IRC log for #brlcad on 20100805

00:09.37 CIA-43 BRL-CAD: 03starseeker * r40008 10/brlcad/branches/cmake/CMakeLists.txt:
00:09.37 CIA-43 BRL-CAD: Probably not in a working state (no cmake handy to test) but start fleshing out
00:09.37 CIA-43 BRL-CAD: the contents of this file - gonna need to express what's in configure.ac in
00:09.37 CIA-43 BRL-CAD: here, although for most options the 'auto' setting in configure.ac will just
00:09.37 CIA-43 BRL-CAD: translate to some up-front testing...
01:08.31 ``Erik brlcad: http://www.ogre3d.org/2009/12/31/ogre-1-7-0-rc1-cthugha-released is when iphone support was added to ogre
01:08.44 ``Erik wonders if the license switch to MIT is related O.o
03:05.58 starseeker ``Erik: is LGPL a no-go on the iphone?
03:14.27 ``Erik I d'no, there seems to be confusion about it all *shrug*
03:16.39 CIA-43 BRL-CAD: 03starseeker * r40009 10/brlcad/branches/cmake/CMakeLists.txt: Fix a few gotchas - CMake is now doing the pretty-print banner a.l.a configure - figured out how to generate the spacers.
03:16.42 starseeker woot!
03:16.58 starseeker was wrong - CMake is able to do string lengths
04:09.41 CIA-43 BRL-CAD: 03starseeker * r40010 10/brlcad/branches/cmake/ (3 files in 2 dirs):
04:09.41 CIA-43 BRL-CAD: Add in a minor variation on Wireshark's CMake Lex/Yacc find routines, which
04:09.41 CIA-43 BRL-CAD: conveniently were dual licensed GPL/BSD. Probably need to think about Windows
04:09.41 CIA-43 BRL-CAD: in these Find* files and may need considerably more sophistication in these
04:09.41 CIA-43 BRL-CAD: macros, but it's a start.
04:25.03 *** join/#brlcad packrat (~packrator@99-67-225-40.lightspeed.livnmi.sbcglobal.net)
04:27.22 CIA-43 BRL-CAD: 03starseeker * r40011 10/brlcad/branches/cmake/CMakeLists.txt: Use flex/bison find_package by default - will need to set up the other macros as non-flex/bison fallbacks.
04:49.58 brlcad willdye: no problem, welcome to lurk, welcome to code, welcome to do as you please ;)
04:50.06 *** join/#brlcad Ralith (~ralith@216.162.199.202)
04:50.13 brlcad ``Erik: thanks
04:50.44 brlcad I vaguely remember seeing that months ago
04:50.52 brlcad after you mentioned it
05:30.43 brlcad shouldn't still be awake!
06:50.21 *** join/#brlcad merzo (~merzo@smartbussiness.mobicom.net.ua)
09:05.00 *** join/#brlcad mafm (~mafm@83.40.126.66)
11:05.13 *** join/#brlcad Zaebos (~irc@pd95b7f5e.dip0.t-ipconnect.de)
11:36.08 CIA-43 BRL-CAD: 03brlcad * r40012 10/brlcad/tags/rel-7-16-10/: Tagging release 7.16.10 (finally! woo hoo!)
11:39.30 CIA-43 BRL-CAD: 03brlcad * r40013 10/brlcad/trunk/ (NEWS README include/conf/MINOR include/conf/PATCH): release is tagged, bumping revisions. next release is a minor, so let the good times roll on in again.
11:55.24 CIA-43 BRL-CAD: 03brlcad * r40014 10/brlcad/trunk/HACKING:
11:55.24 CIA-43 BRL-CAD: so quotes on the mkdir call are particularly necessary, otherwise it creates the
11:55.24 CIA-43 BRL-CAD: wrong directory. while we're at it, make it so that the instructions really can
11:55.24 CIA-43 BRL-CAD: be just copy-pasted without edits (even though SFUSERNAME will probably have to
11:55.24 CIA-43 BRL-CAD: be set manually).
12:13.09 CIA-43 BRL-CAD: 03brlcad * r40015 10/brlcad/trunk/HACKING: instead of whoami, try to use the auth files directly. look for any sourceforge entry and use that username. echo the results.
12:21.08 brlcad tagged and bagged
12:21.33 brlcad now up on sf.net with release notes, I'll send the announcement later today
12:22.06 *** topic/#brlcad by brlcad -> BRL-CAD Open Source Solid Modeling || http://brlcad.org || http://sf.net/projects/brlcad || #brlcad logs: http://ibot.rikers.org/%23brlcad/ || BRL-CAD release 7.16.10 is posted! (20100805)
12:35.15 *** join/#brlcad typemore (~typemore@unaffiliated/typemore)
12:35.39 typemore is there a public repository of CAD (not necessairly brl-cad) parts? like a wikipedia of mechanical stuff?
13:16.23 CIA-43 BRL-CAD: 03starseeker * r40016 10/brlcad/branches/cmake/CMakeLists.txt: Start looking for libraries
13:41.57 CIA-43 BRL-CAD: 03starseeker * r40017 10/brlcad/branches/cmake/CMakeLists.txt: Add in some more tests - this will need a LOT Of wiring up, cross platform testing, etc...
14:17.56 brlcad that's twice now, he keeps leaving too quickly
14:18.30 brlcad so this should be a full binary release...
14:23.33 brlcad I think we'll need to post 1) Mac OS X (big endian, 32-bit), 2) Mac OS X (little endian, 64-bit), 3) Linux (little endian, 32-bit), 4) Linux (little endian, 64-bit), 5) BSD pkg (little endian, 32-bit), 6) Windows (little endian, 32-bit), 7) Windows (little endian, 64-bit)
14:24.07 brlcad along with the usual package distro updates for ports, portage, fink, and apt
15:21.52 *** join/#brlcad e-motion (~circuit@88.118.29.54)
15:21.54 e-motion hi
15:25.04 *** join/#brlcad Stattrav (~Stattrav@117.192.155.167)
16:05.15 yukonbob brlcad: is big releaes?
16:05.18 yukonbob *release
16:49.52 starseeker blinks - how do I identify the big endian vs. little endian nature of a platform?
16:55.07 yukonbob C bit twiddling?
16:56.31 starseeker there's no posix utlity to report it?
16:58.03 yukonbob auto* might have something -- I'm guessing, though.
16:58.43 yukonbob man -k endian on my FreeBSD box yields nothing... which isn't promising for a POSIX spec...
16:59.27 yukonbob I'd bet tests involve building a trivial C prog that exits 0 for little, 1 for big, for example..
17:02.55 ``Erik no, just use a union, pack and compare
17:03.46 ``Erik or use something like if(0x01020304 == htonl(0x01020304)) printf("Big endian!");
17:05.37 ``Erik um, gcc should also have a define for compile time
17:07.22 starseeker notes both autoconf and CMake can tell you...
17:09.45 starseeker brlcad: I can't remember - for the sourceforge binaries, do we use just /usr/brlcad or /usr/brlcad/rel* as the prefix?
17:15.20 ``Erik he says rel*
17:15.56 ``Erik ideally with symlinks
17:16.07 starseeker righto
17:16.19 starseeker starts on the easy one (64 bit linux...)
17:17.52 ``Erik on what, the h9000? rs6000? r12000? G5? :D
17:17.55 ``Erik oh, x86-64
17:18.03 starseeker yeah
17:18.33 starseeker we're gonna need the _OS_CPU_ additions to the tarball names this time
17:21.23 ``Erik the ones I did were always like brlcad-7.10.4_fbsd_ia32.tbz
17:21.37 starseeker nods
17:27.05 starseeker brlcad: I don't suppose you already have a Windows build? (saw some commits that hinted you were testing there...)
17:28.34 ``Erik (doing 'ignite' talks right now)
17:29.52 starseeker ah
17:32.10 starseeker hmm, cool: http://www.ibm.com/developerworks/aix/library/au-endianc/index.html
17:33.48 ``Erik charcast, ew, union is a hair more readable
17:35.06 ``Erik the htonl/ntohl family, "network order" means big endian
17:36.01 starseeker shudders slightly - where's a good document on the various floating point representation schemes we'll eventually need to handle for dbupgrade?
18:21.22 *** join/#brlcad Stattrav (~Stattrav@117.192.155.167)
18:36.52 ``Erik um, assume IEEE754 until we see otherwise? :D
18:39.15 brlcad yukonbob: not so much in terms of features, but it is hopefully the last in the 7.16 line, major in the sense that it should be pretty stable
18:39.24 brlcad at least considerably more stable than previous releases
18:40.15 ``Erik hm, vax wasn't ieee, damn
18:40.30 brlcad starseeker: I have a build, but didn't have nsis to make the exe
18:40.36 ``Erik http://www.codeproject.com/KB/applications/libnumber.aspx
18:42.57 brlcad gcc isn't even necessarily ieee if you use -O3 or -fast
18:43.49 brlcad oh, nice resource
19:09.57 starseeker eeep: http://www.quadibloc.com/comp/cp0201.htm
19:15.38 starseeker http://www.mrob.com/pub/math/floatformats.html
19:20.59 starseeker hey, cool: http://www.mpfr.org/
19:29.13 starseeker that would probably be a cool library to use, if I understood it :-P
20:07.57 *** join/#brlcad cjdevlin (~devlin@99-74-181-148.lightspeed.cicril.sbcglobal.net)
20:19.18 *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1)
20:39.00 *** join/#brlcad _psilva (~psilva@static-96-255-52-7.washdc.fios.verizon.net)
20:45.29 *** join/#brlcad merzo (~merzo@38-0-132-95.pool.ukrtel.net)
21:19.24 CIA-43 BRL-CAD: 03n_reed * r40018 10/brlcad/trunk/src/tclscripts/boteditor/botEditor.tcl: added exit confirmation dialogs to bot editor
21:54.11 ``Erik hm, kagan was confirmed to the supreme court
22:03.57 *** part/#brlcad e-motion (~circuit@88.118.29.54)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.