am I imagining that I remember (system,shell command here)?
maybe it was never in emc2
it is in emc1
(which always ran under sudo)
thanks, at least I wasn't crazy.
jepler: the motion.spindle-speed-cmd-rps is working great. i think I have the shifting comp to the point of testing on the machine. (still have to add spindle lock - but I don't see that being a big deal)
EMC: 03cmorley 07v2.4_branch * r57a26171e128 10/src/emc/usr_intf/pncconf/pncconf.py: Use constants in references to mesa data array
EMC: 03cmorley 07v2.4_branch * r3751b59d6799 10/src/emc/usr_intf/pncconf/ (pncconf.glade pncconf.py): Add 'board title' to mesa array
EMC: 03cmorley 07v2.4_branch * r205f28311ab2 10/src/emc/usr_intf/pncconf/pncconf.py: add directory entry to firmware array
EMC: 03cmorley 07v2.4_branch * rfc1c7e848515 10/src/emc/usr_intf/pncconf/pncconf.py: Fix loading of multiple mesa cards
EMC: 03cmorley 07v2.4_branch * r1f95eef45241 10/src/emc/usr_intf/pncconf/pncconf.glade: fix labels on second mesa page tabs
EMC: 03cmorley 07v2.4_branch * rb7656268a2c9 10/src/emc/usr_intf/pncconf/pncconf.py: Fix spindle encoder signals for threading / rigid tapping
EMC: 03cmorley 07v2.4_branch * r54f5a54d41f1 10/src/emc/usr_intf/pncconf/pncconf.py: fix layout of spindle signals in HAL file
CrabbyCoconut is now known as CoconutCrab
Does somebody know how I can get to know the time needed for a function that I have added to a thread in hal layer ?
I have the time for the thread from halcmd, show thread
but not for the function
maximilian_h: the parameters <function-name>.time, .tmin give information about the execution time of a hal function.
look at the pastebin moment please
Jepler: Thanks, that is what I was looking for
question answered, thanks
EMC: 03jthornton 07v2.4_branch * rd65f2ee6ef02 10/docs/src/hal/components.lyx: fix whitespace error
EMC: 03jthornton 07v2.4_branch * r30341a87170d 10/docs/src/hal/components.lyx: add new components
our debian/configure script fetches the distro name from /etc/lsb-release, then modifies our debian build deps based on what it finds
if it doesnt recognize the distro name it just adds "lyx-qt" by default
the only platform that we build-test on that's not recognized by debian/configure is debian lenny
which doesnt have lyx-qt
it does have just "lyx"
so i think i should add a "lenny" entry to debian/configure to say lyx instead of lyx-qt
but the default is probably wrong in general and i'm not sure if it's useful
please feel free to add lenny to debian/configure in v2.4
ok will do
also, there is a lot of overlap amongs our per-distro EXTRA_BUILD variables, maybe some should be moved into the regular build deps?
if it's fixing what's not broken, please do it on master -- but I'm in favor of that
confusingly, debian/control.in has this: lyx-qt | lyx
so that should probably go, to be replaced by whatever's coming from EXTRA_BUILD
another thing is to determine whether we need to ask for a specific version of libreadline5-dev
both the packaging platforms on the buildbot (hardy & lucid) are fine without a versioned dependency on libreadline5-dev
I meant whether we could remove the "5" and just ask for libreadline-dev (and get readline6 on 10.04)
JT-Work: does Master_Integrator.pdf build for you after your two recent 2.4 commits? it won't build at all for me on hardy or lenny, and I'm testing lucid now... (and yes i think i have the build deps installed right!)
Master_Integrator is the only lyx doc we have that includes hal/components.lyx, which you just touched
hmm, it builds fine on lucid
maybe some new feature of lyx was used, so it doesnt work on old lyxes?
i wish lyx would say *why* it balks on hardy
it makes me sad when users don't understand "it should do what I want, even though I understand that I'm telling it to do something other than what I want" is not a design a programmer can implement, and I can't seem to explain why that is
seb_kuzminsky: looking into the lyx thing
* jepler wonders where he has an 8.04 machine to test on :-/
pretty sure I have (a slow) one online
if you're gonna bail with an error exit code, please please tell me why!!
seb_kuzminsky or cradek: can you try reverting d65f2ee6ef023baca50d57ca8c5078d30d397515 and see if it builds
* seb_kuzminsky frowns at lyx
jepler: not until after lunch
I think that lyx files require trailing whitespace some places, and bjt was trying to appease git
it's just impossible to do that without eating first
man, fuck git and its whining about whitespace
commit 30341a... fails in the same way
if git was written in python would it whine that you used tabs instead of spaces?
that is so much not its job
seb_kuzminsky: OK, thanks for checking
v2.4_branch HEAD^2 works fine, HEAD^1 and HEAD do not work
all's fine on lucid, it's just the older lyxes on hardy & lenny that have this problem
yay for buildbot not being broken and actually finding a real problem this time
"lyx -dbg any" make it spit out a ton of debug info, some of which may be useful to folks who know lyx
here's a failing run on lenny: http://bioserve.colorado.edu/~seb/lyx-dbg.txt
there's some messages down near the bottom that say things like "Log line: ! Missing $ inserted.
Desc: Missing $ inserted.
I've inserted a begin-math/end-math symbol since I think
you left one out. Proceed, with fingers crossed.
problem? i dont know
i'm stopping work on this now and waiting for someone who knows lyx to take a look at it, let me know if i can help more somehow
EMC: 03seb 07v2.4_branch * rb76294cae734 10/debian/configure: update debian/configure script for Debian Lenny
cradek: it's just a smop!
really I think it's just my inability to explain things
i think you explain very clearly, and the communication problem lies elsewhere...
my command of the language is good and all, but I don't know how to explain to a nonprogrammer why a programming problem is hard or impossible
what I want to happen is that everyone takes my examples and generalizes them and imagines the various problem cases that come about
but nonprogrammers don't do that kind of thing. they just see the single current problem case that has an "obvious" solution and the error message they wish they didn't have.
I think people don't undersand how hard it is. You would have to read an infinatly far ahead to see if you where going to gouge in any line segments.
yes that's definitely one of the problems
and find the line segment that 'is' the correct tangent for the 'correct' corner
I wouldn't know where to start :)
even if you read infinitely far ahead, pretty much every contour path overlaps itself, at the beginning/end. you'd have to know that's the overlap you want
so you'd maybe only want to read a little bit ahead?
and I can't be convinced that throwing out part of a program and continuing with the rest is safe
that's way into you-haven't-thought-this-through land
I would think your solution of offsetting the path and using a 0 diameter tool (then changing +/- a little ) seems like a good one.
yes I think that would really help him
I do not know if he understood it. I tried twice to explain it.
now the repo wont build on 2.4 or master... in 2.4 there's the old-lyx bug, and in master there's the bogus-builddeps problem
we can't easily merge the lenny-builddep fix into master without also getting the broken-lyx commit
if you want to make a new merge base, you can merge and then revert on master the commits you don't want
(I don't know anything about the changes, that's just a comment about use of git)
I don't think that's a good suggestion right now
the lyx problem is bad for both branches so merging it and then reverting it in only one branch is not the right answer
I've just about got a fresh git on my 8.04 machine so I'll look into the real lyx problem
cradek: maybe three times' the charm? (re: tool path compensated in the cam package)
Daniel G's advice is excellent
oh is that you? haha
i have my good moments
now if i just learn to explain my emc problems to you and hear "excellent, i know what you did wrong, quick fix:" instead of "huh?" :P
"ummmm can you ask a more specific question?"
asking questions well is as hard as answering them
i just need to stop asking questions moments after my boss dropped me a new goodie to research, but i admit, when documentation is spread over html and wiki, and sometimes one is old and not the other, i do fail more often to find the good info i want, once school starts, meaning when i have true "off time" i will just have to start soaking it all in
summer break is too long
sometimes a question that leads to a specific doc url is fine. it can be hard to find the right thing unless you already kind of know where it is.
anyways, you made my day with complementing Daniel G :P
as long as the asker knows that a url is not a blowoff
mozmck's 10.04 build info trips me out, i STILL don't see it on the wiki page with all the other older and the 9.04 docs
urls are great, even if i still feel dumb when the url is something i think i should have found myself and didnt
you know what, i go get a oilchange and while being there, call the Siemens guy, i need him to hook me up with a good local distributor, and i never get around to calling him at work, not wehn running machines especially
morfic: my build info is linked at the top of the main wiki page. In the first section there's a link to Ubuntu10.04Notes - that's it.
awe man, don't put it way up on top with big bold letters, where i just scroll down to the links and go there: http://wiki.linuxcnc.org/emcinfo.pl?Installing_EMC2
Maybe I should put a link from the Installing_EMC2 page.
someone should decruft the Installing_EMC2 page
seb_kuzminsky: I can't try from here I'll have to try when I get home
there is a lathe todo that talks about stuff that is implemented now, wiki decrufting can turn into a full time job
EMC: 03jepler 07v2.4_branch * rcf87de9e0e88 10/docs/src/hal/components.lyx: Fix html doc building error introduced in 30341a8
EMC: 03jepler 07v2.4_branch * r3c8a877ded15 10/docs/src/hal/components.lyx: Fix doc build error introduced at 30341a87
seb_kuzminsky: let the buildbot chew on that for a bit
EMC: 03jepler 07v2.4_branch * r6e4d25c5c442 10/docs/src/lyxtree.py: get rid of unwanted underscores in labels, indexes, and references
Merge made by recursive.
src/emc/usr_intf/pncconf/pncconf.glade |36872 ++++++++++++++++----------------
I'm absolutely stunned that one of these glade megachanges actually merge
morfic: feel free to update wiki
jeff for the win
seb_kuzminsky: are all the 2.4 builds OK now? If so, one of us should go ahead and push the merge of v2.4_branch into master
I can sure do it, I've already done the merge itself
it got past building the pdfs, which is where it choked before
the debs are not finished building yet, but soon
your last commit, "get rid of unwanted underscores in labels, indexes, and references", hasnt started building yet, do you want to wait for that one to move through?
EMC: 03jepler 07master * r44bcdf232c1c 10/src/emc/usr_intf/pncconf/pncconf.py: fix sensitivity of invert encoder
EMC: 03jepler 07master * r441b0d8478e1 10/ (9 files in 6 dirs): Merge branch 'v2.4_branch' of ssh://email@example.com/git/emc2 into v2.4_branch
EMC: 03jepler 07master * r16b99eb3439a 10/src/emc/usr_intf/pncconf/pncconf.py: fix tool table to work with random tool changer
EMC: 03jepler 07master * r7b056ead747a 10/ (9 files in 7 dirs): Merge branch 'v2.4_branch' of ssh://firstname.lastname@example.org/git/emc2 into v2.4_branch
EMC: 03jepler 07master * r573a7ffd9547 10/src/emc/usr_intf/pncconf/pncconf.py: fix typo from last commit to pncconf
EMC: 03jepler 07master * r4dc63b266db9 10/docs/src/common/machining_center.lyx: add missing numbered parameters to list
seb_kuzminsky: I'll take my chances
EMC: 03jepler 07master * rfe501a7cc4ca 10/docs/src/hal/components.lyx: add component info
EMC: 03jepler 07master * ra089185d0081 10/docs/src/gcode/main.lyx: seperate g33 and g33.1
EMC: 03jepler 07master * re30574bc5122 10/src/configure.in: Linux 2.6.34+ moved utsrelease.h
EMC: 03jepler 07master * r205f28311ab2 10/src/emc/usr_intf/pncconf/pncconf.py: add directory entry to firmware array
EMC: 03jepler 07master * r3751b59d6799 10/src/emc/usr_intf/pncconf/ (pncconf.glade pncconf.py): Add 'board title' to mesa array
EMC: 03jepler 07master * r57a26171e128 10/src/emc/usr_intf/pncconf/pncconf.py: Use constants in references to mesa data array
EMC: 03jepler 07master * rfc1c7e848515 10/src/emc/usr_intf/pncconf/pncconf.py: Fix loading of multiple mesa cards
EMC: 03jepler 07master * r1f95eef45241 10/src/emc/usr_intf/pncconf/pncconf.glade: fix labels on second mesa page tabs
EMC: 03jepler 07master * rb7656268a2c9 10/src/emc/usr_intf/pncconf/pncconf.py: Fix spindle encoder signals for threading / rigid tapping
EMC: 03jepler 07master * r54f5a54d41f1 10/src/emc/usr_intf/pncconf/pncconf.py: fix layout of spindle signals in HAL file
EMC: 03jepler 07master * r30341a87170d 10/docs/src/hal/components.lyx: add new components
EMC: 03jepler 07master * rd65f2ee6ef02 10/docs/src/hal/components.lyx: fix whitespace error
EMC: 03jepler 07master * rb76294cae734 10/debian/configure: update debian/configure script for Debian Lenny
EMC: 03jepler 07master * r3c8a877ded15 10/docs/src/hal/components.lyx: Fix doc build error introduced at 30341a87
EMC: 03jepler 07master * r6e4d25c5c442 10/docs/src/lyxtree.py: get rid of unwanted underscores in labels, indexes, and references
EMC: 03jepler 07master * rfbee58ea999d 10/ (7 files in 6 dirs): Merge remote branch 'origin/v2.4_branch'
EMC: 03jepler 07master * rcf87de9e0e88 10/docs/src/hal/components.lyx: Fix html doc building error introduced in 30341a8
jepler: you are a crazy daredeveil who lives on the edge
seb_kuzminsky: you must have me confused with someone who cares much whether master builds or not :-/
jepler: what was the doc error?
JT-Work: two problems. First, apparently it's not OK to put a "_" in index entries without putting a "\" first.
(that tripped up building the pdf documentation on 8.04 only)
ok, thanks for fixing that
second, there was a smart apostrophe that screwed up the HTML documentation only; I replaced it with a plain ' and that fixed it.
don’t vs don't
thanks, I thought I caught all of the smart quotes but I guess not.
JT-Work: if git ever says there are "whitespace errors" in lyx files, just ignore it and go on with your life
seemed like it was only in lyx-code sections that it cared
at least we got a troublesome automatic up and running at the customers place today so time for a cold one for me
thanks for working on the docs .. it's worth the occasional minor hiccough.
micges forgot one thing, knowing a little is dangerous, wouldn't want to remove superseded stuff that turns out it is very much not superseded, i'd wait until i know i am right rather than think i am right ;)
Anybody know if there's a reason we have this channel set +s ("secret")
because we're grumpy and dont like visitors?
speak for yourself
I'm grumpy AND bitter
ChanServ: whatever you say
if you change a setting in the synaptic package manager will it stick when you make a livecd?
like settings/repositories/Updates/Release Upgrade to Never