#hazzy Logs

Sep 07 2020

#hazzy Calendar

04:37 AM lmclaren[m] is now known as Mufdvr0978[m]
04:50 AM JT-Cave: morning
10:03 AM roguish[m]: good morning..... due for another sweltering day.....
10:06 AM JT-Shop: actually a nice day here
10:07 AM JT-Shop: wow is this keyboard dirty...
10:29 AM * roguish[m] posted a file: probe a profile-inch.ngc (1KiB) < https://matrix.org/_matrix/media/r0/download/matrix.org/pwqOHkyRORfKKilXRmFHTPTO/probe a profile-inch.ngc >
10:29 AM roguish[m]: JT-Shop: check it out for probing...
10:37 AM JT-Shop: roguish[m], cool
10:44 AM roguish[m]: JT-Shop: and https://www.youtube.com/watch?v=JO8lLaV6pbw
10:53 AM JT-Shop: it locks up linuxcnc when I load it...
10:54 AM JT-Shop: damn windblows line endings lol
10:55 AM JT-Shop: hazzy-m, can we catch windblows line endings when loading a file?
10:55 AM roguish[m]: oop. sorry. that might have been me....
10:55 AM JT-Shop: File Save As Linux
10:56 AM JT-Shop: wow one of the pullets is in the nest box
10:57 AM hazzy-m: JT-Shop: yes, a neat solution might be to add a filter program for .ngc files, to strip the winblows endings in the filter
10:57 AM hazzy-m: that would be seamless
10:57 AM JT-Shop: cool
11:05 AM JT-Shop: well carp it locks up mill touch v6 no matter what I try and do
11:06 AM JT-Shop: but it runs ok in axis sim...
11:08 AM JT-Shop: carp the owner is nobody lol
11:09 AM hazzy-m: JT-Shop: that's strange ...
11:09 AM JT-Shop: I think it happens when I copy a file from here to the BP
11:10 AM hazzy-m: can you share the file?
11:10 AM hazzy-m: ll
11:10 AM JT-Shop: yea
11:10 AM JT-Shop: dpaste or post the file to my website?
11:12 AM hazzy-m: will dpaste preserve the line endings?
11:13 AM JT-Shop: not sure
11:16 AM JT-Shop: https://gnipsel.com/files/emc/probe.ngc
11:18 AM JT-Shop: just changed owner and group for all the files to me
11:26 AM JT-Shop: it's not the owner or the group it still locks up mtv6
11:26 AM JT-Shop: opens ok in axis...
11:28 AM JT-Shop: looks like a comment was borked
12:31 PM JT-Shop: is this a deal breaker for qtpyvcp (AXIS,stop)
12:35 PM JT-Shop: roguish[m], did Andy write that probe routine?
12:38 PM JT-Shop: ah (AXIS,stop) stops drawing the preview
12:44 PM roguish[m]: JT-Shop: I think so, but not absolutely sure. see the youtube video link I posted. it's andy using it.
12:45 PM Not-8b99: [02qtpyvcp] 07KurtJacobson pushed 031 commit to 03master [+0/-0/±3] 13https://git.io/JUZS5
12:45 PM Not-8b99: [02qtpyvcp] 07KurtJacobson 03e9b87dd - ENH: add support for enum type settings in menubar
12:46 PM Not-8b99: [02qtpyvcp] 07KurtJacobson pushed 031 commit to 03gh-pages [+523/-0/±0] 13https://git.io/JUZSA
12:46 PM Not-8b99: [02qtpyvcp] 07traviscibot 039189601 - Deploy kcjengr/qtpyvcp to github.com/kcjengr/qtpyvcp.git:gh-pages
12:48 PM hazzy-m: JT-Shop: I don't think (AXIS, stop) should be an issue
12:48 PM hazzy-m: it's just a comment that gets treated specialy by axis
12:50 PM JT[m]: yea, I had forgot what that did
01:03 PM JT[m]: hmm probe.ngc locks up my mtv6 sim as well
01:09 PM JT-Cave: hazzy-m, I think it's trying to draw an endless plot... can we turn off plotting?
01:10 PM JT-Cave: I can hear my hard drive thrashing after loading the file...
02:31 PM roguish[m]: hazzy: i've installed Adargel's hole stuff and now get an error just starting the vcp editor. 'No module named float_line_edit'
02:31 PM roguish[m]: waz up?
02:35 PM roguish[m]: adn same error when I touch the 'HoleCircleWidget' in the Widget Box in the designer.
02:38 PM JT-Shop: https://dpaste.com/DL3WRGHWJ this file gives me an error no percent or program end
02:50 PM AaronDargelGitte: roguish: did you install the latest code? the conversational controls were removed from the qtpyvcp and added here: https://github.com/kcjengr/qtpyvcp.conversational-gcode.
02:50 PM AaronDargelGitte: This works for me: pip install git+https://github.com/kcjengr/qtpyvcp.conversational-gcode.git
02:52 PM roguish[m]: Aaron Dargel (Gitter): thanks for popping in. i'll do that exactly right now..
02:58 PM roguish[m]: Aaron Dargel (Gitter): naw. same error. where does that install put the probe_basic gui .ui ??? or what ever I should be editting in the gui editor?
03:03 PM AaronDargelGitte: roguish: it doesn't install probe_basic, just the widgets. you can try to open the hole_circle.ui file from the project in the designer. there is no probe_basic release yet with these controls in it.
03:09 PM roguish[m]: ah, ok. well that ui is way down in the .local, and i still get the same error.....
03:10 PM roguish[m]: but hole_circle.ui does open with editvcp
03:13 PM AaronDargelGitte: roguish: if hole_circle.ui opens without error, then you should see a Conversational G-code section in the widget box. Those are the controls that this project installs.
03:15 PM roguish[m]: I have a problem in editvcp, i think.
03:21 PM roguish[m]: Aaron Dargel (Gitter): for the drilling routines, there are 2 retract modes. g98 and g99. I see in the drill_ops.py you only have G98.
03:22 PM roguish[m]: http://linuxcnc.org/docs/devel/html/gcode/g-code.html#gcode:g98-g99
03:23 PM roguish[m]: Aaron Dargel (Gitter): also, not baggin on you. just helping out a bit.
03:38 PM TurBoss: morning
03:53 PM roguish[m]: Aaron Dargel (Gitter): ya may want to add another input box to choose between G98 and G99.....just a suggestion
04:15 PM * roguish[m] uploaded an image: image.png (191KiB) < https://matrix.org/_matrix/media/r0/download/matrix.org/aroxoAYfyWTmMtnJrCQbQyPC/image.png >
04:15 PM roguish[m]: well, it is a bit toasty today. 110 now.
04:52 PM hazzy-m: roguish: I guess you could potentially save a little time with G99 if you optimized the retracts? Other than that it seems like G98 would be fine the vast majority of the time
04:54 PM hazzy-m: looks like he's setting the retract height with the initial Z pos
04:59 PM AaronDargelGitte: yeah. the clearance height will be used for the retract on a g98, i can add g99 as an option which will then use the retract distance.
04:59 PM roguish[m]: it's just an option.........
05:00 PM roguish[m]: personally I'm a chicken and retract high....
05:00 PM AaronDargelGitte: should be easy to do. seems like it would be useful to have.
05:00 PM roguish[m]: carbide drill bits are too pricey.....
05:02 PM roguish[m]: hazzy: yeah, i think that's what it's for. at least that is what my Haas programming manual says.
05:03 PM AaronDargelGitte: yeah, the R value is set to retract height in the G-code, but because of g98 it will use the Z height prior to the drilling op, which is set to the clearance height.
05:04 PM hazzy-m: Aaron Dargel (Gitter): trying to come up with a way to make it easier to keep everything in sync as you develop. What about making the entire conversation page in ProbeBasic a widget, made up of the individual widgets? Then it can be dropped in PB once, and as the conversational package gets updated it will automatically update in PB
05:05 PM JT-Shop: hazzy-m, does the debug comment work with qtpyvcp?
05:05 PM hazzy-m: JT-Shop: yes!
05:06 PM hazzy-m: right, we need a way to stop the backplot
05:06 PM AaronDargelGitte: hazzy: sounds good to me.
05:06 PM JT-Shop: yes I see it now :)
05:06 PM * hazzy-m got distracted trying to unstick the F350 service truck engine again
05:10 PM JT-Shop: stuck piston?
05:11 PM TurBoss: there are issues with the backplot?
05:12 PM JT-Shop: need a way to turn it off either in g code or a button for endless loops
05:15 PM TurBoss: oh
05:15 PM TurBoss: we can unregister the file watcher
05:18 PM TurBoss: using debug, eval we could call a backplot function that does that
05:18 PM TurBoss: or a button?
05:22 PM JT-Shop: a debug would be nice
05:22 PM TurBoss: cool
05:22 PM JT-Shop: or a button too
05:22 PM TurBoss: :)
05:22 PM JT-Shop: :)
05:55 PM hazzy-m: TurBoss: Unrelated, but what about using a separate path actor for each tool? Then we could hide/show the part each tool, and even show them in different colors
05:56 PM TurBoss: interesting
05:58 PM hazzy-m: JT-Shop: stuck something, but a 24in breaker bar on the crank, and it still won't budge. That's after sitting for a month with marvel oil in the cylinders and crankcase full of 16quarts of used diesel engine oil. Hoping the H2SO4 in the oil would help loosen things up, I guess not :(
05:59 PM JT-Shop: wow that's stuck, looks like a tear down now
05:59 PM * TurBoss uploaded a video: disable_plot.webm (1119KiB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/jqaJtZevNFskhfAryvWemFQs/disable_plot.webm >
06:00 PM JT-Shop: sweet
06:01 PM * TurBoss sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/KSWzlELwaSiIxQNTRLeGtkpn/message.txt >
06:02 PM TurBoss: in load_program function
06:02 PM TurBoss: wich i don't think is the best
06:02 PM TurBoss: 🤔
06:03 PM hazzy-m: TurBoss: can we make it like axis? Maybe something like (PLOTTER, stop) (PLOTTER, start) will turn off the back plot between those comments?
06:04 PM hazzy-m: I have an idea, sec
06:04 PM JT-Shop: nice
06:04 PM AaronDargelGitte: Drilling retract mode option:
06:04 PM AaronDargelGitte: https://github.com/kcjengr/qtpyvcp.conversational-gcode/pull/3
06:05 PM TurBoss: <hazzy-m "I have an idea, sec"> eval debug does that
06:05 PM TurBoss: I created a slot wich can be accessed from there
06:05 PM TurBoss: but I wait :)
06:05 PM TurBoss: <hazzy-m "TurBoss: can we make it like axi"> response was to this one ^^
06:06 PM hazzy-m: Aaron Dargel (Gitter): that was quick!!
06:06 PM hazzy-m: roguish: ^
06:07 PM TurBoss: how is (PLOTTER, start/stop) defined
06:07 PM TurBoss: its on AXIS or linuxcnc
06:08 PM hazzy-m: I think axis
06:09 PM TurBoss: then is python
06:09 PM JT-Shop: http://linuxcnc.org/docs/2.7/html/gui/axis.html#axis:preview-control
06:10 PM * JT-Shop wanders up to the deck to fire up the barbie
06:20 PM TurBoss: cant figure from where it comes
06:20 PM TurBoss: linuxcnc <-> python is a mistery to me
06:20 PM TurBoss: found
06:20 PM TurBoss: https://github.com/LinuxCNC/linuxcnc/blob/a6c727dd6ce21a292198274bb0f688119f5e2317/lib/python/rs274/glcanon.py#L131
06:29 PM TurBoss: hazzy: !!!! just added:
06:29 PM TurBoss: ```
06:29 PM TurBoss: def comment(self, msg):
06:29 PM TurBoss: to VTK
06:29 PM TurBoss: i'm able to read coments!!!!
06:29 PM TurBoss: \o/
06:29 PM * TurBoss sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/xNIJUnefEcJjqgNzVmCIIQKu/message.txt >
06:37 PM hazzy-m: TurBoss: hurray!
06:46 PM * TurBoss uploaded an image: Captura de pantalla de 2020-09-08 01-45-31.png (87KiB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/sZfEMXQpMssqVBQJjVNNHqef/Captura de pantalla de 2020-09-08 01-45-31.png >
06:46 PM TurBoss: comment and message ar 2 diferent functions
06:50 PM TurBoss: maybe add (color:red)
06:50 PM TurBoss: will change path to red
06:54 PM TurBoss: or (eval, stuff)
06:56 PM TurBoss: * or (eval, expresion)
06:59 PM TurBoss: but that will only work if there is a plotter :/
07:00 PM TurBoss: and PB has 2
07:03 PM Not-8b99: [02qtpyvcp] 07TurBoss pushed 031 commit to 03vtk_commands [+0/-0/±2] 13https://git.io/JUneQ
07:03 PM Not-8b99: [02qtpyvcp] 07TurBoss 036ad2886 - startup point
07:03 PM Not-8b99: [02qtpyvcp] 07TurBoss created branch 03vtk_commands - 13https://git.io/fhCou
07:05 PM travis-ci: kcjengr/qtpyvcp#1756 (vtk_commands - 6ad2886 : TurBoss): The build passed.
07:05 PM travis-ci: Diff view : https://github.com/kcjengr/qtpyvcp/commit/6ad28869164d
07:05 PM travis-ci: Build details : https://travis-ci.org/kcjengr/qtpyvcp/builds/725086589
07:05 PM TurBoss: is this stuff compatible with your idea hazzy ?
07:05 PM hazzy-m: yes!!
07:06 PM TurBoss: maybe def comment can be added to other place in the open signal
07:16 PM hazzy-m: JT-Shop: sure enough, that program does lock up the GUI
07:43 PM Not-8b99: [02qtpyvcp] 07KurtJacobson pushed 031 commit to 03vtk_commands [+0/-0/±2] 13https://git.io/JUnvV
07:43 PM Not-8b99: [02qtpyvcp] 07KurtJacobson 034e16dbe - ENH: VTKBackplot - Add support for AXIS style backplot control comments
08:08 PM Not-8b99: [02qtpyvcp] 07KurtJacobson created branch 03vtk_commands 13https://git.io/JUnf3
08:09 PM Not-8b99: [02qtpyvcp] 07KurtJacobson deleted branch 03vtk_commands
08:09 PM Not-8b99: [02qtpyvcp] 07KurtJacobson deleted branch 03vtk_commands - 13https://git.io/fhCou
08:10 PM travis-ci: kcjengr/qtpyvcp#1758 (vtk_commands - 4e16dbe : Kurt Jacobson): The build has errored.
08:10 PM travis-ci: Diff view : https://github.com/kcjengr/qtpyvcp/compare/4e16dbe4de92...4e16dbe4de92
08:10 PM travis-ci: Build details : https://travis-ci.org/kcjengr/qtpyvcp/builds/725096917
09:02 PM TurBoss: ops you deleted?
09:05 PM Not-8b99: [02qtpyvcp] 07KurtJacobson pushed 033 commits to 03master [+0/-0/±6] 13https://git.io/JUnUU
09:05 PM Not-8b99: [02qtpyvcp] 07TurBoss 036ad2886 - startup point
09:05 PM Not-8b99: [02qtpyvcp] 07KurtJacobson 034e16dbe - ENH: VTKBackplot - Add support for AXIS style backplot control comments
09:05 PM Not-8b99: [02qtpyvcp] 07KurtJacobson 03c57bb92 - Merge branch 'vtk_commands'
09:05 PM hazzy-m: TurBoss: it failed to push, strange
09:05 PM TurBoss: :O
09:06 PM TurBoss: but does it work?
09:06 PM hazzy-m: yes!
09:06 PM TurBoss: lol HURRAY!
09:06 PM hazzy-m: JT should be happy
09:07 PM Not-8b99: [02qtpyvcp] 07KurtJacobson pushed 031 commit to 03gh-pages [+523/-0/±0] 13https://git.io/JUnUm
09:07 PM TurBoss: do suppres -= 1 work fine ?
09:07 PM Not-8b99: [02qtpyvcp] 07traviscibot 03877feb8 - Deploy kcjengr/qtpyvcp to github.com/kcjengr/qtpyvcp.git:gh-pages
09:07 PM TurBoss: lol
09:07 PM TurBoss: nice!
09:07 PM hazzy-m: it seems to, I didn't have add much code, must if it was already there in the base_backplot
09:08 PM hazzy-m: * it seems to, I didn't have add much code, most if it was already there in the base_backplot
09:08 PM hazzy-m: * it seems to, I didn't have add much code, most of it was already there in the base_backplot
09:08 PM TurBoss: Glorious victory awaits
09:17 PM TurBoss: just tested here
09:17 PM TurBoss: thanks!!
09:19 PM hazzy-m: It seems fine?
09:19 PM TurBoss: (AXIS,hide/show) works fine
09:20 PM TurBoss: hazzy: i'm trying a wild experiment
09:20 PM TurBoss: add PrintCanon to something
09:20 PM TurBoss: to get commenets
09:20 PM TurBoss: so we can eval
09:21 PM TurBoss: what can be the subject?
09:22 PM TurBoss: * to get comments
09:23 PM TurBoss: meanwhile i'll try what you said about the path colors
09:46 PM hazzy-m: TurBoss: yes, we could run the file thru the PrintCannon, but that will evel everything when the file is first loaded, not as it is run
09:47 PM hazzy-m: a solution would be to queue up the eval statements basd on line number, and then eval them based on the current line number, but that would not be 100% accurate
09:48 PM TurBoss: ok
09:48 PM hazzy-m: that's an interesting idea though ...
09:48 PM * hazzy-m must sleep!
09:48 PM hazzy-m: gn8
09:49 PM TurBoss: n8