IRC log for #brlcad on 20120518

00:19.22 CIA-65 BRL-CAD: 03starseeker * r50600 10/brlcad/trunk/src/tclscripts/rtwizard/lib/ (PictureTypeB.itcl PictureTypeC.itcl PictureTypeD.itcl): Make picture types B, C and D use rtimage
00:29.18 CIA-65 BRL-CAD: 03starseeker * r50601 10/brlcad/trunk/src/tclscripts/rtwizard/lib/PictureTypeE.itcl: Picture type E converted
00:38.35 CIA-65 BRL-CAD: 03starseeker * r50602 10/brlcad/trunk/src/tclscripts/rtwizard/lib/PictureTypeD.itcl: Fix background color for type D picture
00:40.28 CIA-65 BRL-CAD: 03starseeker * r50603 10/brlcad/trunk/src/tclscripts/rtwizard/lib/PictureTypeF.itcl: And last but not least, type F images
00:45.46 CIA-65 BRL-CAD: 03starseeker * r50604 10/brlcad/trunk/src/tclscripts/rtwizard/lib/ (PictureTypeBase.itcl tclIndex): Remove some unused commands
00:54.01 CIA-65 BRL-CAD: 03starseeker * r50605 10/brlcad/trunk/src/tclscripts/rtwizard/rtwizard.tcl: Add option for perspective, although it doesn't seem to interact well with zoom for some reason
00:56.12 CIA-65 BRL-CAD: 03starseeker * r50606 10/brlcad/trunk/src/tclscripts/rtwizard/rtwizard.tcl: Ah - just needed to crank it into the initial view setup
00:57.01 starseeker Still needs a lot more testing and documentation work, but now the NEWS item about PNG should properly apply
02:23.14 *** join/#brlcad louipc (~louipc@archlinux/fellow/louipc)
02:39.10 CIA-65 BRL-CAD: 03starseeker * r50607 10/brlcad/trunk/NEWS: rtwizard now properly supports PNG output from all of its image types. Bob's new tcl proc cadwidgets::RtImage is the heart of the new features.
03:01.37 CIA-65 BRL-CAD: 03starseeker * r50608 10/brlcad/trunk/src/tclscripts/rtwizard/rtwizard.tcl: remove debugging puts statement
03:30.49 CIA-65 BRL-CAD: 03brlcad * r50609 10/brlcad/trunk/NEWS:
03:30.49 CIA-65 BRL-CAD: tom added an italian translation of the intro to brl-cad. do not know who the
03:30.49 CIA-65 BRL-CAD: original author is. my records indicate it dates back to at least sometime
03:30.49 CIA-65 BRL-CAD: before july 2007 but is an unattributed contribution. files had not identifying
03:30.49 CIA-65 BRL-CAD: markers and no other record in my logs all the way back to 2004.
05:24.39 CIA-65 BRL-CAD: 03phoenixyjll * r50610 10/brlcad/trunk/src/librt/primitives/hyp/hyp_brep.cpp: Fixed the direction of shearing the hyp.
05:30.05 CIA-65 BRL-CAD: 03Phoenix 07http://brlcad.org * r3677 10/wiki/User:Phoenix/GSoc2012/Reports: /* Community Bonding */
05:49.04 *** join/#brlcad Stattrav_ (~Stattrav@117.192.128.176)
06:14.43 CIA-65 BRL-CAD: 03Stattrav 07http://brlcad.org * r3678 10/wiki/User:Stattrav/GSoC2012_log: Updation of logs
06:26.35 *** join/#brlcad archivist (~archivist@host81-149-189-98.in-addr.btopenworld.com)
06:42.51 *** join/#brlcad Stattrav_ (~Stattrav@117.192.128.176)
06:56.01 *** join/#brlcad d_rossberg (~rossberg@BZ.BZFLAG.BZ)
07:38.05 *** join/#brlcad KimK (~Kim__@2001:470:1f0f:1042:4261:86ff:fe43:bcad)
07:57.44 *** join/#brlcad stas (~stas@82.208.133.12)
08:21.43 *** join/#brlcad KimK (~Kim__@209.248.147.2.nw.nuvox.net)
08:36.07 *** join/#brlcad Al_Da_Best (~Al_Da_Bes@elvyn-248-109.halls.student.lut.ac.uk)
10:23.27 *** join/#brlcad cristina_ (~cristina@188.24.66.147)
10:46.26 *** join/#brlcad kane_ (~Mesut@e181169167.adsl.alicedsl.de)
11:18.24 *** join/#brlcad Stattrav_ (~Stattrav@223.235.172.114)
11:22.16 Stattrav_ the benchmark for sphflake fails after the recent updates
11:23.39 *** join/#brlcad Stattrav_ (~Stattrav@223.235.172.114)
11:47.32 *** join/#brlcad Stattrav_ (~Stattrav@223.235.172.114)
12:23.25 brlcad Stattrav: hm, good to know -- did you isolate the commit that causes the failure?
12:23.37 brlcad and did you get the two images starseeker pasted earlier?
13:19.31 *** join/#brlcad crdueck__ (~AndChat43@184.151.61.79)
13:21.10 brlcad morning crdueck__
13:21.13 Al_Da_Best kane_: Did you say you managed to compile brl-cad with bullet? I'm running into a lot of errors when trying to do so, could do with some advice
13:21.56 brlcad the vm image has a bullet-enabled compile included, could inspect it for reference
13:22.40 Al_Da_Best Yeah I have done. As far as I can tell my CMake is setup correctly, going by the cmakecache on the VM, but visual studio is having issues
13:22.46 crdueck__ brlcad: morning, have you gotten a chance to take a look at the patch I submitted on SF?
13:23.33 *** join/#brlcad AndChat-430625 (~AndChat43@184.151.61.29)
13:27.46 *** join/#brlcad crdueck__ (~AndChat43@184.151.61.29)
13:55.59 CIA-65 BRL-CAD: 03bob1961 * r50611 10/brlcad/trunk/src/libtclcad/tclcad_obj.c: Simplified to_png(). Added to_pix(). Not tested on Windows.
13:57.17 CIA-65 BRL-CAD: 03bob1961 * r50612 10/brlcad/trunk/src/tclscripts/lib/Ged.tcl: Added cadwidgets::Ged::pix.
14:10.12 CIA-65 BRL-CAD: 03bob1961 * r50613 10/brlcad/trunk/src/libtclcad/tclcad_obj.c: More simplification of to_png() and to_pix().
14:25.05 starseeker Al_Da_Best: so far as I know, the bullet stuff has never been tried on Windows before
14:25.13 starseeker not surprising there would be issue
14:25.24 starseeker or issues even
14:26.53 Al_Da_Best Yeah that's what I thought. I'll check that its not my configuration before diving in and trying to fix things that aren't broken, should be moving to a fresh install of windows+vs2010 when I get round to it, which would be ideal for testing
14:28.38 kane_ Al_Da_Best: I have tried it with the Bulletlibraries, i got 61 build-errors
14:29.04 Al_Da_Best Ah :P Yeah I get similar amounts. Oki, I'll try and work out why :D
14:38.37 *** join/#brlcad Al_DC_Mobile (~Al_Da_Bes@2001:630:301:9110:f0fa:9bcd:9fc8:1e4e)
15:10.51 *** join/#brlcad Al_Da_Mobile (~Al_Da_Bes@2001:630:301:9110:b553:a4e8:6a5f:ce27)
15:19.33 *** join/#brlcad ibot (~ibot@rikers.org)
15:19.33 *** topic/#brlcad is BRL-CAD Open Source Solid Modeling || http://brlcad.org || logs: http://ibot.rikers.org/%23brlcad/ || Congratulations to our 11 accepted GSoC students!!
15:22.52 *** join/#brlcad crdueck__ (~AndChat43@184.151.61.29)
15:34.10 brlcad crdueck__: not yet
15:35.21 brlcad I usually try to let other devs get first dibs, but I'll take a look if nobody else comments soon
15:40.31 crdueck__ brlcad: alright thanks
15:42.09 Stattrav brlcad: yes
15:42.25 brlcad crdueck__: so one issue I notice right off the bat is that you've had 9 patch submissions
15:42.54 brlcad 5 of which are "invalid" or deleted or rejected
15:43.13 *** join/#brlcad Stattrav_ (~Stattrav@117.192.143.52)
15:43.18 Stattrav brlcad: doing a binary search on the commits between them I checked r 50443
15:43.23 Stattrav that was working fine
15:44.09 brlcad crdueck__: that's usually a sign that you're not thinking through or testing your own patch before submitting it, so a little more rigor on your end is probably warranted
15:44.52 brlcad don't want to be in a situation where 5/9ths of your commits are problematic .. that'd be quite a bit of havoc :)
15:45.38 brlcad Stattrav: do you know how to binary search efficiently or are you doing full-blown checkouts each time? :)
15:46.02 Stattrav brlcad: svn update -r <r.no>
15:46.19 brlcad okay, good just checking
15:46.21 Stattrav brlcad: svn update -r <revison no> <- this was what I was doing. so yeah
15:55.22 crdueck__ brlcad: agreed, i'll try to be more thourough in the future and especially when committing
15:57.53 brlcad crdueck__: that said, I'm hoping you have reviewed your four patches that are open to make sure they don't have obvious room for improvement, like wrong style or not compiling, etc
15:59.31 brlcad if a patch has an issue, you do not need to close out the patch tracker, just add a new attachment and mark it v2 or similar
16:05.59 crdueck__ i dont believe there's any obvious issues with the open patches, but i'll look them over again when i get the chance. i'm away from my desktop at the moment
16:07.49 CIA-65 BRL-CAD: 03bob1961 * r50614 10/brlcad/trunk/src/libtclcad/tclcad_obj.c: Added calls to glPixelStorei(GL_PACK_ALIGNMENT, 1) in both to_png() and to_pix().
17:06.37 *** join/#brlcad Al_Da_Mobile (~Al_Da_Bes@2001:630:301:9110:69b9:8754:c10d:fc6f)
18:08.24 *** join/#brlcad Al_Da_Mobile (~Al_Da_Bes@2001:630:301:9110:69b9:8754:c10d:fc6f)
19:09.07 *** join/#brlcad ksuzee (~ksu@46.149.81.166)
19:39.44 CIA-65 BRL-CAD: 03Ksuzee 07http://brlcad.org * r3679 10/wiki/User:Ksuzee/Reports:
19:40.22 CIA-65 BRL-CAD: 03Ksuzee 07http://brlcad.org * r3680 10/wiki/User:Ksuzee/Reports:
19:41.06 CIA-65 BRL-CAD: 03Ksuzee 07http://brlcad.org * r3681 10/wiki/User:Ksuzee/Reports:
19:41.46 CIA-65 BRL-CAD: 03Ksuzee 07http://brlcad.org * r3682 10/wiki/User:Ksuzee/Reports:
19:46.29 brlcad ksuzee: so curious, what criteria are you using for selecing these reductions?
19:51.54 CIA-65 BRL-CAD: 03Sean 07http://brlcad.org * r3683 10/wiki/Google_Summer_of_Code/2012: move additional info to the end, prep for announcement
19:56.16 CIA-65 BRL-CAD: 03Sean 07http://brlcad.org * r3684 10/wiki/Google_Summer_of_Code/2012: deengrishify
20:03.31 *** join/#brlcad DarkCalf (~DarkCalf@173.231.40.99)
20:13.49 CIA-65 BRL-CAD: 03brlcad * r50615 10/brlcad/trunk/src/librt/primitives/ell/ell.c: apply sf patch 3515075 (rt_ell_surface_area) from crdueck adding support for calculating ellipsoid surface areas.
20:39.16 Stattrav brlcad: 50580 works fine and 50581 breaks
20:45.34 brlcad Stattrav: that can't be right
20:46.03 brlcad 50580 and 50581 are both edits to our release notes text file, having nothing to do with runtime behavior
20:46.34 Stattrav wait or 582
20:46.43 brlcad well which is it? :)
20:46.50 Stattrav let me check that once :P
20:47.28 Stattrav did somebody else report that the benchmark test is reporting that sphflake is throwing out the error ?
20:47.37 brlcad actually, I'll buy 50582 breaking sphflake -- let me read into it
20:51.20 brlcad Stattrav: you don't have a patch yet, do you? want to try and tackle that one now that you have it isolated?
20:51.39 brlcad I can certainly explain it to help make it easy if that commit really did break sphflake
20:52.49 brlcad actually I think I already see the bug
20:53.32 Stattrav brlcad: let me confirm once
20:53.43 Stattrav as in let me check that once again.
20:53.56 brlcad sure, but I'm pretty sure I see why it's failing
20:54.49 Stattrav I am checking 50580
20:54.58 brlcad yeah, that's the issue -- no worries, not even patch-worthy because the edit was just wrong
20:56.02 Stattrav 580 is good
20:56.34 Stattrav 582 had a change in parse.c
20:58.08 CIA-65 BRL-CAD: 03brlcad * r50616 10/brlcad/trunk/src/libbu/parse.c: Revert r50582 as it was just wrong. 'shader' is the whole string, not just the name.
20:58.49 Stattrav cool, I shall test 50616 now
21:01.31 CIA-65 BRL-CAD: 03brlcad * r50617 10/brlcad/trunk/src/libbu/parse.c: go ahead and still make the comparison case insensitive. thanks to Stattrav (via irc) for pinpointing the cause of sphflake failure introduced by r50582.
21:02.15 Stattrav Yaay! my first useful contribution.
21:02.38 ksuzee hello
21:03.18 ksuzee Sean, I've just read you wrote me)
21:03.25 Stattrav Now, I shall get down to writing the patch.
21:05.03 *** join/#brlcad Stattrav_ (~Stattrav@117.192.143.52)
21:07.18 brlcad Stattrav_: so what patch are you working on?
21:08.54 Stattrav the same thing pixcmp.c, couldn't find the bug w.r.t starseeker 's image with normal testing, seeing into it now. Instead I want to write the off-by-<val> where val is specified by -d
21:09.34 Stattrav by write the off-by-<val>, I meant implement that feature
21:14.15 starseeker src/libbu/parse.c:1767:2: error: implicit declaration of function 'bu_strnasecmp'
21:14.35 CIA-65 BRL-CAD: 03brlcad * r50618 10/brlcad/trunk/src/tclscripts/mged/skt_ed.tcl:
21:14.36 CIA-65 BRL-CAD: qualify all of the calls to the dist proc as being in the top-level namespace.
21:14.36 CIA-65 BRL-CAD: apparently if you qualify one in a proc, but not others, some version(s) of tcl
21:14.36 CIA-65 BRL-CAD: assume the non-qualified one are for a non top-level scope even if a local one
21:14.36 CIA-65 BRL-CAD: doesn't exist. problem reported by Csanyi Pal via brlcad-users mailing list
21:14.36 CIA-65 BRL-CAD: trying to use sketch editor.
21:15.57 starseeker oh, I see
21:16.15 CIA-65 BRL-CAD: 03starseeker * r50619 10/brlcad/trunk/src/libbu/parse.c: looks like a typo?
21:20.33 brlcad oops, sorry about that -- didn't have a build to compile-test that on yet
21:25.17 brlcad ksuzee: so you read it, do you intend to reply? :)
21:26.32 brlcad Stattrav: if you can't reproduce the bug under normal use, that may imply reader error or platform-specific conditions -- no need to go on a witch hunt for it ;)
21:26.45 ksuzee oh, yes)) I didn't want to interrupt you=)
21:26.51 ksuzee well,
21:27.08 Stattrav_ brlcad: ahaha, sure.
21:27.26 brlcad ksuzee: it's not an interruption
21:27.34 brlcad multithreaded group discussion ;)
21:27.37 Stattrav_ oh god, I am replying from two places, browser and irssi, shall switch to irccloud
21:28.41 ksuzee ok:) I made the file which contains duplications for every folder in /src
21:29.54 ksuzee and now I'm searching the most obvious situations
21:30.20 ksuzee in this files
21:30.58 ksuzee I think this one https://sourceforge.net/tracker/?func=detail&aid=3528023&group_id=105292&atid=640804
21:31.04 ksuzee is important
21:31.28 ksuzee the previouse one is just an example from my proposal
21:32.39 brlcad it's actually an interesting social bias in play that I found interesting with the htonf.c change
21:33.03 brlcad I knew/know about that duplication
21:33.11 brlcad it's goes WAY back 15+ years
21:33.36 ksuzee eee
21:33.38 brlcad left it in there just because it was so old, but technically there's no reason for it
21:33.59 brlcad so it's nice to see your new eyes questioning its existence
21:34.38 ksuzee these functions were absolutly similar
21:35.07 ksuzee but after making...I saw everything you told me before=)
21:37.34 ksuzee about my previous patch I mean
21:38.11 Stattrav brilliant, the benchmark tests on 50619 are good. thanks starseeker and brlcad.
21:42.07 *** join/#brlcad Stattrav_ (~Stattrav@117.192.143.52)
21:48.44 brlcad Stattrav: aha, I found the bug
21:48.55 brlcad it's actually in the pixdiff utility, not pixcmp
21:49.11 brlcad moreover, the summary differences reported by the two don't match each other:
21:49.28 brlcad [sean@gcc1-power7 .cmake]$ bin/pixcmp eto_implicit.pix eto_brep.pix
21:49.29 brlcad pixcmp pixels: 206784 matching, 31160 off by 1, 24200 off by many
21:49.37 brlcad [sean@gcc1-power7 .cmake]$ bin/pixdiff eto_implicit.pix eto_brep.pix |bin/pix-png -o diff.png
21:49.40 brlcad pixdiff bytes: 702101 matching, 84304 off by 1, 27 off by many
21:49.57 Stattrav_ aah :D
21:49.59 brlcad so not a bug-perse because pixdiff does qualify and say bytes
21:50.14 brlcad but they are different, which is cause for concern
21:50.27 brlcad probably only/best solved by fully merging the two tools
21:50.52 Stattrav_ ohh. so what do you suggest I do now, work on it ?
21:51.22 brlcad adding the off-by-# option is a fine simple patch to start with regardless
21:52.25 brlcad dc
21:52.57 Stattrav_ cool. that would be the thing I shall work on.
21:54.11 CIA-65 BRL-CAD: 03Stattrav 07http://brlcad.org * r3685 10/wiki/User:Stattrav/GSoC2012: Pointer changed in the roadmap
21:54.25 ksuzee so what shall I do with this "good old" duplication? =)
21:54.28 Stattrav_ I could actually look into fixing that a bit later though
21:59.23 CIA-65 BRL-CAD: 03Stattrav 07http://brlcad.org * r3686 10/wiki/User:Stattrav/GSoC2012_log: Updation of logs
22:51.22 CIA-65 BRL-CAD: 03Anuragmurty 07http://brlcad.org * r3687 10/wiki/User:Anuragmurty: /* community bonding period */
23:53.48 *** join/#brlcad crdueck__ (~AndChat43@184.151.61.220)

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