#emc-devel | Logs for 2010-12-03

[00:51:20] <andypugh> addr2line is brilliant. :-)
[00:52:15] <andypugh> It just saved me a lot of head scratching.
[00:54:45] <voxadam> Ancient nix tool to the rescue!
[00:55:10] <andypugh> Yeah, it could be a little less manual, but it does the tick.
[01:04:40] <andypugh> Need a daemon watching the dmesg output and auto-running addr2line when it spots "Vector 14" :-)
[01:23:24] <voxadam> Is it possible to write a more or less full feature UI for EMC using only the HAL interface?
[01:25:23] <voxadam> I notice that emcmodule.cc in Axis includes the rcs, emc, emc_nml, kinematics and numerous other headers while a cursory look at Touchy only shows a hal interface.
[01:41:15] <jepler> touchy uses the python emc module too. touchy.py:import emc
[03:58:41] <cradek> touchy uses HAL for its panel controls such as cycle start button. a more typical gui doesn't really need hal at all.
[07:00:24] <alex_joni> voxadam: there is a sample config that uses halui and pyvcp, and it can almost completely control emc
[07:00:30] <alex_joni> you cannot load a file though
[08:19:03] <psha> heh, emc2 configure don't check for g++ :)
[09:57:29] <alex_joni> psha: you sure?
[09:57:56] <psha> yes
[09:58:14] <psha> i've tried to build emc on my wifes atom notebook...
[09:58:50] <psha> 629 AC_PROG_CXX
[09:58:54] <psha> configure.in
[09:59:02] <psha> this only _checks_ for g++
[09:59:13] <psha> and if it's not found - sets CXX to g++
[09:59:39] <psha> http://www.gnu.org/software/autoconf/manual/html_node/C_002b_002b-Compiler.html#index-AC_005fPROG_005fCXX-914
[09:59:49] <psha> Otherwise, if the macro is invoked without an argument, then search for a C++ compiler under the likely names (first g++ and c++ then other names). If none of those checks succeed, then as a last resort set CXX to g++.
[10:02:29] <alex_joni> right, read the page
[10:03:01] <alex_joni> so maybe another AC_PATH_PROG(CXX,$CXX) afterwards?
[10:04:21] <psha> not enought
[10:04:28] <psha> i think we have to fail there
[10:04:31] <psha> and not pass
[10:04:42] <psha> additional argument to CHECK is needed
[10:05:54] <alex_joni> AC_PATH_PROG(CXX, $CXX, "none")
[10:06:04] <alex_joni> if test $CXX = "none"
[10:06:05] <alex_joni> then
[10:06:25] <alex_joni> AC_MSG_ERROR([C++ compiler not found])
[10:06:27] <alex_joni> fi
[10:06:39] <psha> i've just added if which $CXX >/dev/null; then....
[10:06:50] <psha> but AC_PATH_PROG is more correct i think
[10:07:11] <alex_joni> which isn't necessarely present :)
[10:07:24] <psha> ah
[10:07:28] <alex_joni> I think..
[10:07:31] <psha> i've found correct way
[10:08:44] <psha> i think :)
[10:09:41] <alex_joni> I looked at how it's done for python
[10:09:50] <alex_joni> couple lines down
[10:10:25] <psha> i see
[10:10:50] <psha> i've tried to pass [AC_MSG_ERROR(...)] as 'not-found' argument
[10:11:39] <psha> but failed
[10:12:41] <psha> i think it's better to copy one from python :)
[10:29:02] <alex_joni> psha: easy to read if it follows other examples in there :)
[10:29:19] <psha> :)
[10:29:23] <alex_joni> but it's been a while since I worked in there ...
[10:31:37] <psha> there is also missing check for xinerama heaers (needed for emcsh)
[11:36:06] <SWPadnos> ?
[11:36:12] <SWPadnos> why would emcsh need xinerama?
[13:15:41] <jepler> SWPadnos: so that the splash screen shows in the right location on multimonitor systems
[13:24:26] <jepler> psha: someone will be happy to look at configure improvements
[13:54:15] <SWPadnos_> SWPadnos_ is now known as SWPadnos
[13:54:25] <psha> jepler: while i'm in configure... there is warning 'run in place is obsolete, use --prefix'
[13:54:44] <psha> but without RIP emc-environment refuses to load...
[13:56:02] <psha> i'll test today emc-env without RIP flag
[14:03:57] <psha> jepler: what's policy to do if xinerama not present?
[14:04:07] <psha> disable emcsh or abort configure?
[14:08:04] <psha> abort since it's needed for emc.so
[14:16:35] <psha> jepler: http://psha.org.ru/cgit/psha/emc2.git/commit/?h=configure-fix
[14:16:37] <psha> configure-fix
[14:16:59] <psha> oops
[14:17:09] <psha> invalid place for xinerama checks
[14:45:26] <jepler> psha: I'd start by making it fatal
[14:45:58] <jepler> if this is the first we've noticed the problem, it must not affect a lot of developers
[14:53:33] <jepler> psha: did you delete the configure-fix branch?
[15:25:56] <jepler> meanwhile, I think I've successfully made all non-current documentation "noindex,nofollow" and added the red banners I had added to the very very old "handbook" pages
[15:26:01] <jepler> let me know if you see any problems
[15:31:55] <cradek> yay, thanks
[15:37:30] <psha> jepler: reuploaded correct one
[15:37:45] <psha> i've first placed xinerama check into 'with x' one
[15:42:57] <jepler> thank you, looks good
[15:44:10] <CIA-41> EMC: 03jepler 07master * r8c27f92bf6f5 10/src/configure.in: configure: Fail if g++ or Xinerama not found
[16:21:24] <SWPadnos> hmmm. now why would dpkg and apt tell me different things about available packages?
[16:23:10] <psha> what's your question? :)
[16:23:40] <SWPadnos> well, I just set up a new 10.04 server machine, and dpkg -l doesn't show me some packages that aptitude (and apt-get) do
[16:23:43] <SWPadnos> such as git
[16:23:56] <psha> dpkg -q or dpkg -l | grep?
[16:24:01] <SWPadnos> I'm wondering why they wouldn't show the same thing
[16:24:08] <SWPadnos> dpkg -l | grep
[16:24:34] <psha> and how you ask apt?
[16:24:52] <SWPadnos> actually, command-line completion will complete package names
[16:25:01] <SWPadnos> and aptitude search also shows them
[16:25:01] <psha> avaiable
[16:25:04] <psha> not installed
[16:25:53] <psha> look apt-cache policy 'package name'
[16:26:15] <psha> it will show you available version along with installed status
[16:27:51] <SWPadnos> well, I guess I have my answer. dpkg/dselect and apt keep separate lists of available packages
[16:27:57] <SWPadnos> which seems silly to me
[16:28:15] <psha> no
[16:28:20] <psha> dpkg is wokring with installed list
[16:28:24] <psha> apt with available
[16:28:40] <SWPadnos> that's according to the dpkg manpage, actually
[16:29:43] <SWPadnos> (in the description of the "--update-avail" command)
[16:30:50] <psha> i think it's for historical reasons
[16:31:00] <psha> when there was only dpkg/dselect
[16:31:07] <SWPadnos> sure, could be
[16:31:16] <SWPadnos> a little confusing though :)
[16:34:15] <psha> it seem that when i've jumped on debian dselect was already obsolete
[16:34:22] <psha> somewhat in potato time
[16:34:38] <SWPadnos> potato time?
[16:34:45] <psha> release before woody
[16:34:58] <SWPadnos> ah. debian potato :)
[16:35:01] <psha> :)
[18:14:22] <CIA-41> EMC: 03jepler 07master * rdc12848308b5 10/src/hal/components/estop_latch.comp: estop_latch: make watchdog behavior match documentation
[18:14:24] <CIA-41> EMC: 03jepler 07master * r471643e574f5 10/src/emc/usr_intf/pncconf/pncconf.py: fix firmware directory names for 7i43 and 5i22
[18:14:25] <CIA-41> EMC: 03jepler 07master * re0fbe2669c1b 10/ (VERSION debian/changelog): release 2.4.6
[18:14:25] <CIA-41> EMC: 03jepler 07master * r4bbdc38db7e2 10/docs/src/hal/comp.lyx: add info on the period variable
[18:14:28] <CIA-41> EMC: 03jepler 07master * r8750a3b827c5 10/docs/src/gcode/main.lyx: add info on O parameters
[18:14:29] <CIA-41> EMC: 03jepler 07master * rf6f7637be741 10/ (5 files in 5 dirs): Merge remote branch 'origin/v2.4_branch'
[18:57:45] <CIA-41> EMC: 03jepler 07v2.4_branch * racc3e9080ffa 10/docs/src/index.tmpl: fix top-level ULs being too far to the left
[19:03:33] <jepler> I'd long thought it was something wrong with linuxcnc.org's main stylesheet, since the /docs/ version looked OK to me and the /docview/ version didn't
[19:03:39] <jepler> .. turns out I was wrong, and just missing a <UL>!
[23:48:35] <dgarr> for consideration (git master): http://www.panix.com/~dgarrett/stuff/0001-interp-add-EXISTS-unary-function.patch