#emc-devel | Logs for 2011-04-09

Back
[00:03:33] -!- theorbtwo has quit [Read error: Operation timed out]
[00:03:39] theorb is now known as theorbtwo
[00:25:47] EDocTooR_ is now known as EDocTooR
[00:29:23] -!- roberth_ has quit [Ping timeout: 246 seconds]
[01:08:00] -!- crazy_imp has quit [Read error: Operation timed out]
[01:24:44] -!- ve7it has quit [Read error: Connection reset by peer]
[01:33:09] -!- ve7it [[email protected]] has joined #emc-devel
[01:41:30] -!- sumpfralle has quit [Ping timeout: 248 seconds]
[01:43:52] -!- Tom_shop has quit []
[02:44:02] -!- garage_seb [[email protected]] has joined #emc-devel
[03:47:53] -!- rooks has quit [Ping timeout: 240 seconds]
[04:01:21] -!- brendanjerwin has quit [Quit: brendanjerwin]
[04:05:00] -!- brendanjerwin has quit [Client Quit]
[04:49:31] -!- emc2-buildmaster has quit [Ping timeout: 260 seconds]
[04:53:05] -!- emc2-buildmaster [[email protected]] has joined #emc-devel
[05:09:16] -!- ve7it has quit [Remote host closed the connection]
[05:24:21] rooks_ is now known as rooks
[05:34:52] -!- Valen has quit [Quit: Leaving.]
[05:42:51] -!- psha [[email protected]] has joined #emc-devel
[06:10:21] -!- nullie has quit [Quit: Ex-Chat]
[06:47:49] -!- EDocTooR has quit [Read error: Connection reset by peer]
[06:51:46] -!- EDocTooR [[email protected]] has joined #emc-devel
[06:53:09] -!- isssy has quit [Quit: Visitor from www.linuxcnc.org]
[07:19:14] -!- jstenback has quit [Read error: Connection reset by peer]
[07:21:21] -!- capricorn_one has quit [Remote host closed the connection]
[08:11:28] -!- EDocTooR has quit [Read error: Connection reset by peer]
[08:15:40] -!- EDocTooR [[email protected]] has joined #emc-devel
[08:19:34] -!- Connor has quit [Quit: Leaving.]
[08:21:19] -!- garage_seb has quit [Quit: Leaving]
[08:29:26] -!- mhaberler [[email protected]] has joined #emc-devel
[08:37:29] -!- psha has quit [Read error: Connection reset by peer]
[08:38:40] -!- psha [[email protected]] has joined #emc-devel
[08:55:52] -!- robh__ [[email protected]] has joined #emc-devel
[09:02:56] -!- Dannyboy has quit [Remote host closed the connection]
[09:12:43] <mhaberler> jthornton: around?
[09:25:58] <psha> heh, asciidoc refuses to use strings with colons (like sec:Halcmd) as section id's
[09:26:01] <psha> crap...
[09:52:15] -!- CO19_KUL_CARI_CO [[email protected]] has joined #emc-devel
[09:52:15] -!- CO19_KUL_CARI_CO [[email protected]] has parted #emc-devel
[09:58:29] -!- mhaberler has quit [Quit: mhaberler]
[10:15:34] -!- mhaberler [[email protected]] has joined #emc-devel
[10:20:30] <mhaberler> ah. You noticed. I've started fixing some of these already.
[10:30:42] <psha> i'm sitting second day inside asciidoc and since have not found fix for that...
[10:49:00] <mhaberler> well, why not rewrite refs from '<<ref:foo>>', to '<<ref:foo,foo>>'?
[10:49:37] <mhaberler> emacs defun and off you go ;-)
[10:50:15] <mhaberler> jthornton: post-coffee mode?
[10:57:23] <jthornton> waiting on the coffee now
[10:57:54] <psha> mhaberler: problem is not in refs
[10:58:28] <psha> in links
[10:58:30] <psha> but in anchors
[10:58:37] <psha> there is construction named BlockId
[10:58:45] <psha> that overrides section id
[10:58:56] <psha> but if you have : in it - boom
[11:17:12] -!- mhaberler has quit [Quit: mhaberler]
[11:26:55] -!- mhaberler [[email protected]] has joined #emc-devel
[11:28:23] <mhaberler> jthornton - q: do you do doc changes straight off master/v2.x_branch, or do you have a documentation branch? I just started factoring all docs work into a documentation branch because it becomes hairy pulling doc-only changes together from different devel branches
[11:29:07] <jthornton> I've been doing it on the branch unless it applies to master only
[11:29:23] <mhaberler> I see
[11:30:17] <jthornton> when they do a merge thingy the branch stuff goes to master I think
[11:32:08] <mhaberler> yes, on branch merge
[11:32:08] <mhaberler> docs need some more work on the plumbing side, like what psha's working on, and stuff like mscgen or svg images
[11:32:08] <mhaberler> one possible organisation would be branches like so:
[11:32:09] <mhaberler> off master:
[11:32:09] <mhaberler> documentation
[11:32:09] <mhaberler> docs-tools
[11:32:09] <mhaberler> documentation is rebased on master to get it to format properly
[11:32:09] <mhaberler> the v2.[45]branches dont need that methinks
[11:33:17] <mhaberler> the reason for documentation branch is - it's quite massive changes and it touches many files, so I dont want to incrementally push that into master
[11:33:47] <mhaberler> psha: how do you think about that?
[11:33:47] <mhaberler> these branches could be on git.linuxcnc.org
[11:34:09] <jthornton> atm, I really don't understand what you said :/
[11:34:48] <mhaberler> documentation tools need work, and while being worked on break documentation
[11:35:15] <jthornton> yes, I see that
[11:35:34] <mhaberler> but they are different items so leave master with working documentation and tools, and shift doc-tools and documentation work to separate branches
[11:36:28] <jthornton> the lyx docs?
[11:36:51] * mhaberler these are history IMV and should be removed to prevent confusion
[11:37:02] <mhaberler> oops wrong key
[11:37:17] <mhaberler> I think we should exclusively work on .txt
[11:37:26] <jthornton> ok
[11:37:49] <jthornton> should I not work on docs until psha has the plumbing finished?
[11:38:13] <mhaberler> example: I added highlighting for gcode and sh in gcode/main.txt in my 'documentation' branch, see http://emc.mah.priv.at/docs/html/gcode/main.html#sec:Subroutine-return-values e.g. ocodes section or M100
[11:39:00] <jthornton> the g code?
[11:39:01] <mhaberler> this is a case of simultaneous work on fixing source-highlight and changing .txt
[11:39:51] <mhaberler> well look at the top, Polar coordinates gcode example
[11:40:11] <mhaberler> that has now syntax-driven highlighting
[11:40:34] <jthornton> cool
[11:41:42] <mhaberler> the next issues are mscgen and svg images, see for instance this image (forget the text, its BS): http://emc.mah.priv.at/docs/html/gcode/tool_handling.html#_abstract_tool_geometry
[11:42:14] <mhaberler> this is a svg image and it would be great if we could pull the svg into the pdf instead of converting to png before
[11:42:48] <jthornton> nice image and very clear
[11:43:43] <mhaberler> my inkscape learning example
[11:47:25] <jthornton> the front angle and back angle are from a line parallel with the X axis?
[11:52:56] <mhaberler> yes, that is what I reverse-engineering from existing docs
[11:53:49] <jthornton> that makes more sense than the pitiful drawings I did
[11:54:20] <jthornton> so a right hand tool is the same but opposite?
[11:55:20] <mhaberler> this is what I took as a base: http://www.linuxcnc.org/docs/2.4/html/lathe_lathe-user.html
[11:55:51] <mhaberler> figure: lathe tool orientations
[11:56:55] <jthornton> I did that one in an attempt to understand what front angle and back angle was by trying different numbers then looking at Axis after loading the tool... quite time consuming
[11:57:02] <mhaberler> this just defines angle start and use, all the orientations would be something similar to what you had
[11:57:28] <mhaberler> I guess axis is helpful in showing how it's handled
[11:58:04] <jthornton> yes, AFAIK it is only used for the tool display in the backplot of Axis
[11:58:30] <mhaberler> well, it could be used for more, like gauging detection
[11:58:41] <mhaberler> but thats more of a CAM issue
[11:59:53] * mhaberler anyway what I'll do is
[11:59:53] * mhaberler - create a documentation branch with my stuff that is ready for preview
[11:59:53] * mhaberler - create a doc-tools branch where probably me, Pavel and Jeff would fixup new features
[12:02:17] <jthornton> ok, since I'm only fixing things that seemed to get hosed during the conversion to .txt and adding missing text and stuff like that I assume my work should be on branch 2.5
[12:02:28] <mhaberler> I need to spin out doc-tools from documentation first and clean docs so far , then I'll do these branches later this weekend
[12:02:46] <mhaberler> I think those can be rebased upon easily, dont worry about that
[12:03:11] <mhaberler> when that stuff reaches master I'll rebase and fix conflicts
[12:03:19] <jthornton> ok
[12:03:52] <mhaberler> eventually when the tool_handling.txt stuff and doc-tools are ready to go I'll merge both into master
[12:04:07] <mhaberler> this is really for psha - where to do docs-tools changes
[12:04:19] <jthornton> ok
[12:04:49] <jthornton> lol, the more I look at my drawing of lathe tool orientations the more confused I get
[12:04:59] <mhaberler> se..
[12:05:02] <mhaberler> see..
[12:05:29] <mhaberler> good - will let you know when the new branches are done; I always have the output of my documentation branch at http://emc.mah.priv.at/docs/html/
[12:06:57] <jthornton> ok, I've got that bookmarked here now
[12:06:59] <mhaberler> let me see.. I think we need to do a laundry list of docs-tools changes: links syntax, pdf links, svg images, mscgen,
[12:07:00] <mhaberler> let me know if you encounter any wrong highlighting examples, I fixed five or so bugs this morning
[12:07:23] <jthornton> what is mscgen?
[12:07:26] * mhaberler aja, html output comparison...
[12:07:34] <mhaberler> ahttp://emc.mah.priv.at/docs/html/
[12:07:50] <mhaberler> revision bars for html: aja, html output comparison...
[12:08:31] <mhaberler> these graphs: http://emc.mah.priv.at/docs/html/gcode/tool_handling.html#change-M6-phase
[12:08:52] <jthornton> ah, ok
[12:09:21] <mhaberler> scroll down in http://emc.mah.priv.at/docs/src/gcode/tool_handling.txt to find the source for that example
[12:09:21] <jthornton> be nice to have one that describes what happens when EMC starts up
[12:09:31] <mhaberler> uh, yes
[12:09:43] <mhaberler> I think that would be more of a flowchart type thing
[12:10:07] <jthornton> can you do a flow chart?
[12:10:08] <mhaberler> there's a way to do that too and it would be handy to have to add such charts to the manual
[12:10:21] <mhaberler> yes I did locally, let me see..
[12:14:39] <mhaberler> http://static.mah.priv.at/public/asciidoc-graphviz-sample.html
[12:15:08] <mhaberler> ah, this was it: http://static.mah.priv.at/public/ditaa.html (just a start)
[12:15:56] <jthornton> cool
[12:16:08] <mhaberler> that would be an option too: http://wiki.asciidoc-aafigure-filter.googlecode.com/hg/asciidoc-aafigure-readme.html
[12:16:47] <mhaberler> matplotlib is great but I don see much use for something like this, do you? http://volnitsky.com/project/mplw/index.html
[12:17:40] <jthornton> can't think of anything atm
[12:18:12] <jthornton> aafigure looks more flexible to me
[12:18:17] <mhaberler> but I think we'll have to do a separation as to what gets converted during docs buld and what is converted separately, like the old dxf examples
[12:18:17] <mhaberler> for intsnace converting svg to png during build is possible with inkscape but a tad slow
[12:18:29] <mhaberler> inkscape that is
[12:19:23] <mhaberler> so it would be 'create output from illustrations' a single pre-build step, and docs build would use the pre-built output but wouldnt convert everything on each reformat
[12:19:27] <jthornton> seems best to convert only once manually to svg
[12:19:49] <mhaberler> right, but that needs adapting the filter conf.. Paaaavel ;-)
[12:20:11] <mhaberler> thats the kind of thing for docs-tools
[12:20:53] <mhaberler> I wonder if anybody in the developers crowd can write latex style files, any suspects?
[12:21:34] <mhaberler> I was forced to use latex many moons ago but it still looks like a core dump to me
[12:22:01] <jthornton> I don't know of anyone
[12:22:14] <mhaberler> anyway, girlfriend drafts me for gardening ;-) so I'll have some time to think
[12:22:21] <jthornton> ok
[12:22:25] * mhaberler cu!
[12:22:46] -!- mhaberler has quit [Quit: mhaberler]
[12:30:19] -!- MOGLI has quit [Ping timeout: 260 seconds]
[12:42:18] -!- sumpfralle has quit [Ping timeout: 248 seconds]
[12:52:06] -!- cv has quit []
[12:54:32] -!- lilalinux has quit [Quit: deswahnsinns.de]
[13:14:42] <jthornton> ah ha!!! it don't make any difference which is the frontangle and backangle
[13:15:06] <archivist> wha...
[13:15:40] <jthornton> on the lathe tool table you can swap the numbers for frontangle and backangle with no apparent effect
[13:16:21] <archivist> that feels wrong
[13:16:57] <jthornton> but it makes sense why I had so much trouble before trying to sort it out...
[13:17:16] <archivist> Ive had serious trouble too
[13:17:19] <jthornton> it is just two angles from Z+ in a clockwise direction
[13:18:08] <archivist> except front/back have special meaning to a normal lathe tool user
[13:18:34] <jthornton> how is that?
[13:21:32] <archivist> I need a drawing to show
[13:21:42] <jthornton> ok
[13:23:49] <jthornton> AFAIK the frontangle and backangle are only used to display the tool in axis
[13:24:21] <jthornton> I'm not sure if the orientation has any effect on anything except the backplot tool display
[13:24:49] * jthornton wishes someone who knew could spill the beans :)
[13:25:41] <archivist> what you just said does fit with the numbers on http://www.linuxcnc.org/docview/html/lathe_lathe-user.html#r1_2
[13:26:16] <jthornton> yes, I did that trying to sort out what they meant
[13:26:46] <archivist> except for the words possibly being swapped on one diag
[13:27:16] <archivist> or ...
[13:27:19] <jthornton> the frontangle/backangle?
[13:27:22] <archivist> yes
[13:27:46] <jthornton> I just did some testing and it doesn't matter which is which
[13:28:27] <archivist> why, is axis swapping them?
[13:29:37] <jthornton> no, I was trying to figure out why I did the drawing that way and just swapped one to see and no change on the tool display so I did another with the same result
[13:30:42] <jthornton> trying to arrive at some logical description of how frontangle and backangle are... and the drawing confused me more.
[13:43:14] -!- jthornton has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.16/20110323162424]]
[14:06:27] -!- Valen has quit [Quit: Leaving.]
[14:25:31] -!- mhaberler [[email protected]] has joined #emc-devel
[15:16:57] -!- dgarr [[email protected]] has joined #emc-devel
[15:19:17] -!- brendanjerwin has quit [Quit: Colloquy for iPad - http://colloquy.mobi]
[15:35:19] -!- skunkworks [[email protected]] has joined #emc-devel
[15:44:15] -!- tom3p [[email protected]] has joined #emc-devel
[15:50:59] -!- psha has quit [Read error: No route to host]
[15:51:48] -!- psha [[email protected]] has joined #emc-devel
[15:58:15] -!- mhaberler has quit [Quit: mhaberler]
[16:01:48] -!- mhaberler [[email protected]] has joined #emc-devel
[16:06:14] -!- WalterN has quit [Read error: Connection reset by peer]
[16:07:47] -!- ve7it [[email protected]] has joined #emc-devel
[16:32:58] <tom3p> is there a map of sys vars? (like what is, say... #5161?, what is free? what is used?, what ranges are for what purpose?
[16:33:24] -!- me345 has quit [Read error: Connection reset by peer]
[16:44:41] <mhaberler> yes there is. It's merged but not in the online docs I think. http://emc.mah.priv.at/docs/html/gcode/overview.html#_a_id_sec_parameters_a_parameters_variables_in_g_code_programs
[16:44:41] <mhaberler> use that for now
[16:46:23] <mhaberler> rules.. well. 1-30 are subroutine params; I would use anything abouve 5000 as user values
[16:47:18] <mhaberler> note this page is my 'documentation work in progress' result so it reflects some stuff not merged into master, like the '#<_value>' parameter
[16:54:34] <mhaberler> I suggest you look at src/emc/rs274ngc/interp_array.cc - the required_parameters array tells most what's reserved;
[16:54:56] <mhaberler> I hope I caught all in the parameters chapter
[17:03:00] <tom3p> mhaberler, thanks, im writing a shuttle rack tool changer and the motion needs to be specail ( cant 3d move to toolchanger_position) so have a bundle of positions to store once tooling is trammed in.
[17:04:29] <mhaberler> if you are really adventuous you might try this branch, which collects all my current toolchange work; newbynobi currently makes his Heckler&Koch BA20 use it
[17:04:55] <mhaberler> http://git.mah.priv.at/gitweb/emc2-dev.git/shortlog/refs/heads/ns-toolchange-wip
[17:09:59] <tom3p> mhaberler, thx, i grabbed a snapshot
[17:12:02] <tom3p> in this implementation there is no T<N> M6 , there is no prep, no intermediate position, so M6 is nulled, yet allowed.
[17:12:23] <mhaberler> make me understand this?
[17:12:54] * mhaberler oh you mean your changer!
[17:13:18] <tom3p> the tdemo.ngc has all the motion , and io needed to unclamp and clamp the exiting tool, put it away, and grab the new tool
[17:13:33] <tom3p> / existing
[17:13:52] <mhaberler> sounds doable
[17:14:02] <tom3p> yes, thx to you!\
[17:14:20] <mhaberler> so you're using some of this already?
[17:15:01] <mhaberler> last days changes made it possible to fail a prepare or change in gcode, by returning a value > 0 or <= 0
[17:15:13] <tom3p> i do it in a heidhenhain plc i wrote , it has a half dozen toolchangers i wrote in it, and cfgd by params. i now rewrite for emc2
[17:15:34] <tom3p> yes the on_abort is a great feature
[17:15:55] <tom3p> i need to see if i can pass strings to error dialog
[17:16:11] <tom3p> or indexes to array of strings
[17:16:34] <mhaberler> you could pass indexes, yes, like through a pin, and have gladevcp do the right thing
[17:17:29] <tom3p> ah yes ( shamefaced) i have to change i/f from pyvcp to glade, i used glade already but made quik&dirty pyvcp at first for this
[17:18:04] <tom3p> back to it, thx!
[17:18:52] <mhaberler> well if you get stuck or need an example for some odd bit, drop me a line
[17:19:41] <tom3p> uh, one idea, i had probs on heid with memeory & initialization so made M10 code t forget any memory of tool in spindle ( chuck for me )
[17:19:42] <tom3p> and
[17:20:13] <mhaberler> well m61 set tool number will work in the ns-toolchange-wip branch
[17:20:33] <tom3p> the rules were... start with spindle empty, M10 at top of program, and all tools in rack ( let the damn tool changer change the tools! :)
[17:20:35] * mhaberler you can do M61 Q0 to 'unload toolÄ
[17:20:50] <tom3p> the idea of unloading a tool is cool
[17:21:25] <mhaberler> you could put that in rs274ngc_startup code to set any tool at startup inlcuding none
[17:22:04] <tom3p> but knowing whether some salesman had moved a tool in the carousel was problematic , (parking car in full garage... ouch )
[17:23:04] <mhaberler> now we're talking full scale image recogintion - that would be Pavel's remit I guess ;-)
[17:23:13] <tom3p> we added prox sensors and used probing moves to try to make it foolproof, but the fools always found a way
[17:23:39] <tom3p> i ended up using a simple rule
[17:23:40] -!- dgarr has quit [Ping timeout: 276 seconds]
[17:24:19] <tom3p> hey thx again, off to play with your cool stuff
[17:25:20] <mhaberler> ;-)
[17:35:29] -!- mhaberler has quit [Quit: mhaberler]
[17:37:40] -!- KimK has quit [Ping timeout: 246 seconds]
[17:39:15] -!- acemi has quit [Quit: WeeChat 0.3.2]
[17:47:00] -!- KimK [[email protected]] has joined #emc-devel
[17:56:34] -!- pcw_home has quit [Remote host closed the connection]
[17:58:07] -!- Ikarus has quit [Ping timeout: 276 seconds]
[18:08:27] -!- mhaberler [[email protected]] has joined #emc-devel
[18:09:48] -!- mhaberler has quit [Client Quit]
[18:17:50] -!- isssy has quit [Quit: Visitor from www.linuxcnc.org]
[18:49:05] -!- tom3p [[email protected]] has parted #emc-devel
[18:59:05] <alex_joni> hi all
[19:53:30] -!- micges [[email protected]] has joined #emc-devel
[19:53:43] -!- psha has quit [Quit: Lost terminal]
[20:02:37] -!- micges_ [[email protected]] has joined #emc-devel
[20:05:14] -!- micges has quit [Ping timeout: 258 seconds]
[20:06:16] -!- nullie has quit [Quit: Ex-Chat]
[20:16:18] -!- SteveStallings has quit []
[20:21:31] -!- SteveStallings [[email protected]] has joined #emc-devel
[20:22:22] SteveStallings is now known as steves_logging
[20:24:41] -!- micges__ [[email protected]] has joined #emc-devel
[20:27:05] -!- micges_ has quit [Ping timeout: 258 seconds]
[20:36:56] micges__ is now known as micges
[20:40:11] -!- me345 has quit [Read error: Connection reset by peer]
[21:04:17] -!- Dannyboy has quit [Ping timeout: 240 seconds]
[21:40:34] -!- garage_seb [[email protected]] has joined #emc-devel
[21:40:34] -!- garage_seb has quit [Client Quit]
[21:53:25] -!- Fox_Muldr has quit [Ping timeout: 276 seconds]
[21:57:56] -!- brendanjerwin has quit [Quit: Colloquy for iPad - http://colloquy.mobi]
[22:03:42] -!- alex_joni has quit [Read error: Operation timed out]
[22:04:00] -!- alex_joni [alex_joni!~alex_joni@emc/board-of-directors/alexjoni] has joined #emc-devel
[22:04:18] -!- izua has quit [Ping timeout: 260 seconds]
[22:05:35] -!- izua has quit [Changing host]
[22:05:42] -!- micges_ [[email protected]] has joined #emc-devel
[22:08:23] -!- micges has quit [Ping timeout: 260 seconds]
[22:11:54] -!- JT-Shop [[email protected]] has joined #emc-devel
[22:12:00] -!- micges [[email protected]] has joined #emc-devel
[22:12:57] -!- micges_ has quit [Ping timeout: 252 seconds]
[22:26:46] -!- brendanjerwin has quit [Quit: Colloquy for iPad - http://colloquy.mobi]
[22:37:59] -!- micges_ [[email protected]] has joined #emc-devel
[22:40:12] -!- micges has quit [Ping timeout: 252 seconds]
[23:02:08] -!- micges_ has quit [Ping timeout: 276 seconds]
[23:34:38] -!- robh__ has quit [Ping timeout: 276 seconds]
[23:56:00] -!- EDocTooR_ [[email protected]] has joined #emc-devel
[23:57:29] -!- EDocTooR has quit [Ping timeout: 240 seconds]