#emc-devel | Logs for 2010-09-08

[00:52:45] <KimK> This was odd, I put some stuff in the tool table, pressed check entries, it checked good. http://imagebin.org/113158 I pressed write, and got an error: http://pastebin.ca/1935413 Are comments not supposed to start with a number? Or must it be a floating point number? Or is that a "copy from comments to radius" hidden feature? Comments welcome, I'll be back later.
[01:15:07] <cradek> sure doesn't sound like a feature to me
[01:15:21] <cradek> I did not understand what you said about IJQ vs QIJ
[01:17:23] <cradek> I can't immediately reproduce any problem with the comment becoming the diameter - can you give me a step-by-step?
[02:33:28] <jepler> in master, run tooledit configs/sim/sim.tbl and in the 'comment' field enter '0/0'
[02:33:35] <jepler> click "write tool table"
[02:33:52] <jepler> doesn't appear to be in v2.4_branch
[02:34:09] <jepler> or else my v2.4_branch is not up to date
[02:34:14] <jepler> * jepler checks
[02:35:58] <jepler> my 2.4 was stale, and the problem is in 2.4
[02:36:04] <cradek> yuck
[02:38:02] <cradek> something interprets it as a number, because if I make a comment 0.01234000 it changes to 0.01234
[02:38:22] <cradek> but a comment of asdf is fine, and doesn't error
[02:38:58] <cradek> as is 1;23kl;jklvcx;43
[02:39:03] <cradek> what's special about 0/0?
[02:39:08] <jepler> not sure yet
[02:39:20] <jepler> it may be related to the fix for leading zeros in other columns
[02:39:29] <jepler> e.g., treat 099 as 99, not an error
[02:39:39] <jepler> and 077 as 77, not 63
[02:39:47] <cradek> oh right, I remember
[02:40:21] <jepler> + # correct entries with leading zeros
[02:40:23] <jepler> + if {[string first 0 [string trim $::te($type,$i,$h)]] == 0} {
[02:40:23] <jepler> + set ::te($type,$i,$h) [format %g $::te($type,$i,$h)]
[02:40:23] <jepler> + }
[02:40:38] <jepler> yeah, I think this needs to not be done for the comment column
[02:46:28] <jepler> http://emergent.unpy.net/files/sandbox/0001-tooledit-fix-bug-with-0-at-start-of-comment.patch
[02:47:05] <cradek> that looks promising, but I'm pretty useless at reviewing tcl
[02:47:12] <cradek> $h is always in english?
[02:48:34] <jepler> I don't think tooledit is translatable at the moment
[02:49:10] <jepler> h is each thing from the array $::te(unified,header) in turn -- basically, each column
[02:53:08] <jepler> hah http://en.wikipedia.org/wiki/Honeywell_316#Kitchen_Computer
[02:53:45] <cradek> aren't they cool?
[02:54:11] <jepler> I suppose in those days $10,600 was real money
[02:54:37] <jepler> KimK: when you come back, please tell me if the patch fixes the problem you ran into. http://emergent.unpy.net/files/sandbox/0001-tooledit-fix-bug-with-0-at-start-of-comment.patch
[02:54:45] <jepler> if so I'll put it in v2.4_branch tomorrow
[02:54:46] <jepler> 'night all
[02:55:00] <cradek> thanks for fixing it
[07:16:18] <CIA-5> EMC: 03micges 07master * r8725a7353ad2 10/src/emc/rs274ngc/ (interp_convert.cc rs274ngc_return.hh): interp: Remove redundant checks in convert_probe()
[08:05:21] <CIA-5> EMC: 03cmorley 07v2.4_branch * r36906267dfbf 10/src/emc/usr_intf/pncconf/pncconf.py: fix typo in firmware name SVST2_4_7I47
[14:19:08] <Al_Smt> wondering if these could & should be added to tkemc 1st http://pastebin.ca/1935265 2nd http://pastebin.ca/1935731
[15:21:45] <KimK> jepler, cradek: Yes, jepler's patch got rid of the error (at least when re-using my same comments from the "imagebin", I didn't try to fool it a second time with more imaginative comments, lol.) Although it did seem to change all the other cells with "0.0" and "0." to simply "0", but I will understand if I have to live with that. Thanks very much for the fast fix!
[15:23:05] <KimK> Now this error was discovered on master, and your patch was only applied to master, but I'm sure it will be fine on 2.4 as well.
[15:25:00] <KimK> s/applied to master/applied locally to my master/
[15:27:01] <Jymmm> KimK: your wife?
[15:29:27] <KimK> BTW, that was the first time I applied an outside patch like that, it took me a couple of tries to make it work. Neat, I'm "ahead of master" for a few minutes, that's pretty cool. Now, when jepler applies the patch to git, what will happen when I pull next time? What's the usual way to cancel a local test patch in favor of the main git, a hard reset?
[15:29:37] <KimK> Jymmm: LOL
[15:30:09] <Jymmm> ;)
[15:31:52] <KimK> Sorry to be such a noob, thanks for your help.
[15:39:04] <KimK> * KimK gives jepler and cradek plaques engraved with "It's impossible to make anything fool-proof because fools are so ingenious".
[15:41:37] <KimK> Jymmm: How's the dust collector project coming along? You were working on something about lid clamps the last time you talked about it.
[15:42:05] <KimK> Or the last time I heard, anyway.
[15:43:49] <Jymmm> KimK: Still working on it, changes designs as it goes. I *think* I have the basic design down, still need to work on the interior (for structural strength) as I'm doing some oddball mounting of the entire enclosure to accommodate size restrictions.
[15:45:57] <Jymmm> KimK: I WAS going to do this (note the blue tabs), but going to make the bottom flush with the interior of the steel legs but drop it down two inches and bolt on some aluminum angle to the side of the enclosure to rest of the legs http://i51.tinypic.com/281g6eh.jpg
[15:47:13] <Jymmm> KimK: I just hope that by side mounting, that I'm not stressing the joints of the bottom of the enclosure too much. I 'd really hate if the bottom fell out - literally.
[15:48:59] <KimK> You could always notch a few angle irons as "floor joists" and lay them across the gap? Maybe fasten them with countersink head screws?
[15:50:02] <Jymmm> KimK: The floor joist idea won't work, as I need to drop the enclosure BELOW the top of the legs by 2"
[15:50:27] <KimK> Ah, OK. Good idea while it lasted, lol.
[15:50:32] <Jymmm> I'm working with a tight height limitations.
[15:50:46] <Jymmm> brb coffee
[15:50:51] <KimK> OK
[16:19:44] <jepler> CIA-5: you there?
[16:19:51] <CIA-5> EMC: 03jepler 07v2.4_branch * rfcadb506d15a 10/src/emc/usr_intf/tooledit.tcl: tooledit: fix bug with 0 at start of comment
[16:24:51] <CIA-5> EMC: 03jepler 07master * r534c92709b1d 10/src/emc/usr_intf/pncconf/ (pncconf.glade pncconf.py): add option to set joystick jog speed with buttons
[16:24:51] <CIA-5> EMC: 03jepler 07master * r28dbe2b931fb 10/src/emc/usr_intf/pncconf/ (pncconf.glade pncconf.py): Add usb joystick jog option
[16:26:57] <CIA-5> EMC: 03jepler 07master * ra79585fc5fb3 10/src/emc/usr_intf/pncconf/ (pncconf.glade pncconf.py): add ability to search for pncconf made device rules
[16:27:02] <CIA-5> EMC: 03jepler 07master * r5180334d3f7b 10/src/emc/usr_intf/pncconf/ (pncconf-help/help-extcontrols.txt pncconf.glade pncconf.py): Merge branch 'joystick' into v2.4_branch
[16:27:11] <CIA-5> EMC: 03jepler 07master * r7d8ecad97737 10/src/emc/usr_intf/pncconf/pncconf-help/help-extcontrols.txt: Update help page for USB joystick jogging
[16:28:16] <CIA-5> EMC: 03jepler 07master * r19b60670ee48 10/scripts/runtests: testsuite: fix checking exit values
[16:28:17] <CIA-5> EMC: 03jepler 07master * rd7dc1300b942 10/src/emc/usr_intf/pncconf/pncconf.py: fix illegal filename characters in USB device rule name
[16:28:17] <CIA-5> EMC: 03jepler 07master * r5aa42c9a61e8 10/tests/overrun/test.sh: testsuite: fix overrun test
[16:28:18] <CIA-5> EMC: 03jepler 07master * rdeb305e11e15 10/ (scripts/runtests tests/README): testsuite: allow tests written in scripting langs
[16:28:19] <CIA-5> EMC: 03jepler 07master * rd7ee56fb9e32 10/tests/interp/bad/ (9 files): testsuite: check that invalid gcodes are rejected
[16:28:20] <CIA-5> EMC: 03jepler 07master * r36906267dfbf 10/src/emc/usr_intf/pncconf/pncconf.py: fix typo in firmware name SVST2_4_7I47
[16:28:21] <CIA-5> EMC: 03jepler 07master * recbe2a8ad018 10/tcl/tkemc.tcl: tkemc: get rid of unneeded 'upvar's
[16:28:23] <CIA-5> EMC: 03jepler 07master * rfcadb506d15a 10/src/emc/usr_intf/tooledit.tcl: tooledit: fix bug with 0 at start of comment
[16:28:24] <CIA-5> EMC: 03jepler 07master * r59da96161957 10/tcl/tkemc.tcl: tkemc: add some padding
[16:28:26] <CIA-5> EMC: 03jepler 07master * r3d481344320a 10/ (17 files in 8 dirs): Merge remote branch 'origin/v2.4_branch'
[16:28:27] <CIA-5> EMC: 03jepler 07master * r3f116eab7600 10/src/emc/motion/control.c: motion: improve G38.4 error message
[16:38:57] <jepler> hmmmm -- g5.1 is not modal?
[16:43:50] <CIA-5> EMC: 03jepler 07master * re3ac9c55085a 10/tcl/tkemc.tcl: tkemc: show coordinate system in offset widget
[16:53:42] <jepler> READ => g2 j1
[16:53:42] <jepler> 7 N..... ARC_FEED(0.0000, 0.0000, 0.0000, 1.0000, -1, 0.0000, 0.0000, 0.0000, 0.0000)
[16:53:45] <jepler> READ => g5.1 x0y0i1j1
[16:53:48] <jepler> 8 N..... NURBS_FEED(3, ...)
[16:53:50] <jepler> READ => j1
[16:53:53] <jepler> 9 N..... ARC_FEED(0.0000, 0.0000, 0.0000, 1.0000, -1, 0.0000, 0.0000, 0.0000, 0.0000)
[16:54:07] <cradek> ick
[16:55:46] <micges> ouch
[16:56:02] <micges> jepler: patch for 2.4: http://codepad.org/S4lnhhNG
[16:58:58] <KimK> It appears I may have stumbled across another issue. In EMC2 master I'm running lathe-sim. I set up the tool table as shown, and was walking through the tools from T1 to T9. When I asked for T9, it error boxed, saying "Requested tool 9 not found in tool table". Comments and suggestions welcome. http://imagebin.org/113248
[16:59:57] <cradek> did you write the tool table and then reload it?
[17:02:14] <KimK> Ha! Well, don't I feel stupid, lol. That was it. I forgot to reload. As a friend of mine says, "Sorry I blew up", lol.
[17:02:26] <cradek> ha
[17:03:13] <KimK> Love those easy fixes, lol. What is it, PEBKAC?
[17:04:13] <jepler> micges: presuambly you'd actually provide a translation of the message in one language too
[17:04:55] <micges> update translation before 2.4.4 was next step
[17:07:27] <jepler> you considered whether the strings like 'error', 'system command', 'probing' should be translated, and decided not to?
[17:07:59] <micges> I don't understand
[17:08:48] <jepler> when exec_state is emc.EXEC_ERROR the string displayed will be 'error'
[17:09:09] <jepler> I'm asking if you considered whether the string 'error' should be translated
[17:09:21] <micges> ah ok
[17:10:42] <micges> (small typo: http://codepad.org/XLPNOARN)
[17:10:48] <micges> (2.4)
[17:11:12] <jepler> no, it's right for the wm iconname to be translated
[17:11:46] <jepler> window managers have the option of showing the iconname when there is less space to show a window's name than in the title bar (for instance when it's an icon or on the window list at the bottom of the screen)
[17:12:21] <micges> I see
[17:12:57] <Al_Smt> jepler, thanks
[17:13:00] <jepler> bbl, it's lunchtime here
[17:13:05] <jepler> Al_Smt: thank you
[17:13:54] <ries_> ries_ is now known as ries
[17:18:36] <micges> jepler: I think translation of content of emctop items would just mess it up, our users just run it, copying it to file and send it to us
[17:43:17] <jepler> micges: OK, go ahead and push that change to v2.4_branch
[18:23:41] <CIA-5> EMC: 03micges 07v2.4_branch * r4e73474ae112 10/src/emc/usr_intf/axis/scripts/emctop.py: emctop: make it translatable
[20:51:54] <moopy> anyone know how kinematics hooks joints into emc??
[21:03:34] <alex_joni> wonder if I could write a CD directly from http
[21:04:45] <micges> kinematics maps joints positions(1..n) to axis positions (xyzabcuvw)
[21:06:47] <cradek> and the other way
[21:19:24] <moopy> axis gui does not seem to be showing more than 6 joints
[21:20:11] <moopy> the max5kins demo shows 9
[21:22:53] <jepler> AXIS decides which axes to display by using the 'axis_mask' attribute of the stat object (s.axis_mask). It display joints from 0..num_joints-1, where num_joints comes from the inifile [TRAJ]JOINTS.
[21:24:03] <jepler> if [TRAJ]JOINTS is not specified then it defaults to live_axis_count, which is the number of axes seen in live_axis_count
[21:26:56] <jepler> er, "... which is the number of axes seen in axis_mask"
[21:27:37] <jepler> so if you are specifying 6 axes (e.g., XYZABC) but want a different number of joints to appear, then specify [TRAJ]JOINTS=9 (or whatever)
[21:30:00] <jepler> for instance it might be more sensible to specify 'COORDINATES = X Y B C C U V W' in max5kins, since there's no A rotary IIRC
[21:30:28] <jepler> but if you did that without specifying[TRAJ]JOINTS=9 then joint 8 (remember, joints are numbered from 0, not from 1) would disappear
[21:30:39] <jepler> bbl
[21:32:17] <moopy> i have: AXES = 7
[21:32:17] <moopy> COORDINATES = X Y Z A B C D
[21:32:32] <alex_joni> D is not a coordinate
[21:32:49] <moopy> so it should be U V W
[21:32:50] <alex_joni> you get to pick from: XYZABCUVW
[21:33:24] <moopy> but if i define JOINTS=10 it will show all 10 in axis?
[21:33:52] <alex_joni> nope, it's max 9 afaik
[21:34:06] <alex_joni> and having joints!=axes in 2.4.x won't really work
[21:34:26] <moopy> so XYZABCUVW is all thats possible
[21:35:25] <moopy> i have no idea if that hack i did to genserkins will really work, did you look at it alex?
[21:36:39] <moopy> i think maybe some of the changes could be wrong like : go_matrix_init(Jfwd, Jfwd_stg, 6, genser->link_num);
[21:37:09] <alex_joni> moopy: briefly
[21:37:13] <alex_joni> and no, that looks ok
[21:37:30] <moopy> should all the references to '6' be changeds to 'genser->link_num'??
[21:37:35] <alex_joni> not all
[21:38:23] <moopy> basically all i did was do a replace of the 'num_joints /*GENSER_MAX_JOINTS*/'
[21:38:47] <moopy> and i have made a couple of changes sice the version i pastebin yesterday
[21:39:41] <moopy> i hope i am not just giving you shit to trawl through, as i think maybe i really dont properly understand the code
[21:40:06] <alex_joni> that's quite likely :)
[21:40:27] <moopy> lol
[21:41:52] <moopy> all the pyvcp dials seem to be working for 7 joints when i run the demo now
[21:42:18] <moopy> but i dont thing the gcode is plotting correctly
[21:43:20] <moopy> i notice the genser->num_links was being set to 15 due to an error in the code i added