Back
[07:36:07] <jepler> seb_kuzminsky: thanks for more releases
[07:46:22] <KGB-linuxcnc> 03John Thornton 052.7 5524071 06linuxcnc 10docs/src/user/user-concepts.txt Docs: remove apostrophe from section title. * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=5524071
[10:06:22] <seb_kuzminsky> the 2.7.8 release debs are on their way to wlo
[10:15:46] <cradek> thank you seb!
[10:41:57] ChanServ changed topic of
#linuxcnc-devel to:
http://linuxcnc.org | Latest releases: 2.7.8 and 2.6.13 | (this channel is logged by the zlog robot)
[10:57:59] <seb_kuzminsky> hmm
[10:58:22] <KGB-wlo> push to master branch:
http://linuxcnc.org/
[10:58:23] <seb_kuzminsky> ah
[10:58:24] -linuxcnc-github:#linuxcnc-devel- [13wlo] 15SebKuzminsky pushed 1 new commit to 06master: 02
https://github.com/LinuxCNC/wlo/commit/5f56bac97048f3708ebacd245e8bf5a790a12482
[10:58:25] -linuxcnc-github:#linuxcnc-devel- 13wlo/06master 145f56bac 15Sebastian Kuzminsky: new post: 2.7.8 release announcement
[10:59:47] <KGB-wlo> push to master branch:
http://linuxcnc.org/
[10:59:58] -linuxcnc-github:#linuxcnc-devel- [13wlo] 15SebKuzminsky pushed 1 new commit to 06master: 02
https://github.com/LinuxCNC/wlo/commit/8a016b264bbe5bc78b0049c619648d0be11bc90d
[10:59:58] -linuxcnc-github:#linuxcnc-devel- 13wlo/06master 148a016b2 15Sebastian Kuzminsky: new 2.7.8 release for the top bar
[11:03:51] <terkaa> Hi everyone
[11:04:13] <seb_kuzminsky> hi terkaa
[11:04:33] <seb_kuzminsky> are you Tero Kaarlela from the mailing list?
[11:04:41] <terkaa> Yes
[11:04:46] <seb_kuzminsky> hi! :-)
[11:05:26] <terkaa> Also from Forum :)
[11:06:14] <seb_kuzminsky> i haven't looked at your g84 patch since that first feedback i sent you, but thanks for adding documentation
[11:06:37] <seb_kuzminsky> i'll try to read the code and review it in more detail when i can
[11:06:54] <seb_kuzminsky> i forget, did you also add a test?
[11:07:23] <seb_kuzminsky> ah, i see now that you asked for my help to add a test and i never replied to you, sorry
[11:07:47] <seb_kuzminsky> my advise is, take a look at tests/motion-logger/basic
[11:08:09] <seb_kuzminsky> there are several little sub-tests there for testing what motions different G-codes ask for
[11:08:33] <seb_kuzminsky> for example, g1.ngc should results in the motions described by expected.g1
[11:08:36] <terkaa> Ok. I Will check that
[11:08:49] <seb_kuzminsky> great
[11:10:44] <terkaa> If I can make that I will also update diff file
[11:11:45] <seb_kuzminsky> have you read the Contributing to LinuxCNC document? It requests contributions in the form of git branches rather than diff files
[11:11:57] <seb_kuzminsky> if that's an obstacle for you, diff files will be ok of course
[11:12:21] <seb_kuzminsky> http://linuxcnc.org/docs/devel/html/code/contributing-to-linuxcnc.html
[11:14:17] <terkaa> I will take a look at that too. maybe I can manage that also
[11:14:48] <seb_kuzminsky> it's a bit of a learning curve for sure, but i find it's helpful for me to keep the stuff i'm working on organized
[11:14:55] <jepler> files produced by "git commit" followed by "git format-patch" are better for our needs than ones produced by "diff" or "git diff" too, even if you prefer not to sign up for github to publish your branches there.
[11:16:34] <terkaa> I did sign up few days ago :)
[11:19:51] <terkaa> Ill try to study that
[11:38:06] <terkaa> Would it be a good idea To mimic tests in interp g81 folder?
[11:51:12] <seb_kuzminsky> terkaa__: yeah that would work too
[11:51:47] <terkaa__> Ok, thanks for pointing me to right direction
[11:51:48] <seb_kuzminsky> the tests/interp tests use 'rs274', the stand-alone-interpreter, to see what canon calls result from snippets of gcode
[11:52:17] <seb_kuzminsky> the motion-logger tests use a mock of the motion controller to verify the motion commands that result from the canon calls that result from snippets of gcode
[11:52:49] <seb_kuzminsky> slightly more coverage, but if you're just changing interp, not canon, then rs274 is probably fine
[11:53:13] <seb_kuzminsky> brb
[11:53:47] <terkaa__> Ah, ok
[12:02:37] <seb_kuzminsky> so really, either one would work for what you're doing, feel free to use whichever you prefer
[12:03:18] <jepler> tests using 'rs274' are faster to execute
[12:05:00] <terkaa__> Ok, tests for G84 are almost same. I will modify those to suit G84/G74 also
[12:05:42] <terkaa__> Tests for G81
[14:49:09] <skunkworks_> either 4.8.6-rt isn't stable or 4.8.6 + virtualbox isn't stable..
[15:17:26] <jepler> skunkworks_: sorry to hear it
[15:17:35] <jepler> I don't think I've done 4.8 yet, I've list track though.
[15:17:48] <jepler> 4.1 4.4 4.6 maybe?
[15:18:07] <jepler> and of course no virtualbox
[15:18:20] <jepler> I've been using kvm + libvirt recently and that's actually pretty nice for linux guests
[15:19:19] <skunkworks_> I was goofing around with libvirt. it works too - I was really looking for better graphic performace to run fusion 360 but so far no go.
[15:19:27] <skunkworks_> *kvm
[15:19:32] <jepler> yeah
[15:19:59] <jepler> I learned to my chagrin for $DAY_JOB that 3D support in virtualization is pretty sad, particularly if you want to pay $0 for virtualization SW and a $0 surcharge on your graphics
[15:20:03] <jepler> cards
[15:20:19] <jepler> by the time you go up to an extreme tier of vmware + extra expensive virtualization-oriented nvidia cards it's apparently nice
[15:21:01] <jepler> so I got to make a build of mesa opengl in llvmpipe mode to bundle with the software as an alternative. ugh.
[15:23:27] <skunkworks_> If I want to really play with fusion - I reboot to windows 10 on the laptop.
[15:29:33] <KGB-linuxcnc> 03Dewey Garrett 052.7 3c9b549 06linuxcnc 10docs/man/man1/halcmd.1 10src/hal/utils/halcmd_commands.c halcmd manpage & help: update for 'save' command * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=3c9b549
[15:34:12] -linuxcnc-github:#linuxcnc-devel- [13linuxcnc] 15fixerslo commented on issue #203: The same issue is also with active G-Codes, it shows read-ahead values instead of current values and current program line doesn't return any useful value when there is non-linear program flow, for instance when using subroutines and loops. 02
https://github.com/LinuxCNC/linuxcnc/issues/203#issuecomment-259526468
[15:36:20] -linuxcnc-github:#linuxcnc-devel- [13linuxcnc] 15cradek commented on issue #203: Please help us test the statetags branch, which is mean to fix these problems.... 02
https://github.com/LinuxCNC/linuxcnc/issues/203#issuecomment-259526902
[15:36:35] <cradek> t
[15:36:35] -linuxcnc-github:#linuxcnc-devel- [13linuxcnc] 15cradek commented on issue #203: Please help us test the statetags branch, which is meant to fix these problems.... 02
https://github.com/LinuxCNC/linuxcnc/issues/203#issuecomment-259526902
[15:36:40] <cradek> heh
[15:44:42] <jepler> rebooting forum, new kernel..
[15:45:51] <jepler> cradek: thanks for taking care of the backups of that thing
[15:47:02] <cradek> welcome!
[16:45:30] <andypugh> cradek: Have you seen that modification to Touchy that I want to make?
[16:45:41] <andypugh> Any opinion on the matter?
[16:47:07] <cradek> I saw a small patch on pastebin? if it's useful and you think it's likely to not break things you should make it
[16:47:57] <andypugh> Well, I don’t _expect_ it to break anything
[16:48:46] <andypugh> http://pastebin.com/FYXGRx16
[16:49:27] <andypugh> The code is actually courtesy of Chris Morley
[16:50:08] <andypugh> Actually, that paste is only part of it
[16:50:23] <andypugh> http://pastebin.com/ibcRZkth
[16:50:31] <andypugh> Has the actual stuff that does something
[16:50:57] <andypugh> I will lose the “print” and document it.
[16:51:18] <andypugh> Ideally Gscreen, Gmoccapy and Touchy would send the same message
[16:51:30] <cradek> totally
[16:51:43] <andypugh> It basically provides a messgae to the embedded tab that it has just come to the front.
[16:51:58] <cradek> and you're right that documentation that says how to use it is important
[17:03:11] <JT-Shop> that code is so the GUI can send a message to?
[17:05:10] <JT-Shop> pink is on reprap asking about LinuxCNC...
[17:10:33] <andypugh> JT-Shop: Better there than here asking about reprap
[17:10:51] <JT-Shop> yep
[17:10:59] <JT-Shop> what does the send message do?
[17:11:01] <andypugh> And, the code is so the host GUI can tell an embedded tab that it is active/visible
[17:11:25] <andypugh> Otherwise the embedded tab embedded code can’t tell
[17:11:59] <JT-Shop> ah ok that makes sense now
[17:12:22] <JT-Shop> pretty slick
[17:23:11] <KGB-linuxcnc> 03Dewey Garrett 05master 8163b09 06linuxcnc 10debian/changelog 10docs/src/Master_Documentation.txt 10docs/src/common/emc-history.txt 10src/hal/utils/halcmd_commands.c Merge remote-tracking branch 'origin/2.7' * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=8163b09
[17:23:11] <KGB-linuxcnc> 03Dewey Garrett 05master fe1595f 06linuxcnc 10docs/man/man1/halcmd.1 10src/hal/utils/halcmd_commands.c halcmd: new save types "unconnectedinpins, "allu" * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=fe1595f
[17:23:11] <KGB-linuxcnc> 03Dewey Garrett 05master 1beca56 06linuxcnc 10lib/hallib/sim_lib.tcl basic_sim.tcl: save all state for equiv hal cmds * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=1beca56
[17:37:10] <andypugh> What’s the incantation to generate the html docs?
[17:37:45] <andypugh> I have edited the asciidoc, but want to see what it looks like formatted
[17:39:22] <JT-Shop> ./autogen.sh
[17:39:22] <JT-Shop> ./configure --enable-build-documentation > /home/john/Desktop/configure.txt 2>&1
[17:39:22] <JT-Shop> make clean > /home/john/Desktop/make_clean.txt 2>&1
[17:39:22] <JT-Shop> make > /home/john/Desktop/make.txt 2>&1
[17:39:26] <JT-Shop> that's what I do
[17:39:46] <andypugh> Yes, thanks, just found it in ./configure -help
[17:39:51] <JT-Shop> I send the output to a file so I can look at it if need be
[17:40:34] <andypugh> I seem to lack a dependency
[17:40:39] <andypugh> I wonde rhow many?
[17:42:12] <JT-Shop> http://paste.ubuntu.com/23453115/
[17:42:22] <JT-Shop> that's what I do when I build from fresh
[17:43:13] <seb_kuzminsky> andypugh: not sure if this helps:
http://linuxcnc.org/docs/devel/html/code/building-linuxcnc.html#_building_for_run_in_place
[17:44:11] <seb_kuzminsky> also, it's not mentioned explicitly in that document but you can say "make ../docs/html/gcode/g-code.html" to build a particular page, or "make docs" to build all the docs
[17:45:06] <JT-Shop> make docs is a new one for me, does it do all the link checking?
[17:48:37] <jepler> if not it's an oversight
[17:51:05] <seb_kuzminsky> yes it does
[17:51:20] <JT-Shop> cool I use that and save a bunch of time
[17:51:36] <seb_kuzminsky> 'make docs' is what the docs buildslave uses, so it's pretty well tested
[17:55:19] <linuxcnc-build_> build #2769 of 1400.rip-wheezy-i386 is complete: Failure [4failed compile runtests] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/1400.rip-wheezy-i386/builds/2769 blamelist: Sebastian Kuzminsky <seb@highlab.com>, Dewey Garrett <dgarrett@panix.com>, John Thornton <bjt128@gmail.com>
[18:11:02] <seb_kuzminsky> hmm, i wonder why all the buildslaves are failing to check the links on the docs
[18:11:14] <seb_kuzminsky> they say they can't find checklinks, but it's clearly installed...
[18:11:31] <seb_kuzminsky> when i run it by hand it finds it
[18:29:07] <andypugh> So, I made a new commit, then Dewey made some, and so I couldn’t push without a pull, but now if I push there is a merge as part of the push. Can I just reset —hard and then cherry-pick my commit to the tip of master, or will the reset delete the commit?
[18:32:18] <cradek> you can git pull --rebase and it'll move yours on top of his
[18:32:38] <cradek> er, not if you've already created the merge
[18:33:16] <andypugh> Well, git pull did the merge without asking
[18:33:53] <cradek> yeah, it does that unless you use --ff-only
[18:34:04] <cradek> the merge is fine, don't sweat it
[18:34:15] <KGB-linuxcnc> 03Sebastian Kuzminsky 052.7-checklink 03b6e80 06linuxcnc 10docs/src/checkref get some debug output from the buildbot * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=03b6e80
[18:34:26] <cradek> you could change it around if you care, but whatevs
[18:35:07] <KGB-linuxcnc> 03andypugh 05master 1bcd51e 06linuxcnc 10docs/src/gui/gladevcp.txt 10src/emc/usr_intf/touchy/touchy.py GladeVCP: Send a message from Touchy to an embedded GladeVCP tab to tell * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=1bcd51e
[18:35:56] <andypugh> reset —hard desn’t stop you cherrypicking the commit (it wasn’t much of a gamble, I had the changed files still open in a text editor_
[18:52:04] <pcw_mesa> Can the numpy fix be added to Master?
[18:56:34] <cradek> is it simple?
[19:00:47] <linuxcnc-build_> build #4625 of 0000.checkin is complete: Failure [4failed] Build details are at
http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/4625 blamelist: Sebastian Kuzminsky <seb@highlab.com>, Dewey Garrett <dgarrett@panix.com>, John Thornton <bjt128@gmail.com>
[19:18:40] <pcw_mesa> its in 2.7.8
[19:24:30] <cradek> there are no changes in 2.7.8 that are not also in master
[19:25:37] <pcw_mesa> hmm... Let me try again
[19:33:43] <pcw_mesa> OK so neither work for some reason (no module named numpy.core)
[19:53:58] <seb_kuzminsky> rut roh
[20:33:32] <skunkworks> zlog
[20:40:24] <seb_kuzminsky> pcw_mesa: do you have python-numpy installed?
[20:43:37] <seb_kuzminsky> it's a runtime dependency of the linuxcnc deb, but src/configure doesn't check for it
[20:43:45] <pcw_home> I thought I did
[20:43:46] <seb_kuzminsky> so if you're running-in-place, it won't warn you
[20:43:46] <pcw_home> I'll check tommorow
[20:44:04] <seb_kuzminsky> ok
[20:44:10] <pcw_home> Yeah I'm RIP
[20:57:20] <seb_kuzminsky> i bet that's it