irclog2html for #brlcad on 20100127

00:04.50starseekerred even
00:05.27brlcad:)
00:07.02starseekerthat'll be a nice little loc reduction once I get done mucking with the editor stuff...
00:08.35``Erikand probably a minor performance boone, to boot
00:08.57brlcadnah, sprintf just truncs automatically
00:09.07brlcadthat's the diff between bu_vls_printf and bu_vls_sprintf
00:09.28brlcadprintf treats it as a stream and appends, sprintf as a string and truncates
00:16.06starseekerwow, writesolid in tedit.c is begging for a primitive callback.
00:18.26starseekerread too, for that matter
00:21.00``Erikhrm, something similar to the 'get' and 'adjust' callbacks, perhaps?
00:23.27CIA-34BRL-CAD: 03brlcad * r37423 10/brlcad/trunk/src/libged/pathlist.c: rewrite pathlist to not edit the provided pathp so the db_full_path can be const
00:58.28CIA-34BRL-CAD: 03brlcad * r37424 10/brlcad/trunk/bench/run.sh: ah, to forget the almighty 'hinv' command that sgi used to provide. check for it too, but remove the comments in the submission recommendations since we now embed system information in the log files.
01:00.41``Erikhm, 'caprica' on syfy
01:04.51starseekeraffects edcolor too
01:04.53starseekerconfound it
01:06.54starseekeralso edmater and red
01:07.34brlcadahh, yes
01:09.00brlcadheh
01:16.17``Erikwell, ya see, starseeker, the "++" operator means to increment... the same as #'INCR
01:16.30starseekerheh
01:16.38starseekerso THAT's what I was doing wrong
01:17.53``Erik:D
01:29.44starseekerjust... would need a little time to unravel
01:30.50``Erikheh, I wrote that to cheat at an irc based trivia game O:-)
01:31.18starseekertisk, tisk
01:31.33starseekerah!  it compiled.  Now I can watch it blow up
01:43.41``Erik<@GaidinBDJ> Wanna feel old?  Just think, if they remade Back to the Future now, they would be travelling back to 1980.
02:01.11brlcadhaha
02:04.17CIA-34BRL-CAD: 03erikgreenwald * r37425 10/brlcad/trunk/src/libged/pathlist.c: Copy db_init_full_path macro from mged/animedit.c to prevent missing symbol link error.
02:04.22*** join/#brlcad branco_ (n=branco@79.114.27.55)
02:05.15branco_hello ?
02:05.41brlcad~ask
02:05.41ibotQuestions in the channel should be specific, informative, complete, concise, and on-topic.  Don't ask if you can ask a question first.  Don't ask if a person is there; just ask what you intended to ask them.  Better questions more frequently yield better answers.  We are all here voluntarily or against our will.
02:07.17branco_cant compile from source on ubuntu 8.04 - some opennurbs error - u responded once on the ubuntuforums that its a badly setup system somehow , but im relatively sure my system is setup ok
02:07.42branco_/usr/include/c++/4.2/cstddef:55: error: '::ptrdiff_t' has not been declared
02:08.05brlcadthat doesn't sound like the first error
02:08.16brlcadif it's the problem I'm aware of, it's an optimization bug in gcc
02:08.23branco_when compiling opennurbs_3dm_attributes.cpp
02:08.30brlcadto see if it is, cd src/other && make CXXFLAGS=-O2
02:08.38brlcador reconfigure without --enable-optimized
02:08.55branco_no I tried several things
02:09.06brlcadthen I need more of a log than that one line
02:09.12brlcadneed it from the compile line to the end of the error
02:09.16brlcad~bzpaste
02:09.17ibotfrom memory, bzpaste is http://pastebin.bzflag.bz/
02:09.23brlcadideally posted there :)
02:09.44brlcadwhat is your configure line?
02:09.58brlcadand what version of brl-cad are you attempting to compile?
02:10.44branco_http://pastebin.bzflag.bz/d3c3bd87c
02:11.12branco_i guess --enable-optimized --with-ogl
02:11.25branco_brlcad 7.16.4
02:11.37brlcaddid you try make CXXFLAGS=-O2 ?
02:11.41branco_last stable source release
02:12.15brlcadbecause that really looks like the optimization bug .. maybe not, but oddly close
02:12.28brlcadit is a bit shorter than I recall..
02:12.49branco_no , but how could that fix it ? it seems very strange that it would fix such an error
02:13.11branco_http://ubuntuforums.org/showthread.php?t=971927
02:13.23branco_i get the exact same output
02:14.00brlcadyou're in std header land
02:14.12brlcaddifferent code paths are taken for optimized, unoptimized
02:14.23brlcadit was worth a shot regardless
02:14.27brlcadsimple enough to test
02:14.55branco_cxxflags is an option from the configure command line ?
02:15.02brlcadit's failing on #include <new>
02:15.20brlcadhm?
02:15.29branco_did you try make CXXFLAGS=-O2 ?
02:15.35brlcadconfused, you tried it or you didn't tryi it?
02:15.40brlcadthat's not for ME to try.. :)
02:15.40branco_didnt
02:15.43brlcadthat's for you to try
02:16.14branco_ok ill see how i can do that
02:16.28brlcadthere's no "how" to it, other than type it in
02:16.31brlcadexactly like that
02:16.34brlcadinstead of "make"
02:16.42brlcadtype "make CXXFLAGS=-O2"
02:16.52brlcadno more, no less
02:16.59branco_ok thnaks , ill try it
02:19.19*** join/#brlcad jesica__ (n=jesica@168.226.178.47)
02:21.53brlcadbranco_: do you have build-essential installed?
02:22.07brlcadit's really looking mostly like you have a busted install of the standard c++ headers
02:22.56brlcadit's basically just failing on the very first c++ file, nothing to do with the actual line of code in question that it stops on
02:34.30*** join/#brlcad branco__ (n=branco@79.114.98.56)
02:49.45branco__i cant wait forever - mabe I'll log in tomorrow - on some reasonable timing for my part of the world - (branco_ to branco__ - got disconnected)
02:50.26*** part/#brlcad branco__ (n=branco@79.114.98.56)
03:03.01brlcadah, heh yes, that too
03:03.15brlcadthe child closes the standard descriptors for safety
03:23.24CIA-34BRL-CAD: 03brlcad * r37426 10/brlcad/trunk/ (45 files in 19 dirs):
03:23.24CIA-34BRL-CAD: make db_walk_tree()'s struct db_full_path pointer be const since no caller
03:23.24CIA-34BRL-CAD: should really need to be modifying it (and for the few that were temporarily
03:23.24CIA-34BRL-CAD: modifying it, make them get a copy). this cascades a const through to all of
03:23.24CIA-34BRL-CAD: the db_walk_tree() callbacks for region start, region end, and the leaf
03:23.25CIA-34BRL-CAD: function, but it's a minor api change. oof.
03:31.06jesica__brlcad ...
03:31.36brlcadyes?
03:31.55jesica__which differences exist if I add an html instead an xml?
03:32.21CIA-34BRL-CAD: 03brlcad * r37427 10/brlcad/trunk/src/libgcv/region_end.c: move NMG debug state save/restore to outside the jump sections.
03:32.39jesica__ask for the converter to pdf mostly
03:33.16brlcaddepends how the html is generated, but generally tons of changes
03:33.22brlcadhtml isn't really useful, it's not our source format
03:33.34brlcadwe generate html and pdf automatically from docbook
03:34.01brlcadwe cannot maintain html files easily, there are too many formats and too many docs
03:35.52CIA-34BRL-CAD: 03brlcad * r37428 10/brlcad/trunk/src/libgcv/region_end.c: for that matter, we don't need to save/restore twice
03:37.33jesica__well, just to know
03:43.19jesica__brlcad when I try to open the xml with just a click, a comment pops-up and tell It is an xml for the extension, but an html for the contents
03:43.57jesica__but I read ir, and its an xml, or at least, It looks like other xml in /lessons/en
03:44.18brlcadwhat are you editing it with?
03:44.23brlcademacs?
03:44.23brlcadvim?
03:44.41brlcadit's just text
03:44.51jesica__a simple editor with hightlight
03:44.52brlcadso any text editor should work fine
03:45.12jesica__use to edit with vim short changes
03:45.14brlcaddoesn't matter if it thinks it's html or xml, so long as the syntax is what you've been doing
03:45.37jesica__I can open it with the same editor
03:45.39brlcadosea if you write docbook, it's fine
03:45.48jesica__well, any editor can open it
03:45.51brlcade.g., if it looks okay in vim, then it's fine
03:46.07jesica__but first it shows the warning
03:46.35jesica__It looks ok everywhere
03:46.55brlcadthen it should be fine
03:47.29CIA-34BRL-CAD: 03brlcad * r37429 10/brlcad/trunk/src/libgcv/region_end.c: refactor out the goto statements (exactly for starters) via a cleanup function
03:47.46jesica__at least, it works until now :)
03:48.28brlcadwhat you have been doing has been working great
03:49.20brlcadthere's going to be a bigger news item regarding your efforts in the next release
03:49.26brlcadhttp://brlcad.svn.sourceforge.net/viewvc/brlcad/brlcad/trunk/NEWS  <-- aqui
03:50.03brlcadthat will go to our mailing lists, the linux cad lists, sourceforge, freshmeat, and a few other places
03:52.07jesica__brlcad remind there's two images with english words in
03:52.13brlcadyeah
03:52.17CIA-34BRL-CAD: 03brlcad * r37430 10/brlcad/trunk/NEWS: Add missing news entry for jesica's work translation tutorials 3 and 6 to spanish. woo hoo.
03:52.27jesica__:)
04:05.46CIA-34BRL-CAD: 03brlcad * r37431 10/brlcad/trunk/NEWS: (log message trimmed)
04:05.50CIA-34BRL-CAD: this one is a biggie. keith applied a fix to the BoT ray tracing code so that
04:05.56CIA-34BRL-CAD: we no longer have a solidity failure when grazing a surface edge. using a FILO
04:06.00CIA-34BRL-CAD: in/out collapse, multiple ins are collapsed to the first in, and multiple outs
04:06.06CIA-34BRL-CAD: to the last out, making grazing points be considered a full hit. this should
04:06.10CIA-34BRL-CAD: fix a bug documented in sf bug 1501921 (Missed volume when raytracing BOTs)
04:06.16CIA-34BRL-CAD: reported by Charles Kennedy (g2asc) and Tim Myers where a BoT with stair-step
04:14.26jesica__brlcad  It's me driving crazy CIA-34 ? :P
04:15.37jesica__or sourceforge drives me crazy?
04:17.05jesica__whe I entry commit it asks me the password for 'login' gnome keyring
04:17.40jesica__... I'm afraid can't do it right
04:17.48CIA-34BRL-CAD: 03brlcad * r37432 10/brlcad/trunk/src/libgcv/region_end.c: make empty_model and empty_region non-static, simplify the logic and cleanup parameters.
04:17.50brlcadwhat's asking for a password?
04:17.57brlcadhow are you trying to commit?
04:18.01brlcadon the command line or with a gui?
04:19.03brlcadthe very first time you commit, it will ask you for your login -- give it your sourceforge username and password, NOT your system username
04:19.54brlcadso for you, your login: is "jesicaguidice"
04:20.10jesica__:) heh, GNOME word confused me
04:20.14brlcadpassword should be your sourceforge password
04:20.59jesica__mmm but it ask me first for the password
04:21.19jesica__I'm shure of that because it don't show my entries
04:21.23brlcadright, it will first try your system username .. so just hit enter
04:21.35brlcadthen it will ask login:
04:21.56brlcadfortunately, you only have to do this ONCE :)
04:22.12brlcadsolo una vez, despues se lo recuerda
04:23.24jesica__mmm
04:25.45CIA-34BRL-CAD: 03brlcad * r37433 10/brlcad/trunk/src/libgcv/region_end.c: make the jump section an else block to be obvious and consistent with other code. make the nmgregion be non-static too (WIP).
04:27.21jesica__but the probles is just with thw comment or the adds werenot applyed?
04:28.08jesica__brlcad I'm too tired, you know
04:28.32jesica__prefer to try it tomorrow earlier
04:29.04jesica__bad idea to do this with closed eyes :P
04:29.18brlcadthe adds were applied
04:29.24brlcadrun "svn status"
04:29.31brlcadshould see lines that start with "A"
04:29.38jesica__yes I saw the status
04:29.40brlcadyou're almost there
04:29.47jesica__they are
04:30.02jesica__good
04:30.06brlcadso are you running "svn commit"?
04:30.12jesica__aha
04:30.30brlcadand what does it say?
04:30.55jesica__could not authenticate to server
04:31.05brlcadcan you log in on sourceforge?
04:31.12jesica__yes
04:31.24brlcadshow me what it shows you
04:32.04brlcadscreenshot or copy/paste text from console will work
04:32.19jesica__<PROTECTED>
04:32.20jesica__Password for 'login' GNOME keyring:
04:32.20jesica__svn: Fall? el commit (detalles a continuaci?n):
04:32.20jesica__svn: MKACTIVITY de '/svnroot/brlcad/!svn/act/f16532de-2a33-44e6-b2f6-0430491a8a3b': fall? la autorizaci?n: Could not authenticate to server: rejected Basic challenge (https://brlcad.svn.sourceforge.net)
04:33.27brlcadoh, you're using gnome keyring...
04:33.36brlcadwonder if you put the wrong key in the keyring..
04:33.53jesica__well, but I've tried with its password
04:33.57brlcadwhat password did you put there?
04:34.07brlcadgnome keyring should be a diff password
04:34.11jesica__gnome keyring
04:34.19jesica__ah?
04:34.27brlcadso maybe you have a bad sourceforge password in your keyring
04:35.58brlcadaha
04:36.08brlcadgrep password-stores ~/.subversion/config
04:36.12brlcadwhat does it report?
04:36.20jesica__mmm possible, remember I'd tryed a lot of times with sourceforge
04:36.58jesica__absolutly nothing
04:37.01brlcadokay
04:37.06``Erikum, didn't sf disable passwords for ssh access a long time ago? you need keys now?
04:37.15brlcadgrep password-stores /etc/subversion/config
04:37.31brlcad``Erik: nah, still allows passwords
04:37.43brlcadused mine just earlier today on a new system
04:38.01``Erikah, 'k, for svn, cvs required ssh which needed a key, I think
04:38.07jesica__<PROTECTED>
04:38.12brlcadthat's the problem
04:38.16brlcadjesica__: add these two lines to ~/.subversion/config
04:38.18brlcad[auth]
04:38.23brlcadpassword-stores =
04:38.46brlcador edit /etc/subversion/config and make that password-stores = nada
04:38.53jesica__just that?
04:39.47brlcadyes
04:39.58jesica__done
04:40.06brlcadalternatively, you could delete your login gnome keyring :)
04:42.45CIA-34BRL-CAD: 03brlcad * r37434 10/brlcad/trunk/src/libgcv/region_end.c: simplify cleanup by pushing the nmg_kr(r) up a level. following the existing logic, looks like it's only actually killing the region in a couple places.
04:42.48brlcadyou should now be able to svn commit, it'll say password, hit enter, it'll say login, enter your sourceforge login
04:42.52brlcadetc
04:43.02jesica__done :D
04:44.01jesica__buuu
04:44.02jesica__Commit blocked by pre-commit hook (exit code 1) with output:
04:44.03jesica__/var/local/mastertree/host/sfp-svn/hook-scripts/check-mime-type.pl:
04:45.22jesica__brlcad Is really difficul to me to stop in the middle of something, but I'm really tired
04:45.32brlcadahhhh
04:45.36jesica__sleeped jus 2 ours the last 2 days
04:45.38brlcadSO close.....
04:45.39brlcad:)
04:45.43brlcadone more command
04:46.06jesica__ok
04:46.26brlcadcurl http://brlcad.org/~sean/subversion.config > ~/.subversion/config
04:46.46brlcadyou'll need to re-add the "password-stores ="
04:46.59brlcadand revert the files
04:47.07brlcadsvn revert -R doc/docbook
04:47.24brlcadsvn add doc/docbook/lessons/es/whatever.xml
04:47.44brlcadthat will set the mime-type automatically after you download my config
04:47.50brlcadmore details here: http://brlcad.org/wiki/Mime-types
04:49.19jesica__can I add 4 xml at the same time, really?
04:49.36brlcadyou should be able to
04:49.47brlcadcd doc/docbook/lessons/es && svn add *.xml
04:51.23jesica__buuuu
04:51.39brlcadque paso'?
04:51.42jesica__I'm falling sleep
04:51.46jesica__nothing
04:51.49brlcadaww
04:52.04brlcadtan cerquita
04:52.06jesica__me qued? dormida con la noteboook encima
04:52.13brlcadjeje
04:52.40brlcad'tas joven
04:52.43jesica__tell me what means the message
04:52.47jesica__:)
04:53.19brlcadyou mean what does the mime-type message mean?
04:53.29jesica__no
04:55.33jesica__brlcad please, tomorrow
04:55.39brlcadmm, okay :)
04:55.45jesica__I wanna understand
04:56.05jesica__enjoy it, but I cannot do it like this :(
04:56.12brlcadno problem, until tomorrow then :)
04:56.32jesica__thanks, night
05:08.34CIA-34BRL-CAD: 03brlcad * r37435 10/brlcad/trunk/src/librt/primitives/nmg/nmg_mk.c: add additional parameter validation to all of the kill routines to allow null parameters. this should make life easier on callers to not have to sanity check everything 'just in case'.
05:22.51CIA-34BRL-CAD: 03brlcad * r37436 10/brlcad/trunk/src/mged/edsol.c: no longer need to check for non-null
05:29.36CIA-34BRL-CAD: 03brlcad * r37437 10/brlcad/trunk/src/libgcv/region_end.c: validate parameters a little more
05:49.34CIA-34BRL-CAD: 03brlcad * r37438 10/brlcad/trunk/src/ (libged/pathlist.c mged/animedit.c): the db_init_full_path macro is unnecessary. it was a typo (one which animedit.c apparently made too). the call is supposed to be db_full_path_init().
06:22.06CIA-34BRL-CAD: 03brlcad * r37439 10/brlcad/trunk/src/librt/db_tree.c: allow db_free_tree() to be given a null and be okay with it.
06:28.55CIA-34BRL-CAD: 03brlcad * r37440 10/brlcad/trunk/src/libgcv/region_end.c:
06:28.55CIA-34BRL-CAD: finally quiet the warning about curtree possibly getting clobbered due to a
06:28.55CIA-34BRL-CAD: longjmp. We make a copy of the curtree and free the original so that we can
06:28.55CIA-34BRL-CAD: play with our copy more carefreefully. this lets strict pass on erik's
06:28.56CIA-34BRL-CAD: teardropmaxbox that was curiously unreproducible on an identical configuration
06:28.58CIA-34BRL-CAD: elsewhere. no matter, it seemeth fixed now.
08:06.29CIA-34BRL-CAD: 03brlcad * r37441 10/brlcad/trunk/src/librt/db_tree.c:
08:06.29CIA-34BRL-CAD: implement support for duplicating an OP_NMG_TESS tree as well so we don't die in
08:06.29CIA-34BRL-CAD: a miserable bomb death. can't figure out why this isn't really working, though,
08:06.29CIA-34BRL-CAD: so just stub the code and cheat by creating a 'fake' copy for now that
08:06.30CIA-34BRL-CAD: still/also references the original region data.
08:08.42CIA-34BRL-CAD: 03brlcad * r37442 10/brlcad/trunk/src/libgcv/region_end.c: helps to test sometimes. can't call db_free_treee() since we don't really have a real copy yet. it's a fake so just leave it be.
08:32.17CIA-34BRL-CAD: 03brlcad * r37443 10/brlcad/trunk/src/librt/primitives/nmg/nmg_mk.c: whoopsie daisy. bad logic there skippo. do the checks if it's not null, not the other way around.
08:40.58brlcadfantastic.. tesselllate havoc once and it bombs.. immediately tess again (with the same binary, same options, exact same model) and it succeeds
08:41.51brlcadi'd swear the overall output looks a heck of a lot cleaner than it did the last time I ran it though, pretty clean
08:42.01brlcadslow but clean
08:46.27brlcadand complete! wow.. don't recall the last time havoc went through cleanly with no obvious missing parts
08:49.53CIA-34BRL-CAD: 03brlcad * r37444 10/brlcad/trunk/BUGS: mged rrt command isn't respecting the view size, just using default autoview (from the right az/el).
08:56.29CIA-34BRL-CAD: 03brlcad * r37445 10/brlcad/trunk/BUGS: rrt even gets azel wrong
09:05.43CIA-34BRL-CAD: 03brlcad * r37446 10/brlcad/trunk/src/adrt/load_g.c: another const pathp callback that snuck through a crack.
09:55.11brlcad``Erik: distcheck looks clean now, libgcv works strict
09:56.59yukonbobmorning #brlcad
09:57.15brlcadhowdy yukonbob
09:57.34yukonbob!Hi there never-sleeps :)
09:58.08yukonbobbrlcad: how's it going?
09:58.22brlcadpretty good
10:23.30CIA-34BRL-CAD: 03brlcad * r37447 10/brlcad/trunk/configure.ac:
10:23.31CIA-34BRL-CAD: re-enable debugging of mac os x dylibs under 10.5 where they apparently swapped
10:23.33CIA-34BRL-CAD: over to DWARF debug symbols, so ask the compiler to generate STABS for us
10:23.35CIA-34BRL-CAD: instead. while we're at it, lets crank debugging up to level 3 so that macro
10:23.39CIA-34BRL-CAD: definitions are also included.
11:19.15*** join/#brlcad mafm (n=mafm@121.Red-81-32-105.dynamicIP.rima-tde.net)
11:52.23d-loMornin brlcad!
12:10.01brlcadhowdy
12:16.36d-loup early or up late? :)
12:27.24*** join/#brlcad parigaudi (n=quassel@pd95b7f5e.dip0.t-ipconnect.de)
13:38.58*** join/#brlcad alex_joni (n=alex_jon@emc/board-of-directors/alexjoni) [NETSPLIT VICTIM]
13:39.00*** join/#brlcad Computer (n=Computer@209.16.114.100) [NETSPLIT VICTIM]
13:58.48CIA-34BRL-CAD: 03davidloman * r37448 10/rt^3/trunk/ (4 files in 2 dirs): D-lo: Fixed some includes issues.
15:14.06*** join/#brlcad Phurl (n=mdupont@ip-81-210-245-60.unitymediagroup.de)
15:39.57``Erikbrlcad: http://pastebin.bzflag.bz/d63989f08
19:14.41CIA-34BRL-CAD: 03starseeker * r37449 10/brlcad/trunk/ (12 files in 3 dirs):
19:14.43CIA-34BRL-CAD: Needs more testing and a review to make sure the approach is sensible, but move
19:14.45CIA-34BRL-CAD: the logic that tells libged how to call an editor to the MGED level - pass a
19:14.47CIA-34BRL-CAD: string to the libged editit command. Can successfully run TextEdit and vim
19:14.51CIA-34BRL-CAD: (without wrapper script) on Mac.
19:15.58CIA-34BRL-CAD: 03indianlarry * r37450 10/brlcad/trunk/include/opennurbs_ext.h:
19:16.02CIA-34BRL-CAD: Fix to remove vertical trims from "above list" that are not close to the hit
19:16.04CIA-34BRL-CAD: points and should not be used for determining trimming. Also adjustments to
19:16.11CIA-34BRL-CAD: BREP_EDGE_MISS_TOLERANCE and BREP_SAME_POINT_TOLERANCE definition values.
19:18.05starseekerah, here's the tkterm code: http://expect.cvs.sourceforge.net/viewvc/*checkout*/expect/expect/example/tkterm?revision=5.32
21:57.02``ErikiPad.. .with wings... O.o
22:11.51``Erikhah, already... http://failblog.files.wordpress.com/2010/01/129091000493619623.jpg
22:25.00yukonbob``Erik: there's also the "iPod is like a min-iPad -- where the device launched today is the max-iPad"...
22:32.53brlcad(erik)
23:43.21*** join/#brlcad Nohla (n=jesica@168.226.179.147)

Generated by irclog2html.pl by Jeff Waugh - find it at freshmeat.net! Modified by Tim Riker to work with infobot logs, split per channel, etc.