#linuxcnc-devel | Logs for 2016-09-08

[00:07:51] <linuxcnc-build> build #4517 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/4517 blamelist: Sebastian Kuzminsky <seb@highlab.com>
[05:00:30] <emcPT> Good morning, I require help/information if possible.
[05:01:16] <emcPT> On saicanon.cc, I need to know if the code passes though something I am creating. Therefore I need something that outputs something.
[05:03:31] <emcPT> For that I tested under a function that the code should pass though, for example void FLOOD_ON() a printf("test"); a CANON_ERROR ("test"); and nothing is displayed.
[05:04:09] <emcPT> any idea?
[05:06:08] <emcPT> I suppose that when the food is connected, the code flows though the above function.
[05:06:15] <emcPT> flood not food
[06:37:40] <KGB-linuxcnc> 03John Thornton 052.7 e88c006 06linuxcnc 10docs/src/config/python-interface.txt Docs: fix incorrect return values and add constants * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=e88c006
[06:37:40] <KGB-linuxcnc> 03John Thornton 052.7 8f85aaf 06linuxcnc 10docs/src/config/python-interface.txt Docs: fix return types * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=8f85aaf
[06:41:02] <jthornton> still need to figure out the tool_table part of the python interface...
[06:49:14] <jthornton> I'm trying to work out some example code but I get tool id 1 twice when I look for it http://paste.ubuntu.com/23149822/
[06:50:29] <jthornton> http://paste.ubuntu.com/23149826/
[07:45:38] <jthornton> ah ha I figured it out :)
[08:19:21] <KGB-linuxcnc> 03John Thornton 052.7 88e22fa 06linuxcnc 10docs/src/config/python-interface.txt Docs: add info about tool_table and example code * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=88e22fa
[08:20:47] <jepler> emcPT: saicanon.cc is used to build the program "rs274", which is a standalone rs274 interpreter. It is not used in task or to perform the preview function in AXIS and other UIs
[08:23:46] <jepler> $ rs274 -g
[08:23:52] <jepler> READ => M8
[08:23:52] <jepler> 6 N..... FLOOD_ON()
[08:23:52] <jepler> 7 N..... CANON_ERROR(hi emcpt)
[08:24:48] <jepler> there are at least 3 places you have to implement a new CANON call: saicanon (for the rs274 program), emccanon.cc (for the milltask program), and emcmodule.cc (For the linuxcnc python module, used by AXIS and other UIs for preview generation)
[08:26:03] <jepler> Yay, my vacation for the Wichita event is approved.
[08:30:54] <jepler> still not sure which day I'll arrive in Wichita, I have to coordinate with cradek and seb, but maybe wed/thu, and then I'll stay through Sunday afternoon
[08:43:33] <cradek> hey mine is too! and I have a laptop that works!
[08:56:03] <emcPT> Wichita event is a linuxcnc meeting?
[08:56:13] <emcPT> if so, when?
[09:27:18] <seb_kuzminsky> emcPT: yes, the wichita event is the ~yearly linuxcnc hack fest
[09:27:45] <seb_kuzminsky> it'll be a developer-centric get-together to work on linuxcnc and discuss ideas and plans for the future of linuxcnc
[09:28:54] <cradek> I think it's important to control expectations for these things: there will not be classes or anything like classes for new users
[09:29:28] <cradek> (in the past we've not been clear about that, and it led to uncomfortable situations)
[09:30:28] <seb_kuzminsky> here's the deets: https://sourceforge.net/p/emc/mailman/message/35294740/
[09:30:32] <seb_kuzminsky> (i miss gmane)
[09:30:41] <jepler> gmane is coming back!
[09:30:41] <seb_kuzminsky> right, good point cradek
[09:30:47] <emcPT> cradek: noticed
[09:32:01] <emcPT> no crew on the captain deck
[09:32:16] <emcPT> in what city it will be?
[09:32:31] <jepler> emcPT: wichita, kansas, usa
[09:33:30] <seb_kuzminsky> emcPT: all the details on time & location are in the email i linked for you
[09:33:48] <emcPT> reading it now
[09:34:36] <emcPT> it will be 10-20 guys with pc's only or pc's and machines?
[09:35:01] <seb_kuzminsky> probably much less than 20
[09:35:02] <cradek> it will be in a machine shop
[09:35:06] <cradek> probably less than 10
[09:35:07] <seb_kuzminsky> i'd say 5-15 people
[09:35:21] <seb_kuzminsky> i plan to bring a couple of little machines i want to work on
[09:37:01] <emcPT> ok, it is clear that I cannot make the C axis on my own, but I also understand that the meeting is not focus on one's particular problem. Neverless it would be nice to know some of you.
[09:37:37] <seb_kuzminsky> you're sure welcome to join!
[09:38:43] <emcPT> humm, air ticket is about 1200€ per travel
[09:38:47] <emcPT> so 2400 total
[09:39:23] <emcPT> but next week I will be installing a machine made by us in central america!
[09:39:35] <archivist> if one cannot go, they are usually all on IRC :)
[09:39:51] <cradek> very true
[09:39:54] <emcPT> not the same thing
[09:40:24] <mozmck> also true!
[09:40:47] <seb_kuzminsky> it's good to hack next to each other, so you can ask questions and collaborate more easily
[09:41:07] <emcPT> you would be able to see my beauty and more focus could be attended to the C axis. Everyone knows that the C axis is the most important problem in the world
[09:41:17] <cradek> we've had virtual participants before, and still worked together
[09:42:17] <archivist> even webcam feeds once upon a time (I remember the cat)
[09:42:54] <emcPT> anyone traveling from chicago?
[09:43:16] <emcPT> not so hard for me to arrive to chicago
[09:43:36] <seb_kuzminsky> chicago's the wrong side for me, i'll be coming from colorado
[09:43:54] <cradek> unfortunately chicago is a day's drive in the wrong direction for me
[09:49:20] <jepler> at least two major airlines have direct flights from ORD in Chicago to ICT in Wichita.
[09:49:35] <jepler> (United and American)
[09:52:50] <emcPT> in the email says 7, 8, 9 and below from 17 to 23.
[09:52:57] <emcPT> october
[09:53:15] <cradek> the date changed a few times
[09:53:37] <emcPT> it is now fixed to 17 -> 23?
[09:54:34] <seb_kuzminsky> emcPT: yes
[10:01:41] <emcPT> For that date is more suitable (friday, saturday and sunday).
[10:01:48] <emcPT> Not joking now. If I was able to land in Wichita, would someone be available to pick me at the airport?
[10:01:54] <cradek> absolutely
[10:03:17] <emcPT> I found one travel at about 1000€, but it is 28hours from start to finish
[10:03:42] <emcPT> +4 in a bus until I get to Lisbon airport
[10:03:53] <cradek> :-/
[10:04:36] <emcPT> the meeting is during day/night were do you sleep ?
[10:04:50] <emcPT> hotel, normally, yes?
[10:05:07] <cradek> yes
[10:07:01] <emcPT> thank you for the info. I would be very interested in knowing all of you. Lets see what I can arrange/organize my life.
[10:07:54] <cradek> I hope you can make it! (but that travel sounds harrowing)
[10:13:03] <emcPT> I only was once in USA, in ElPaso
[10:13:22] <emcPT> for crossing the boarder to Mexico
[10:14:04] <emcPT> scary thing. Police here almost do not use guns
[10:22:38] <jepler> yeah it is starting to be recognized as a problem here.
[10:24:07] <archivist> I was sat in a Dennys and a fat cop sat down with his doughnut and put his chromed gun on the table
[10:24:34] <archivist> unthinkable over here
[10:28:26] <cradek> although cops shoot a lot of people, that danger is nothing compared to accidental shootings and intimate partner shootings, and both are easy to avoid (don't hang out with or date people with guns)
[10:29:46] <seb_kuzminsky> hrm
[10:30:48] * cradek waits for [citation needed]
[10:39:02] <linuxcnc-build> build #3182 of 4003.deb-lucid-i386 is complete: Failure [4failed apt-get-update shell_2] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/4003.deb-lucid-i386/builds/3182 blamelist: John Thornton <bjt128@gmail.com>
[10:42:50] ChanServ changed topic of #linuxcnc-devel to: http://linuxcnc.org | Latest releases: 2.7.7 and 2.6.12 | (this channel is logged by the zlog robot)
[10:47:51] <cradek> yay!
[10:49:15] <KGB-wlo> push to master branch: http://linuxcnc.org/
[10:49:20] -linuxcnc-github:#linuxcnc-devel- [13wlo] 15SebKuzminsky pushed 1 new commit to 06master: 02https://github.com/LinuxCNC/wlo/commit/0bb8aaa0809c7236a9b6cff7a6c48719fe8f76ff
[10:49:20] -linuxcnc-github:#linuxcnc-devel- 13wlo/06master 140bb8aaa 15Sebastian Kuzminsky: new post: linuxcnc 2.7.7
[10:50:38] <cradek> seb_kuzminsky: the "Current" link at the top didn't update
[10:50:49] <KGB-wlo> push to master branch: http://linuxcnc.org/
[10:50:52] -linuxcnc-github:#linuxcnc-devel- [13wlo] 15SebKuzminsky pushed 1 new commit to 06master: 02https://github.com/LinuxCNC/wlo/commit/e7c6cde4dda29199de6be22ae0836c0220099461
[10:50:52] -linuxcnc-github:#linuxcnc-devel- 13wlo/06master 14e7c6cde 15Sebastian Kuzminsky: bump release to version 2.7.7
[10:50:56] <cradek> ha
[10:50:58] <seb_kuzminsky> cradek: oops
[10:52:25] <seb_kuzminsky> i wonder if it's confusing that the Current Release link at the top takes you to a release announcement post instead of a download link or documentation or something
[10:52:39] <seb_kuzminsky> i know i'd never want to read the release announcement if i came looking for a release
[10:52:57] <cradek> "Download" is *right* next to it
[10:53:32] <cradek> I agree with you, but I think it's probably fine
[11:09:58] <seb_kuzminsky> it's just a papercut-class problem
[11:27:32] <jepler> I don't know what we would send them to. we can't open their GUI package manager, but that's how we believe a user gets the current version of the software..
[11:28:07] <seb_kuzminsky> yeah
[11:28:22] <seb_kuzminsky> it can't be a link to the deb because we dont know their architecture and distro
[11:28:29] <jepler> linking to debs is wrong anyway
[11:28:46] <jepler> "download a deb with your web browser and install it with sudo dpkg -i" is as antipattern as anything
[11:28:58] <seb_kuzminsky> yeah
[11:29:14] <cradek> jepler: I want to subscribe to your newsletter
[15:39:18] <andypugh> Ref a discussion last week about new forum mods. Norbert is happy to be the moderator of the Gmoccapy section, but didn’t fancy the work of being mod of the Deutsch section. (I can see his point, sometimes it gets two posts in the same day)
[15:39:49] <andypugh> No reply from TommyLight. He probably mistook me for a Nigerian heiress.
[15:48:38] <jepler> andypugh: funny, you don't look like one
[15:49:05] <JT-Shop> maybe he sounds like one...
[15:49:07] <jepler> let me see if I can figure out how to promote Norbert. That's his username on the site?
[15:49:12] <JT-Shop> when he types
[15:49:36] <JT-Shop> I can do it if you like
[15:50:17] <andypugh> Norbert is Newbynobi
[15:51:44] <jepler> OK, found
[15:52:35] <jepler> OK, newbynobi shows as moderator of Gmoccapy category in the forum
[15:55:17] <andypugh> Already, or you chaged stuff?
[16:01:31] <andypugh> I was just prompted to look at the Forum Stats by the fact that I am close to 10,000 posts. I just realised that seems to mean that I am guilty of 10% of all posts.
[16:01:50] <andypugh> And I _hate_ forums!
[16:02:17] <cradek> well on the bright side, if signal to noise ratio is >= 10% it's a great forum
[16:03:10] <cradek> andypugh: a few days ago I gave that guy an extensive answer about pluto but although he was idling, I don't think he ever saw it.
[16:04:02] <andypugh> Signal to noise is probably > 90% It’s all on topic, and as such less of a “community” than the mailing list.
[16:04:32] <andypugh> cradek: On IRC?
[16:04:36] <cradek> yeah
[16:04:51] <andypugh> I guess we could link to one of the logs?
[16:05:05] <andypugh> (in a forum post)
[16:05:22] <cradek> oh was the question originally from the forum?
[16:07:57] <andypugh> I can’t recall
[16:08:13] <cradek> it seemed like irc-only
[16:08:29] <cradek> you coming to kansas again next month?
[16:08:32] <andypugh> Yeah, it’s coming back to me
[16:08:55] <andypugh> No, i don’t currently plan to
[16:09:02] <cradek> darn
[16:09:32] <andypugh> Though I am at +30 hours flexiitme from the stint in Turkey
[16:09:55] <andypugh> Of course, It is impossible to fly to the US from here...
[16:10:06] <cradek> ?
[16:10:26] <andypugh> The FAA banned flights from Turkey to the US.
[16:10:35] <cradek> oh I figured you were back home by now
[16:10:47] <andypugh> (Actually, might have been cancelled)
[16:10:54] <andypugh> I fly back tomorrow.
[16:11:43] <andypugh> Then have 2 more 1 week trips booked
[16:11:47] <jepler> you mean direct flights, or you couldn't get an itinerary like istanbul -> frankfurt -> chicago ?
[16:12:08] <andypugh> I don;t mind now. I like it here.
[16:12:52] <andypugh> jepler: Direct flights. stupid over-reaction to a swiftlly-contained coup
[16:13:17] <jepler> weird, I think even the 7% of USAians who heard about the coup have already forgotten
[16:15:30] <kirk_wallace> Andy, would you happen to know what the pins do on this? http://www.ebay.com/itm/152210499757
[16:16:35] <andypugh> No. How many can you afford?
[16:17:27] <cradek> there might be a mode 6 pid to read sensed fuel pressure in realtime
[16:17:36] <kirk_wallace> I try not to smoke.
[16:17:47] <andypugh> It will be either 1 GND 4 +5V or the reverse. The other pins will be temperature and pressure. Maybe analogue, maybe PWM
[16:19:13] <kirk_wallace> I was hoping to find a fuel pressure sender that was digital rather than analog although PWM might be good.
[16:19:37] <cradek> heh I figured you were troubleshooting a fuel rail problem :-)
[16:19:43] <andypugh> The rail pressure sensors I am used to work at 3000bar. They are not plastic.
[16:20:54] <andypugh> Try BMW or Merc if you want the data to be CAN. Ford prefer to use cheaper interfaces.
[16:21:27] <kirk_wallace> Okay. thank you.
[16:22:42] <andypugh> Auto industry econmics is fascinating, it really is worth paying a wage for a whole year for someone to find a way to save 10 cents per car.
[16:23:23] <cradek> wow
[16:24:49] <kirk_wallace> My Explorer is running rich with cold starts, and I'd like to try a leak down test to see what's up before pull it apart and forget how to put it back together. A mechanical gauge would work but isn't as much fun.
[16:26:08] <cradek> but they're so reliable and simple
[16:26:22] <kirk_wallace> In deed.
[16:47:51] <andypugh> Bat Flattery
[16:48:37] <andypugh> kirk_wallace: Have you plugged in an OBD reader? That might tell you exactly what is wrong
[16:48:58] <kirk_wallace> I have a 1992 which has an EEC-IV ecu with OBD1 so no Mode 6. Really basic by modern standards. Maybe I could replace it with an Arduino.
[16:49:24] <cradek> ah. pre-96 is a pain.
[16:49:26] <andypugh> Yes, you probably could
[16:49:50] <cradek> but it'll have the fuel rail port to screw on a $50 gauge, right?
[16:50:07] <kirk_wallace> Yes.
[16:50:16] <andypugh> I keep seeing EEC-IV in the skip at work. They lived on as a general-purpose realtime box in all sorts of applications.
[16:50:46] <kirk_wallace> It has a Schrader valve on the rail.
[17:04:07] <cradek> kirk_wallace: heh I have the gauge in my car - how close are you to Lincoln?
[17:05:21] <kirk_wallace> About a three hour flight with one connection.
[17:06:30] <cradek> hmm, buy your own then
[17:06:34] <cradek> :-)
[17:12:24] <kirk_wallace> https://www.youtube.com/watch?v=oYnVWTuNTaI
[17:25:00] <cradek> awesome!
[17:25:46] <cradek> throttle body injection I assume? I wonder if an injector driver is hard to build
[17:28:07] <cradek> nope it says 4 injector outputs (if this is the matching doc)
[17:29:20] <cradek> that'd be quite a retrofit
[17:29:30] <cradek> you'd probably have to mount a new fuel tank with pump
[17:29:47] <kirk_wallace> This shows a throttle body: http://www.just-technology.co.uk/?page=technical_article&group=tech_article&item=jt_efi_development_145#145
[17:30:34] <cradek> that makes much more sense
[17:32:53] <kirk_wallace> There was talk about multi-port injection not working well with the mini because of the siamesed intake ports in the head.
[17:33:12] <cradek> what a fun project
[23:37:49] <KGB-linuxcnc> 03Sebastian Kuzminsky 052.6 3e9834b 06linuxcnc 10(8 files) add a test of early exit from cutter comp * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=3e9834b
[23:37:50] <KGB-linuxcnc> 052.6-issue-160 bbf1259 06linuxcnc 04. branch deleted * 14http://git.linuxcnc.org/?p=linuxcnc.git;a=commitdiff;h=bbf1259