]: url
[19:46:08] <logger[mah]> jepler: Log stored at http://linuxcnc.mah.priv.at/irc/%23linuxcnc-devel/2012-03-21.html
[19:46:30] <JT-Shop> yes, I'm on flaky dial up
[19:47:10] <JT-Shop> jepler: seems most drives that take a direction pin only use positive PWM
[19:47:31] <jepler> I must not understand what you mean when you say "positive PWM"
[19:48:06] <JT-Shop> one second let me get the post
[19:48:29] <JT-Shop> http://linuxcnc.org/index.php/english/component/kunena/?func=view&catid=38&id=18657&limit=6&start=6#18687
[19:48:34] -!- vin321 has quit [Read error: Connection reset by peer]
[19:48:39] <jepler> looking
[19:48:44] <JT-Shop> thanks
[19:49:21] <JT-Shop> this seems to come up with some regularity on the forum
[19:49:38] <jepler> that thread is about stepconf
[19:50:29] <JT-Shop> part is and part is about PWM mode 1
[19:50:49] <JT-Shop> Is it really ABS thats needed or the proper PWM mode?
[19:50:51] <JT-Shop> PWM mode 1 should do the right thing without adding any other comps
[19:50:53] <JT-Shop> if PWM mode 1 does not do the right thing, the PWM comp needs to be fixed
[19:51:27] <JT-Shop> that was a quote from the last post
[19:52:11] <JT-Shop> brb, gotta fill some growlers
[19:53:45] <mhaberler> one can step a 'o<sub> call' in Axis if the sub body contains an M0, and that seems to work thereafter
[19:54:17] <mhaberler> typing an MDI command and hitting 'step' steps the loaded program
[19:55:01] -!- cmorley has quit [Ping timeout: 272 seconds]
[19:59:48] <mhaberler> nope, false alarm - step doesnt work that way. unly unpause works that way
[20:02:08] -!- automata_ has quit [Ping timeout: 240 seconds]
[20:05:59] <mhaberler> anyway, so from requirements 'stop on next source location' is missing
[20:08:11] -!- psha has quit [Quit: Lost terminal]
[20:08:33] -!- cmorley [cmorley!~chris@d64-180-207-117.bchsia.telus.net] has joined #linuxcnc-devel
[20:10:17] <jepler> JT-Shop: since I still don't think I understand the question, all I can do is reiterate that pwmgen output_mode=1 seems to be doing what I expect it to do
[20:10:42] <jepler> if .value is negative then 'dir' is 1 and if it's positive then 'dir' is 0
[20:11:01] <jepler> if .value is +.1 or -.1 (for the default .scale of 1) then 'pwm' is true about 10% of the time
[20:12:04] <jepler> an example of a device where this is appropriate is an L298 full-bridge driver, if 'dir' is hooked to in1, '!dir' (inverted dir) is hookd to in2, and 'pwm' is hooked to 'ena'
[20:12:39] <erasmo> helo everyone
[20:12:53] <erasmo> I would like to join linuxdevel team
[20:13:31] <jepler> hello
[20:13:34] <jepler> what do you intend to work on?
[20:13:37] <erasmo> I have experience in C C++ and C# a litle in python
[20:14:07] <JT-Shop> jepler: thanks for looking at it
[20:14:31] <erasmo> I dont know is there someone who could do good use of me...
[20:14:39] <erasmo> ill bet there is
[20:15:52] <JT-Shop> I wonder why they need to use abs when the pwm is just off and on?
[20:16:13] <erasmo> I can work on hardware/software...
[20:16:40] <jepler> generally, linuxcnc contributors work on things that they choose themselves
[20:16:48] <GoSebGo> Erasmo: maybe take a look at our bug tracker on sourceforge and see if anything looks interesting
[20:17:06] <jepler> when you have some questions to ask, you can ask them on irc or the mailing lists
[20:17:22] <jepler> and when you have some interesting work to share, you can do it the same way (git has commands for helping you do this, like 'git format-patch')
[20:17:55] <GoSebGo> Or send a link to a branch on a public repo
[20:18:02] <jepler> yes, we'd be happy to see known bugs in the software resolved.
[20:18:19] -!- raynerd has quit [Ping timeout: 260 seconds]
[20:18:39] <GoSebGo> And to see more active developers :-)
[20:18:45] -!- raynerd has quit [Client Quit]
[20:19:33] <erasmo> ok I'll try the bug tracker and ill see what is interesting for me there :) thanks for help
[20:20:16] <jepler> feel free to ask questions -- the more specific, the better.
[20:27:04] -!- automata__ has quit [Ping timeout: 260 seconds]
[20:29:23] <GoSebGo> Erasmo: hej, du ar ju svensk! Valkommen!
[20:32:10] <jepler> I hope somebody writes a verilog synthesizer for minecraft soon, so that people don't have to keep hand-designing their minecraft scientific calculators. http://www.metafilter.com/114058/As-you-can-probably-imagine-this-took-some-effort-to-make
[20:33:38] <GoSebGo> Boy, me too
[20:33:59] * GoSebGo looks askance at jeplet
[20:34:18] <cradek> v2.4_branch fix for jv4779's bug, needs review: http://timeguy.com/cradek-files/emc/0001-Fix-inappropriate-UVW-movement-during-certain-arcs.patch
[20:39:12] <jepler> cradek: 6e29f79f926e8d9410d17d2d844ae96f2851a76f
[20:39:49] <cradek> wait what
[20:41:35] <cradek> what did I do that makes this not in 2.4?
[20:41:43] <GoSebGo> Wow that commit removed some goofy-looking code
[20:42:54] <cradek> jepler: oh I bet this means "XXX rotation" is really an XXX
[20:43:09] <cradek> but that's ok, rotation is all busty in 2.4 anyway
[20:43:22] <GoSebGo> That commit is in 2.5
[20:43:33] <jepler> yes, that commit is in 2.5 and not 2.4
[20:43:44] <cradek> a little piece of it should have been in 2.4
[20:43:46] <erasmo> I'm just starting emc adventure could somebody help me choose bug at http://sourceforge.net/tracker/?group_id=6744&atid=106744
[20:43:57] <jepler> 318fbd9a00e9cfc306b2ff53b2d7531ebb493313 also in 2.5 but not 2.4 was the one that had temporarily disabled arc to line conversion
[20:44:02] <erasmo> at which I could work on?
[20:44:13] <jepler> so .. the fact that you've found essentially the same fix twice is positive, I guess
[20:44:18] <cradek> sigh
[20:44:38] <skunkworks> heh
[20:44:49] <cradek> so I and I agree on the fix, that's good enough for me/us
[20:45:22] <jepler> cradek: besides sticking it on 2.4 you should probably go ahead and do a 2.4->2.5 merge because it'll probably have a conflict right there
[20:45:30] <cradek> sure
[20:45:49] <GoSebGo> Erasmo: i suggest starting with something easy, maybe 3442326
[20:47:01] <skunkworks> I agree with cradek and cradek
[20:47:13] <erasmo> Doesn't somebody is currently working on 3442326
[20:47:16] <erasmo> ??
[20:48:24] <GoSebGo> It's got no activity, so i think not
[20:50:22] <CIA-51> 03cradek 07v2.4_branch * r12a81317f063 10/src/emc/task/emccanon.cc: Fix inappropriate UVW movement during certain arcs
[20:58:28] <jepler> cradek: thanks for that
[20:58:29] <mhaberler> not me
[20:59:36] <CIA-51> 03cradek 07master * r7a84454ed4b8 10/ (65 files in 15 dirs): Merge branch 'v2.5_branch'
[20:59:36] <CIA-51> 03cradek 07master * rf44675698b51 10/docs/src/config/ini_config.txt: Fix minor markup error
[20:59:40] <mhaberler> kirkw brings up an interesting subject wrt line numbers. Are Nxxxx line numbers anything else than syntactic noise currently?
[20:59:51] <cradek> just noise
[21:00:07] <cradek> they are carefully saved in the gcode block in the interp, and never used
[21:00:44] <mhaberler> with a first-class burial at the next line in init_block(), I assume
[21:01:58] <cradek> some gcode dialects use it for something, but ours doesn't
[21:02:07] <cradek> anything past gcode can't possibly care
[21:02:49] <mhaberler> so then it's not worth carrying on in an sloc
[21:03:31] <cradek> I don't feel qualified to agree or disagree
[21:25:25] -!- FinboySlick has quit [Quit: Leaving.]
[21:30:54] -!- skunkworks has quit []
[21:35:57] -!- Thetawaves has quit [Quit: This computer has gone to sleep]
[21:36:31] -!- DJ9DJ has quit [Quit: bye]
[21:43:17] -!- syyl_ws has quit [Remote host closed the connection]
[21:44:53] -!- vladimirek has quit [Remote host closed the connection]
[21:58:04] -!- raynerd has quit [Ping timeout: 260 seconds]
[22:15:21] <erasmo> Goodnight everyone.
[22:15:43] -!- erasmo has quit [Quit: Leaving]
[22:17:26] -!- sumpfralle has quit [Ping timeout: 252 seconds]
[22:41:14] <mhaberler> I think a more versatile approach to RFL would be actually generating the start condition in the interpreter, instead of trying to figure once the interpreter has done its job - a start condition could be an aribtrary expression, not just a line number
[22:41:29] -!- raynerd has quit [Ping timeout: 265 seconds]
[22:42:13] -!- vin321 has quit [K-Lined]
[22:42:24] <mhaberler> that would for instance allow starting inside a loop for a given loopvar value
[22:47:14] -!- factor has quit [Quit: Leaving]
[22:49:11] <rob_h> N numbers on other controls are used for cycles.. ie area clearance, where you give it a number to defind the profile between... or looping on fanuc.. M99 P(number here to loop back to)
[22:49:37] <mhaberler> a, like labels
[22:50:15] <rob_h> yes could say that
[22:53:05] -!- Fox_Muldr has quit [Ping timeout: 265 seconds]
[22:56:44] -!- GoSebGo has quit [Quit: Bye]
[22:56:59] -!- GoSebGo [GoSebGo!~GoSebGo@70-3-248-13.pools.spcsdns.net] has joined #linuxcnc-devel
[22:58:16] -!- GoSebGo has quit [Client Quit]
[22:58:38] -!- GoSebGo [GoSebGo!~GoSebGo@70-3-248-13.pools.spcsdns.net] has joined #linuxcnc-devel
[22:59:43] -!- Valen has quit [Quit: Leaving.]
[23:00:53] -!- syyl has quit [Quit: Leaving]
[23:10:47] -!- GoSebGo has quit [Quit: Bye]
[23:11:00] -!- GoSebGo [GoSebGo!~GoSebGo@70-3-248-13.pools.spcsdns.net] has joined #linuxcnc-devel
[23:17:54] -!- joe9 has quit [Quit: leaving]
[23:18:33] -!- SolarNRG has quit []
[23:29:38] -!- ve7it has quit [Ping timeout: 240 seconds]
[23:49:47] pjm__ is now known as pjm
[23:50:17] pjm is now known as Guest35897
[23:51:14] -!- Guest35897 has quit [Quit: TTFO]
[23:56:51] -!- KimK has quit [Quit: Leaving]
[23:58:43] -!- raynerd has quit [Ping timeout: 260 seconds]