#emc-devel | Logs for 2009-06-22

[04:36:01] <CIA-2> EMC: 03chrisinnanaimo 07master * ra37b61a30f86 10/src/emc/usr_intf/pncconf/pncconf.glade: Add help window
[04:57:29] <CIA-2> EMC: 03chris 07master * r6e9cddd6c921 10/src/emc/rs274ngc/interp_internal.hh: Fix buffer overflow when trying to report error messages.
[04:57:30] <CIA-2> EMC: 03chris 07master * re29353c39c1f 10/src/emc/usr_intf/pncconf/pncconf.glade: Merge branch 'master' of ssh://cradek@git.linuxcnc.org/git/emc2
[05:00:51] <cradek> I don't understand why I keep getting these merges
[05:01:34] <cradek> I try to push, it errors because it's out of date (I guess), so I pull then push, and then I get two commits, one being a merge
[05:01:38] <cradek> shoudl I be using rebase?
[05:01:42] <cradek> jepler: ^
[07:51:32] <micges1> micges1 is now known as micges_plasma
[08:11:06] <christel> [Global Notice] Hi all, I'm going to restart services -- you'll be without NickServ, ChanServ, MemoServ etc for a few moments! Sorry for the inconvenience and thank you for using freenode.
[11:57:49] <jepler> cradek_: yes, if you did a 'git rebase origin/master' after the pull, you'd end up with a linear history instead of a merge
[12:10:01] <jepler> huh, I didn't know we had M61, set current tool number
[12:10:04] <jepler> how long's that been there?
[12:12:48] <micges1> 5 months ?
[12:39:12] <SWPadnos> ugh. does the forum default to sending emails when there are replies?
[13:07:51] <alex_joni> SWPadnos: there's a checkbox when you submit a reply
[13:08:03] <SWPadnos> and it defaults to on it seems
[13:08:10] <alex_joni> it seems ;)
[13:11:25] <cradek_> haha, I like Terry's mail about tool changes and broken tools
[13:11:48] <cradek_> cradek_ is now known as cradek
[13:16:52] <cradek> jepler: I guess what really does happen is a divergence and merge, since we both made separate changes based on an old version.
[13:39:50] <alex_joni> jepler: http://git.linuxcnc.org/gitweb?p=emc2.git;a=commitdiff;h=b7dc6a25e57d025b06d2254db8aa8b24be6acd32
[14:24:10] <CIA-3> EMC: 03jepler 07master * r840bf51bcc90 10/configs/hm2-servo/hm2-servo.hal: Use dac scale from inifile
[14:25:52] <CIA-3> EMC: 03jepler 07v2_3_branch * r40e00830cf13 10/tcl/bin/emccalib.tcl: paper over a problem for hm2-style inifiles where the parameter being set is the result of an inifile substitution
[14:25:53] <CIA-3> EMC: 03jepler 07v2_3_branch * r79f87fe541c8 10/configs/hm2-servo/hm2-servo.hal: Use dac scale from inifile
[14:26:15] <alex_joni> sounds like quite a couple changes on 2.3
[14:26:48] <jepler> turns out I want emccalib for myself now, and used some of the "forbidden" hal syntax that broke it
[14:27:13] <jepler> but, yeah, 2.3.2 needs to be made, and soon
[14:27:25] <alex_joni> did seb commit the changes he meant?
[14:27:30] <alex_joni> some stepgen fixes for hm2
[14:28:31] <jepler> I think so -- 09ed462
[14:32:40] <cradek> I wonder if that function name stack is even used. I thought there used to be a way to get an error message showing the call stack, but I think that comes from calling Interp::stack_name, which nothing does
[14:33:52] <SWPadnos> hmmm. is it possible (or easy enough) to add a "full file" link to the commit emails?
[14:34:05] <cradek> there is no file
[14:34:31] <SWPadnos> sure there is: http://git.linuxcnc.org/gitweb?p=emc2.git;a=blob;f=configs/hm2-servo/hm2-servo.hal;h=149b40ce0134e8c3879ad948f5829afcebeb5312;hb=840bf51
[14:34:37] <SWPadnos> "blob"
[14:35:00] <cradek> you mean for each file that is changed?
[14:35:05] <SWPadnos> hmmm
[14:35:30] <SWPadnos> oh, the diffs are for multiple files if necessary
[14:35:41] <SWPadnos> not "separate" diffs like cvs had
[14:35:56] <cradek> if you follow the link in the email, each file that is changed is one only link away
[14:36:13] <SWPadnos> yeah, that's how I got that blob link :)
[14:36:18] <SWPadnos> that's fine, thanks
[15:43:47] <alex_joni> heh, seems les wants to shoot jepler
[15:44:24] <alex_joni> otoh, I must wonder how hard it is to remember that s stands for status, and c for commands
[15:44:59] <jepler> huh?
[15:52:22] <micges> haha
[17:26:34] <alex_joni> jepler: "Two hours learning Axis and Python, 10 minutes of actual coding. I have to say, Axis is not the easiest piece of code to understand. Whoever decided to use single letters for important variables should have been shot. c.auto(emc.AUTO_PAUSE) is not nearly as easy to follow as something like emcCmd.auto(emc.AUTO_PAUSE)."
[18:03:06] <cradek> anyone who can add a feature to an unfamiliar piece of software in 2h10 is doing great
[18:04:38] <skunkworks_> That would be a great reply :)
[18:16:30] <jepler> alex_joni: hmm .. it's just as well I missed that message the first time
[18:17:58] <jepler> it's not enough that I wrote you AXIS? I have to write it in the coding style you prefer, too?
[18:18:04] <jepler> </argh>
[18:26:06] <alex_joni> maybe you should point to sed ;)
[18:26:22] <lerman> We should arrange for him to be shot at the next Fest. (By then, he'll probably be finished with the next set of features we'd like.) :-) :-)
[18:26:31] <alex_joni> lerman: lol
[18:28:27] <skunkworks_> wait - you guys can do that? (I don't know if I should come anymore...)
[18:28:29] <jepler> nah, actually I'll be just 5% done with the AXIS replacement that will be written in C# and nice and modular and with a screen designer to make it easy for beginning novices to modify.
[18:28:34] <lerman> By that time, what's his name will have rewritten axis anyway.
[18:28:49] <lerman> Drat. You beat me to it.
[18:29:24] <alex_joni> wtf.. bucket-printer anyone? http://www.usa.canon.com/consumer/controller?act=ModelInfoAct&fcategoryid=118&modelid=17141
[18:30:01] <jepler> that's pretty special
[18:30:06] <cradek> someone should make a combination printer/shredder. if you don't pick up your printout in five minutes, it shreds it
[18:30:13] <jepler> excellent!
[18:30:19] <alex_joni> cradek: maybe scanner/shredder ?
[18:30:20] <skunkworks_> I thought it was a printer to print on buckets :(
[18:38:37] <skunkworks_> jepler: we will be needing a video. :)
[18:38:51] <jepler> skunkworks_: of zenbot? not much to see.
[18:42:24] <skunkworks_> jepler: anything servo is cool though...
[18:42:39] <skunkworks_> I think so anyway
[18:49:28] <jepler> you'll have to make do with the pictures for now
[18:49:49] <alex_joni> pictures are cooler
[18:58:41] <skunkworks_> :)
[18:59:04] <jepler> alex_joni: does it do you any good to see a quarter next to the motor mount, to understand how big it is?
[19:00:16] <jepler> (or, in this case, how small..)
[19:01:18] <alex_joni> I think I got a sense from the overall machine size
[19:01:22] <alex_joni> and from the motor size
[19:01:28] <alex_joni> and from the screw sizes ;)
[19:01:43] <alex_joni> otoh, I only somehow know how big a quarter is ;)
[19:15:10] <BJT-Work> how about a 50 groszy coin?
[19:16:27] <jepler> looks like a 1 złoty coin is closer in size to the US quarter
[19:16:36] <jepler> 23mm vs 24.26mm
[19:17:39] <BJT-Work> I was just guessing as I didn't have a quarter in my pocket :)
[19:17:49] <jepler> me either, but I do have wikipedia
[19:18:19] <BJT-Work> * BJT-Work wanders back to the mill to make chips or sparks whatever comes first
[19:18:36] <BJT-Work> I couldn't find the quarter on wikipedia
[19:18:37] <jepler> my favorite wikipedia page: http://en.wikipedia.org/wiki/List_of_lists_of_lists
[19:18:49] <jepler> BJT-Work: http://en.wikipedia.org/wiki/Quarter_(U.S._coin)
[19:28:41] <CIA-3> EMC: 03cradek 07master * r0901fd6ffc09 10/src/emc/rs274ngc/rs274ngc_pre.cc: Use the new define for the stack length
[19:47:31] <alex_joni> jepler: lol
[20:50:42] <CIA-3> EMC: 03jepler 07v2_3_branch * rb6ebdea90852 10/src/ (configure configure.in): fix './configure.lineno: 6816: source: not found'
[20:50:50] <CIA-3> EMC: 03chris 07v2_3_branch * r4048fa9d1c16 10/src/ (configure configure.in): make sure python-tcl and tcl versions match
[20:51:07] <CIA-3> EMC: 03jepler 07master * reaa34efb63ca 10/src/hal/drivers/mesa-hostmot2/ioport.c: allow all pins, including special-purpose output pins, to be read as inputs
[20:51:07] <jepler> now what's weird is that it looks like cradek just made a checkin in v2_3_branch, but he didn't..
[20:51:47] <cradek> was that yours?
[20:51:52] <SWPadnos> oh - did you backport something of his?
[20:52:12] <cradek> Author: Chris Radek <chris@timeguy.com> 2009-05-22 15:35:31
[20:52:13] <cradek> Committer: Jeff Epler <jepler@unpythonic.net> 2009-06-22 15:46:33
[20:52:18] <jepler> SWPadnos: yes, I cherry-picked a commit that says it was authored by chris
[20:52:25] <cradek> oh, so it's a feature
[20:52:27] <SWPadnos> heh
[20:52:27] <jepler> (I think it turns out I'm the one who actually wrote it, but whatever..)
[20:52:32] <SWPadnos> heh
[20:52:36] <SWPadnos> again
[20:52:46] <jepler> (it was part of the 9.04-related fixes at fest)
[20:54:04] <SWPadnos> oh - I wanted to ask about the commit from chris morley (I think) that had "chrisinnanaimo" as the author
[20:54:17] <alex_joni> that's his email addy
[20:54:20] <alex_joni> from the commit
[20:54:33] <SWPadnos> it wasn't on the other commits done at that time
[20:54:42] <alex_joni> probably didn't set up the user on that machine
[20:54:59] <SWPadnos> oh, I guess that's because it was another chris :)
[20:55:41] <alex_joni> the others were by cradek
[20:55:43] <jepler> what's the question?
[20:55:49] <SWPadnos> nothing, nevermind
[20:56:38] <jepler> the confusing thing here is that cradek puts something other than "cradek" before the @ in his e-mail address
[20:56:56] <SWPadnos> yes, that is confusing
[20:57:14] <cradek> I don't like it, but I also don't know what to do
[20:57:23] <cradek> you have to guess an author for CIA somehow
[20:59:10] <jepler> just imagine if we all insisted on using git@ourvanitydomainname.example.com ...
[21:00:26] <SWPadnos> "My Long Name (is really long)" <slightly.shorter.version@my_unique_and_nearly_too_long_for_RFC_compliance.sillydomainnames.com>
[21:02:27] <alex_joni> * alex_joni uses firstname@lastname.tld
[21:02:44] <SWPadnos> me@my.tld :)
[21:03:09] <cradek> maybe I should send the committer's login, instead of a guess at the author's name, to cia
[21:03:28] <cradek> yeah, I should
[21:03:40] <SWPadnos> oh, if that's possible then that gets my vote
[21:03:46] <SWPadnos> (if we're voting :)
[21:03:48] <jepler> probably -- the committer is who you need to talk to about WTF WTF WTF
[21:03:48] <SWPadnos> )
[21:04:07] <alex_joni> that's usually consistent with SF devel names too
[21:04:58] <SWPadnos> hey cool, NewEgg has a couple of those 6-core Opterons now
[21:05:08] <SWPadnos> 115W even
[21:05:53] <cradek> done
[21:06:02] <cradek> well, if I can write one line of perl correctly, that is
[21:06:32] <SWPadnos> no small task
[21:06:36] <BigJohnT> * BigJohnT wonders if it would be better to open up all the dev docs and save them just to get that out of the way so the commits don't have so much stuff in them???
[21:07:06] <SWPadnos> if you're updating to a later version of Lyx, then you absolutely should do that
[21:07:17] <jepler> BigJohnT: yes, if you can put all those useless changes in one single big commit, I'm all for it
[21:07:26] <jepler> separate from any real changes to the docs
[21:07:30] <BigJohnT> yep 1.5.3
[21:07:37] <BigJohnT> consider it done
[21:07:44] <BigJohnT> yep
[21:09:44] <jepler> bbl
[21:10:08] <CIA-3> EMC: 03micges
[21:10:08] <CIA-3> EMC: 07master * r8779be8d6923 10/src/emc/nml_intf/emc.hh: Remove unused definitions
[21:10:22] <alex_joni> cradek: you need to reconsider that line :D
[21:10:28] <cradek> dangit
[21:10:34] <cradek> it's close, you have to admit.
[21:10:35] <micges> ups :)
[21:10:42] <alex_joni> micges broke it
[21:10:52] <SWPadnos> heh
[21:11:01] <micges> heh git is easy :)
[21:11:10] <SWPadnos> ... to break
[21:11:11] <SWPadnos> :)
[21:11:21] <micges> heh
[21:11:59] <cradek> fixed (?)
[21:12:12] <cradek> getting two lines right has about 1/4 the probability
[21:16:13] <alex_joni> g'night all
[21:16:31] <BigJohnT> night Alex
[21:17:16] <micges> good night Alex
[21:20:22] <micges> cradek: any ideas for fixing las G64 bug?
[21:20:31] <micges> last*
[21:42:02] <BigJohnT> git-r-done
[21:43:02] <CIA-3> EMC: 03bigjohnt 07master * r274bc6f02b74 10/docs/src/ (72 files in 13 dirs): Convert Documents to LYX 1.5.3
[21:43:18] <BigJohnT> * BigJohnT likes git
[21:45:52] <BigJohnT> is there a "sets" HAL command?
[21:46:58] <jepler> yes
[21:48:12] <SWPadnos> yes, for signals with no writers
[21:48:54] <BigJohnT> ok, I saw it in a hm2 hal file and was wondering
[21:52:35] <jepler> hmm that's weird usage
[21:52:52] <jepler> the sets there is unneeded
[21:53:06] <micges> good night
[21:56:26] <CIA-3> EMC: 03bigjohnt 07v2_3_branch * rd237f7f6b1f7 10/docs/src/config/ini_config.lyx: Add Lathe ini settings
[21:56:26] <CIA-3> EMC: 03bigjohnt 07v2_3_branch * r3196e5b74993 10/docs/src/hal/rtcomps.lyx: Fix typo in stepgen example
[21:56:51] <BigJohnT> I was wondering why Seb did that?
[21:57:21] <jepler> I think he wants to be really sure that the enable is false before he connects it to the pwmgen enable
[21:57:27] <SWPadnos> using sets can be useful if you want to set a lot of pins to the same value
[21:57:41] <SWPadnos> but that's obviously not the reason he used it :)
[21:58:39] <BigJohnT> when you create a signal bit it defaults to false right?
[22:01:03] <SWPadnos> I think it defaults to 0
[22:01:12] <SWPadnos> which is probably false, but I don't know
[22:01:40] <BigJohnT> I just created one and it came up false in show hal config
[22:01:58] <BigJohnT> how do you sets multiple signals?
[22:02:15] <BigJohnT> it doesn't say in the man page
[22:02:32] <jepler> I'm not sure SWPadnos' idea makes much sense, but I'm thinking he means to hook them all to one signal, then sets the signal
[22:02:42] <BigJohnT> oh, ok
[22:02:49] <SWPadnos> right
[22:02:56] <SWPadnos> for manual testing, it's great
[22:03:09] <SWPadnos> hook up a bunch of enables and 'sets' them all at the same time
[22:03:20] <SWPadnos> it's like command-line pyvcp :)
[22:04:08] <BigJohnT> wow 72 files changed 31142 insertions 20040 deletions
[22:04:29] <jepler> BigJohnT is clearly the winner of the day
[22:04:54] <BigJohnT> LOL
[22:05:27] <BigJohnT> I better do my George Jetson excercise now on my button pushing finger
[22:05:52] <jepler> oops-- did I check in a broken configure script?!
[22:07:30] <CIA-3> EMC: 03jepler 07v2_3_branch * r60aac2b31110 10/src/configure: rebuild
[22:07:31] <BigJohnT> * BigJohnT wanders off to wait for the neighbor to come and as for my backhoe key :)
[22:19:17] <stustev> jepler: I hope les never sees my kinematics module - he would use artillery :)
[22:20:44] <jepler> heh
[22:51:44] <jepler> BigJohnT: the doc building process for the website completed fine after your change
[23:52:32] <cradek> hi stuart
[23:52:39] <cradek> how's it working by now?