IRC log for #brlcad on 20120606

00:36.26CIA-55BRL-CAD: 03crdueck * r50913 10/brlcad/trunk/src/librt/primitives/tgc/tgc.c: added missing break statement
01:25.36crduecki'm not entirely sure how the rt_obj_* interface works, i wrote this interface for volume based off the other obj files in primitives/ and i'd just like someone to check it to see if I've missed anything. http://sprunge.us/cZbb
01:26.13CIA-55BRL-CAD: 03starseeker * r50914 10/brlcad/branches/STABLE/ (189 files in 71 dirs):
01:26.13CIA-55BRL-CAD: Sync trunk and STABLE up to trunk r50908. Reverted previous workaround of vls
01:26.13CIA-55BRL-CAD: issues - should be back to STABLE matching a particular revision of trunk. Will
01:26.13CIA-55BRL-CAD: double check due to complexity of this merge process, but after this syncing
01:26.13CIA-55BRL-CAD: should be more straightforward again.
01:26.40CIA-55BRL-CAD: 03starseeker * r50915 10/brlcad/trunk/doc/docbook/system/man1/en/rtwizard.xml: Get the rtwizard man page closer to complete - still need examples and review.
02:19.33brlcadstarseeker: awesome preemption on the merge
03:29.13brlcadyay, looks like we're in a maintenance window -- cannot commit
03:48.02brlcad``Erik: heh, I don't know what you were smoking with that bot.c change, but cid345 is for a missing break statement (which is indeed missing)
03:49.08brlcadbn_dist_pt2_lseg2() does return [0,5] not [-2,3] as the commit message claimed
03:49.22brlcadso yeah, dunno what went on there, but the fix is simple
04:38.50*** join/#brlcad andrei_ (~andrei@5-12-77-90.residential.rdsnet.ro)
05:16.05*** join/#brlcad andrei_ (~andrei@5-12-77-90.residential.rdsnet.ro)
06:38.15*** join/#brlcad ksuzee (~ksuzee91@193.151.107.42)
06:50.02*** join/#brlcad louipc (~louipc@archlinux/fellow/louipc)
06:51.50*** part/#brlcad ksuzee (~ksuzee91@193.151.107.42)
06:58.23*** join/#brlcad stas (~stas@188.24.41.118)
08:22.49*** join/#brlcad stas (~stas@82.208.133.12)
08:28.21*** join/#brlcad PrezKennedy (~DarkCalf@173.231.40.99)
10:45.43``Erikthe cid was 354, not 345
10:58.18``Erik(now 48539... *look*) hm, that would've been a trivial one to fix, wonder if it was listing differently when I got to it *shrug*
11:00.21CIA-55BRL-CAD: 03erikgreenwald * r50916 10/brlcad/trunk/src/librt/primitives/bot/bot.c: un-comment the 4/5 cases and add in the missing break. Related to Cov354/490354
11:02.40*** join/#brlcad andrei_ (~andrei@5-12-77-90.residential.rdsnet.ro)
12:05.04brlcadright, 354 (now 490354)
12:09.43*** join/#brlcad DarkCalf (~DarkCalf@173.231.40.99)
12:09.46CIA-55BRL-CAD: 03brlcad * r50917 10/brlcad/trunk/src/gtools/g_diff.c:
12:09.46CIA-55BRL-CAD: convert g_diff over from wdb_init_obj/wdb_create_cmd to libtclcad using
12:09.46CIA-55BRL-CAD: to_open_tcl. this removes one of the last remaining vestiges on wdb_obj
12:09.46CIA-55BRL-CAD: (there's still a wdb_get_tcl call that needs to be removed). convert interp
12:09.46CIA-55BRL-CAD: global to upper while we're at it since we added an interp param.
12:09.52CIA-55BRL-CAD: 03brlcad * r50918 10/brlcad/trunk/src/gtools/CMakeLists.txt: g_diff doesn't use libged
12:10.59CIA-55BRL-CAD: 03brlcad * r50919 10/brlcad/trunk/src/gtools/Makefile.am: g_diff doesn't use libged, ged comes before rt
12:12.32CIA-55BRL-CAD: 03brlcad * r50920 10/brlcad/trunk/src/gtools/g_diff.c: add fixme to remove wdb_get_tcl
12:13.11CIA-55BRL-CAD: 03brlcad * r50921 10/brlcad/trunk/TODO: mged is the last remaining caller to wdb_init_obj/wdb_create_cmd needing to be converted (though it should probably convert to libged, not libtclcad)
12:16.18CIA-55BRL-CAD: 03brlcad * r50922 10/brlcad/trunk/src/librt/librt_private.h: declare rt_advance_to_next_cell and rt_plot_cell in the private header since shoot and bundle call it, pointing out that they shouldn't have the rt prefix
12:17.07CIA-55BRL-CAD: 03brlcad * r50923 10/brlcad/trunk/src/librt/bundle.c: reorder to avoid forward declarations, move comment
12:34.35CIA-55BRL-CAD: 03brlcad * r50924 10/brlcad/trunk/NEWS: keith added initial support for importing assembly hierarchies to step-g in r49049
12:35.35CIA-55BRL-CAD: 03phoenixyjll * r50925 10/brlcad/trunk/src/librt/primitives/pipe/pipe_brep.cpp: startedge might points to the wrong place if adding endegde expands the capacity of the edge array, so we need to fix the pointer.
12:40.54CIA-55BRL-CAD: 03Phoenix 07http://brlcad.org * r3814 10/wiki/User:Phoenix/GSoc2012/Reports: /* Week 3 */
12:48.49CIA-55BRL-CAD: 03brlcad * r50926 10/brlcad/trunk/src/librt/primitives/nmg/nmg_fuse.c: it's a sort callback, so it should be reporting when values are equal too with a zero-return. undoubtedly related to the bogus sorting reported in r49898.
12:50.23CIA-55BRL-CAD: 03brlcad * r50927 10/brlcad/trunk/src/librt/primitives/nmg/nmg_fuse.c: er, they're floating point -- use EQUAL for comparison
12:52.47CIA-55BRL-CAD: 03brlcad * r50928 10/brlcad/trunk/TODO: make sure facetize/ev work on windows
12:54.22brlcadonly a few commits still pending review, if anyone wants to help: r50872, r49388, r49603
12:54.42brlcadthere's more, but that's a start
13:01.11CIA-55BRL-CAD: 03brlcad * r50929 10/brlcad/trunk/NEWS: bob added a new select command to archer for picking objects via the gui (using rects), richard then enabled the command for mged by adding it to the command table
13:08.04CIA-55BRL-CAD: 03brlcad * r50930 10/brlcad/trunk/NEWS:
13:08.04CIA-55BRL-CAD: richard changed some tolerances that were preventing some pipe bends from
13:08.04CIA-55BRL-CAD: tesselating. he changed some compares from 0.0 to SMALL_FASTF affecting 90
13:08.04CIA-55BRL-CAD: degree bends (untested) and pipes with inner/outer diameters. he also relaxed
13:08.04CIA-55BRL-CAD: near-zero comparison of the normal (which affects linear segments).
13:25.36*** join/#brlcad xth1 (~thiago@187.106.54.95)
13:35.37CIA-55BRL-CAD: 03brlcad * r50931 10/brlcad/trunk/src/librt/primitives/nmg/nmg_misc.c:
13:35.37CIA-55BRL-CAD: cleanup calling INIT_ZERO scope initializers instead of manually initializing,
13:35.37CIA-55BRL-CAD: for readability and a smidgen performance boost. also add the slew of missing
13:35.37CIA-55BRL-CAD: semicolons on vmath routines that were pooching formatting. finally, remove the
13:35.37CIA-55BRL-CAD: excessive logging hyperbole!!!!!
13:49.50*** join/#brlcad DarkCalf (~DarkCalf@173.231.40.99)
14:08.12CIA-55BRL-CAD: 03starseeker * r50932 10/brlcad/trunk/ (44 files in 24 dirs): Bump version number on BRL-CAD libraries from 19 to 20
14:14.32brlcadless than 25 remaining
14:18.35*** join/#brlcad d_rossberg (~rossberg@BZ.BZFLAG.BZ)
14:57.14``Erikray bradbury passed today
15:05.40brlcadmay his ashes burn at 451 degrees, I salute thee mr bradbury
15:05.53andrei_O_O
15:12.02*** join/#brlcad ksuzee (~ksu@193.151.107.42)
15:23.12d_rossbergbrlcad: "empty string" is different from NULL in C too, e.g. look at strlen()
15:27.14CIA-55BRL-CAD: 03d_rossberg * r50933 10/brlcad/trunk/src/conv/intaval/ (regtab.cpp regtab.h):
15:27.14CIA-55BRL-CAD: the clang warning was about the 'char* = ""' as parameter, the correct declaration is 'const char* = ""'
15:27.14CIA-55BRL-CAD: changed the declaration and implementation accordingly
15:38.39brlcadknows, just conventionally "more" different in C++
15:39.18brlcadksuzee: do you understand my reply?
15:39.58ksuzeebrlcad: hello
15:40.35brlcadhi :)
15:41.17ksuzee:) yes, I saw it. But I can't understand where I must include this *.c file
15:41.41ksuzeeAnd it's just get_args function
15:41.51ksuzeeIt will be used just in 2 files
15:42.17brlcadit doesn't matter what the function is or how few files
15:42.40brlcadyou have to tell the build system about the file, otherwise how does it know to compile it?
15:43.35ksuzeeyes
15:43.51ksuzeewhen i include *.h file
15:43.58ksuzeeeverything woks well
15:44.56ksuzeebut when I put the function to the *.c and do include *.h in this c-file, function isn't seen
15:45.36CIA-55BRL-CAD: 03Plussai 07http://brlcad.org * r3815 10/wiki/User:Plussai/GSoC_2012_log: /* 2 June 2012 */
15:45.51ksuzeeso where else must i write about this *.c file  to make it visible?
15:45.59brlcadksuzee: you keep saying the SAME thing over and over
15:46.17brlcadbut not reading what I'm saying, so I presume that means you don't understand what I'm saying
15:46.55brlcadyou must write about this new c file in the build system
15:47.04brlcadthat's the CMakeLists.txt file and the Makefile.am file
15:47.38ksuzeeI see...
15:49.17ksuzeeSo in Makefile it must be written here: bin_PROGRAMS = \
15:51.23ksuzeeI've found thes places, thanks
15:51.29ksuzeeI'll try now
16:03.27brlcadksuzee: you're using cmake to build, yes?
16:04.06brlcadwe have two build systems, that's why there's a CMakeLists.txt file and a Makefile.am file  (we're transitioning to the cmake build)
16:04.20ksuzeeI used it only first, when i downloaded sources. Now I do just make
16:04.32brlcadokay that's fine
16:05.02brlcadthe reason I ask is because obviously if you modify the Makefile.am and are using cmake+make to build, you're not going to see any change
16:05.08brlcadyou'd have to use the other build system
16:05.32brlcadyou don't have to test the other build system, though, the edits are very simple
16:05.49brlcadlets start with the build system you're not using, the Makefile.am file
16:06.05brlcadready?
16:06.51ksuzeeyes
16:07.11brlcadwhat are the two programs you're modifying
16:07.32ksuzeethere're bw-rle and pix-rle
16:07.40brlcadso we'll start with bw-rle
16:08.00brlcadopen src/util/Makefile.am, and search for bw-rle
16:08.19brlcadyou'll quickly see it's first listed under some "bin_PROGRAMS" variable
16:08.29brlcadyou see it?
16:08.38ksuzeeyes
16:09.05ksuzee$(bw_rle_OBJECTS) \
16:09.09brlcadwithout even knowing what the syntax of that Makefile.am file is, it's pretty clear that bin_PROGRAMS looks like a list of applications but is NOT where source files are described
16:09.36brlcadcontinuing to search for bw-rle .. the next thing you find is what?
16:10.07ksuzeemm, mans?)
16:10.34brlcadno, what is the next reference to bw-rle in that file after the bin_PROGRAMS line?
16:11.36ksuzeesources
16:12.24brlcadbe specific
16:13.14brlcadrunning grep, I find three references to bw-rle:
16:13.14brlcad[sean@gcc1-power7 brlcad]$ grep bw-rle src/util/Makefile.am bw-rle \
16:13.14brlcadbw_rle_SOURCES = bw-rle.c bw-rle.1 \
16:13.17ksuzeesource to bw-rle.c,
16:13.41brlcadthat's not what I mean about being specific :)
16:13.45brlcadI mean, what's the line?
16:13.54ksuzeebw_rle_SOURCES = bw-rle.c
16:13.57ksuzeesorry)
16:13.59brlcadright!
16:14.13brlcadso still I don't need to know the syntax of that file
16:14.29brlcadbut it looks like bw_rle_SOURCES is some sort of variable where you list sources files for the bw-rle program
16:14.39brlcadthe first file listed is bw-rle.c
16:14.44brlcadit's the only file listed
16:15.21brlcadyou moved a function from bw-rle.c into util.c so NOW the source code to bw-rle is in bw-rle.c AND util.c
16:15.32brlcadso you can add your util.c there
16:16.03brlcadunderstand?
16:16.40*** join/#brlcad PrezKennedy (~DarkCalf@173.231.40.99)
16:16.43ksuzeeyeees))))
16:16.59brlcadokay, now after that's done, a couple more things to check
16:17.12CIA-55BRL-CAD: 03Tbrowder 07http://brlcad.org * r3816 10/wiki/EBM: /* Converting PNG into x bitmap file */ show command and response as entered
16:17.34brlcadlook up a few lines before the bw_rle_SOURCES line, do you see the bombardier_SOURCES line?
16:17.47ksuzeeyes
16:17.53brlcadwhat do you see there?
16:18.02brlcadnotice anything interesting?
16:18.06ksuzeebombardier_SOURCES = bombardier.c bombardier.h
16:18.15ksuzeeh - file)
16:18.24brlcadexcellent
16:18.36CIA-55BRL-CAD: 03Tbrowder 07http://brlcad.org * r3817 10/wiki/EBM: /* Inserting ''ebm'' in MGED */ show command as entered (need response)
16:18.53brlcadso that's a pretty good indication that you should probably put util.h on the bw_rle_SOURCES line too
16:19.06ksuzeeyes
16:19.55brlcadone last thing to check
16:20.03brlcadwhen I ran grep, there were three matches
16:20.11brlcadthe first was the bin_PROGRAMS line
16:20.17brlcadthe second was the sources line
16:20.21brlcadwhat is the third?
16:20.35ksuzeebw-rle.1 \
16:20.40ksuzeeman file
16:20.50brlcaddo you think you need to do anything there?
16:20.54ksuzeeno
16:20.58brlcadyou sure?
16:21.26ksuzeemaybe the bw-rle.1 must be correct
16:21.30ksuzeeitself
16:21.49brlcadyou should be sure ;)
16:22.04brlcaddid you modify the man page?
16:22.23ksuzeeno)
16:22.39brlcadso then again
16:22.42brlcaddo you think you need to do anything there?
16:25.01ksuzeeIt's a manual page about how programm works. So why must it be modified?:)
16:25.40brlcadthat's what I'm asking you :)
16:25.51ksuzeeno
16:25.54brlcadyou sure? :)
16:25.58ksuzeeyes)
16:26.02brlcadexcellent! :)
16:26.40ksuzeeI feel like being at lie detector)
16:26.44brlcadhehe
16:27.16brlcadyou should be confident in your basic understanding of what you're doing to a file, whether it's source code or build system files or documentation
16:27.24brlcadhere's a great example :)
16:28.01ksuzeeI see)
16:28.05brlcadyou weren't 100% sure until I made you convince yourself, you should do that to yourself and if you're not sure, figure out why
16:28.20brlcadokay, so now that's done
16:28.32brlcadIF you had commit access, that'd be a perfect time to commit
16:29.06brlcadsince you don't, however, proceed with the second application
16:29.25brlcadyou're basically going to do the exact same thing for pix-rle
16:29.49brlcadfind lines that reference it, see if you need to change anything, add what you need to add
16:30.30ksuzeeok))
16:32.01brlcadwhen you finish that, let me know and we can proceed to the other build system where you can actually test your changes
16:34.20ksuzeeI finished
16:34.49brlcadgreat, so next up is CMakeLists.txt file
16:35.24brlcadagain, even without knowing what the syntax of that file is, you should expect needing to make similar edits to probably denote the addition of util.c and util.h -- you just have to find where/how to declare them
16:36.03ksuzeeI've done)
16:36.04brlcadso where do you find references to bw-rle ?
16:36.22brlcaduse a search mechanism like grep, don't just visually scan the file
16:36.46brlcadctrl-s if you're using emacs for incremental search
16:37.33ksuzeeI've done with grep
16:40.07brlcadwhat editor have you been using?
16:40.26ksuzeegedit - there's ctrl+f
16:40.40CIA-55BRL-CAD: 03tbrowder2 * r50934 10/brlcad/trunk/src/ (10 files in 6 dirs): correct grammar: it's => its (possessive form of 'it')
16:41.33brlcadso how many references to bw-rle did you find?
16:41.41ksuzee2
16:42.05brlcadand they are?
16:42.31ksuzeeBRLCAD_ADDEXEC(bw-rle bw-rle.c "libbu;libsysv;${UTAHRLE_LIBRARY}")
16:42.37ksuzeeamd man
16:42.38ksuzee*and
16:42.55brlcaddo you think you need to do anything with the second one?
16:43.47ksuzeeno)
16:43.50ksuzeeI'm sure)
16:43.52brlcadjust checking ;)
16:43.54ksuzee100%
16:43.55ksuzee)
16:44.19brlcadokay, so looking at the ADDEXEC line, it looks like a compact form of what you were just editing
16:44.35brlcadwithout knowing the syntax, it looks like the name of the program, sources, and libraries
16:45.00brlcadso now you just need to know 1) how to add multiple sources and 2) where does the header go
16:45.28brlcadso look in that file to see if you can find any other ADDEXEC examples that list more than one .c file
16:45.32brlcaddo you see any?
16:46.16ksuzeeNo, I don't
16:46.19ksuzeeaa
16:46.21ksuzeefound
16:46.28ksuzeeBRLCAD_ADDEXEC(dunncolor "dunncolor.c;dunncomm.c" libbu)
16:46.33brlcadexcellent
16:46.48brlcadso it looks like the syntax is to wrap it in quotes and separate them with semicolons
16:46.58brlcadso that answers #1
16:47.07brlcadwhat about #2, where do you think the header goes?
16:47.53ksuzeeCMAKEFILES(bombardier.h morphedit.tcl pldebug.c)
16:48.06ksuzeethe only .h i've found
16:48.52ksuzeein Makefile it was the only too)
16:49.14CIA-55BRL-CAD: 03tbrowder2 * r50935 10/brlcad/trunk/src/util/png-pix.1: correct name of program
16:49.57brlcadso unlike Makefile.am where you could just list it as a source, here, it looks like it's just listed separately
16:50.05brlcadso you can put util.h there
16:50.24brlcaddo it again for the other program and you'll be able to test your build
16:57.34CIA-55BRL-CAD: 03tbrowder2 * r50936 10/brlcad/trunk/TODO: add suggested TODO
17:02.04ksuzeefile is seen!
17:03.11brlcadnow when you compile, the linker error should go away, yes?
17:04.08ksuzeeyes))
17:04.16ksuzeeThank you a lot, Sean!
17:07.01ksuzeeAnd one more...Sorry, if I'm annoying=))) I really don't want to pester) But check, please, my summary and patches)
17:08.44*** join/#brlcad yukonbob (~bch@methodlogic.net)
17:08.53yukonbobhello, #brlcad
17:09.37brlcadksuzee: you're not annoying, glad we could work through it
17:10.02brlcadhopefully you understand why you were getting the linker failure and why the header worked
17:13.07CIA-55BRL-CAD: 03tbrowder2 * r50937 10/brlcad/trunk/doc/docbook/ (articles/en/TEMPLATE.xml system/mann/en/edit.xml): correct grammar: it's => its (possessive form of 'it')
17:22.03ksuzeeyes, I understood everything, thank you a lot) As for patches, all of them except one are opened
17:42.38brlcadthey'll go through another round of reviews sometime soon
18:01.10CIA-55BRL-CAD: 03Tbrowder 07http://brlcad.org * r3819 10/wiki/DSP: /* Preparing your height field data */ adding examples
18:02.09CIA-55BRL-CAD: 03Tbrowder 07http://brlcad.org * r0 10/wiki/Special:Log/upload: uploaded "[[Image:Cs-24-150h.png]]"
18:07.11CIA-55BRL-CAD: 03starseeker * r50938 10/brlcad/trunk/ (7 files in 3 dirs): Checkpoint rework of brlman - no graphical component yet, but a good start on the command line side. Thanks to Nick for the minimal workaround for Tcl not being able to launch man interactively.
18:33.18CIA-55BRL-CAD: 03Tbrowder 07http://brlcad.org * r0 10/wiki/Special:Log/upload: uploaded "[[Image:Dsp.png]]"
18:33.50CIA-55BRL-CAD: 03Tbrowder 07http://brlcad.org * r0 10/wiki/Special:Log/upload: uploaded "[[Image:Dsp2.png]]"
18:42.19CIA-55BRL-CAD: 03Tbrowder 07http://brlcad.org * r3823 10/wiki/DSP: /* Preparing your height field data */ adding examples
19:04.15*** join/#brlcad starseeker (~starseeke@BZ.BZFLAG.BZ)
19:05.01CIA-55BRL-CAD: 03Tbrowder 07http://brlcad.org * r3824 10/wiki/DSP: rearrange examples to fit existing sections better; make a new section to be filled in
19:09.05*** join/#brlcad n_reed (~molto_cre@BZ.BZFLAG.BZ)
19:14.32CIA-55BRL-CAD: 03tbrowder2 * r50939 10/brlcad/trunk/src/conv/asc/asc2pix.c: remove unused var
19:21.08*** join/#brlcad ksuzee (~ksu@193.151.107.42)
19:21.13CIA-55BRL-CAD: 03n_reed * r50940 10/brlcad/trunk/src/other/step/ (3 files in 2 dirs): MSVC fixes from SCL git 6226717 and 879f509
19:21.43CIA-55BRL-CAD: 03brlcad * r50941 10/brlcad/trunk/src/libbn/noise.c: initialize the pointer to NULL in case we end up returning it without setting it.
19:23.39CIA-55BRL-CAD: 03tbrowder2 * r50942 10/brlcad/trunk/src/conv/asc/asc2pix.c: add some comments and white space; adjust other ws for style
19:25.27CIA-55BRL-CAD: 03tbrowder2 * r50943 10/brlcad/trunk/src/conv/asc/asc2pix.c: uppercase labels
19:26.31CIA-55BRL-CAD: 03tbrowder2 * r50944 10/brlcad/trunk/src/conv/asc/asc2pix.c: ws, style
19:28.57CIA-55BRL-CAD: 03tbrowder2 * r50945 10/brlcad/trunk/src/conv/asc/asc2pix.c: more ws, style
19:29.07brlcad<PROTECTED>
19:29.54``Erikstarseeker committed and ran, there're several breakages after that
19:30.01CIA-55BRL-CAD: 03tbrowder2 * r50946 10/brlcad/trunk/src/conv/asc/asc2pix.c: more ws
19:30.31brlcadk
19:33.01CIA-55BRL-CAD: 03tbrowder2 * r50947 10/brlcad/trunk/src/conv/asc/asc2pix.c: more ws, style
19:38.09CIA-55BRL-CAD: 03tbrowder2 * r50948 10/brlcad/trunk/src/conv/asc/pix2asc.c: ease legibility & maintenance with an intermediate var
19:43.05CIA-55BRL-CAD: 03brlcad * r50949 10/brlcad/trunk/HACKING:
19:43.05CIA-55BRL-CAD: there are too many exceptions to the rule that have never been an issue (such as
19:43.05CIA-55BRL-CAD: var[i+2] or for (i=1; ..), remove the space around operators rule. the intent
19:43.05CIA-55BRL-CAD: was and still is to avoid complex math expressions that are error-prone to
19:43.05CIA-55BRL-CAD: interpret without whitespace around some of the operators that logically group
19:43.06CIA-55BRL-CAD: together. don't need to spend time reformatting, up to the author/editor's
19:43.07CIA-55BRL-CAD: discretion as long as the file is self-consistent.
19:43.23CIA-55BRL-CAD: 03tbrowder2 * r50950 10/brlcad/trunk/src/conv/asc/pix2asc.c: add some ws, tidy for ease of reading
19:45.31starseekerah, crud - sorry
19:45.35CIA-55BRL-CAD: 03starseeker * r50951 10/brlcad/trunk/src/brlman/brlman.tcl: whoops - helps to add the tcl file
19:49.55*** join/#brlcad Stattrav (u3131@gateway/web/irccloud.com/x-apaikvsbfbkepjjh)
19:50.56``Erikdamn nfs cache issue :/
19:53.45*** join/#brlcad jbschw (~jbschw@ool-4355ee10.dyn.optonline.net)
19:57.23*** join/#brlcad PrezKennedyX (~DarkCalf@173.231.40.99)
20:01.27*** join/#brlcad cristina (~quassel@188.24.71.66)
20:01.28*** join/#brlcad cristina (~quassel@unaffiliated/cristina)
20:03.01brlcadhi cristina
20:03.14cristinagood morning brlcad
20:05.34brlcadhm, make[2]: *** No rule to make target `/usr/lib/libfontconfig.so', needed by `lib/libtk.so.8.5'.  Stop.
20:19.57*** join/#brlcad anrgmrty (u6512@gateway/web/irccloud.com/x-fnhzxnbpqfmtgeqd)
20:23.26brlcadg'morning anrgmrty
20:23.33brlcadltns
20:32.55CIA-55BRL-CAD: 03brlcad * r50952 10/brlcad/trunk/src/libbn/noise.c: debugging some odd corruption, index into etbl directly instead of incrementing with the loop
20:39.17CIA-55BRL-CAD: 03brlcad * r50953 10/brlcad/trunk/src/libbn/noise.c: ws comment cleanup
20:47.19``Erik*snrkt* trolling politicians in the russian speaking world http://knowyourmeme.com/memes/how-does-one-patch-kde2-under-freebsd
20:54.35brlcadbeats cliff over the head with a wet fish
21:00.47``ErikO.o
21:10.43starseekerbrlcad: what'd I do now?
21:17.29CIA-55BRL-CAD: 03starseeker * r50954 10/brlcad/trunk/src/ (brlman/brlman.tcl tclscripts/man_browser.tcl): Add option to launch html man page viewer in brlman. Needs a little more work to properly customize the gui for this application.
21:21.47starseekerbrlcad: sorry about the breakage, but it should be OK now...
21:22.12brlcadthe build is back to working for me, though I had to blow away my cache for some reason
21:23.17starseekerhuh
21:23.33starseekerwas that the fish issue?
21:23.52brlcadthat was for something else, half-assing that injected a bug and cost others time :(
21:24.23starseekerah - sorry
21:24.46starseekerwill try to fix it - what is it?
21:32.22starseekerah - 49395
21:34.04brlcadtom's already working on it, good exercise
21:34.11brlcadyou didn't mean to (I hope!), just a bad line
21:34.47brlcadjust you even completely recognized it didn't make any sense that warranted a fish
21:34.48starseekernotes that was apparently on Friday... looks like brain burn out
21:44.36CIA-55BRL-CAD: 03n_reed * r50955 10/brlcad/trunk/src/other/step/include/ (21 files in 2 dirs): start adding symbols for MSVC dll import/export; SCL git ba12196
21:49.25brlcadanrgmrty: no activity today?
21:49.47starseekerbrlcad: ugh.  I'm still missing something.  Isn't UCHAR_MAX 256?
21:50.21brlcadeh, probably 255
21:50.37starseekerso moss.pix has values that *are* outside that range
21:50.48brlcadthat makes no sense, they can't be
21:50.54brlcada byte is a byte
21:51.32starseekerTom saw the same thing though - "it has
21:51.34starseekerseveral (20 or so) pixels that do (with values of 256 or 257)"
21:51.54brlcadyes?
21:52.21brlcadhe's reading that count from your mod
21:52.25brlcadyour mod is flawed
21:52.27starseekerah
21:52.44starseekermakes a second test, using brlcad's proposed filter
21:52.58brlcadcheck your pointer arithmetic
21:53.05brlcadyou take *pix, which is pix[0]
21:53.27brlcadint d = (int)(*pix + sizeof(unsigned char) * i);
21:53.28CIA-55BRL-CAD: 03tbrowder2 * r50956 10/brlcad/trunk/src/conv/asc/pix2asc.c: remove unnecessary multiple tests; fail on the first; test is just for Coverity
21:53.51brlcadso (int)(pix[0] + sizeof(unsigned char) * i)
21:54.06brlcadsizeof uchar is always 1 so you have
21:54.16brlcad(int)(pix[0] + 1 * i)
21:54.19CIA-55BRL-CAD: 03tbrowder2 * r50957 10/brlcad/trunk/src/conv/asc/pix2asc.c: make the 'test' valid
21:54.31starseekerah
21:54.36CIA-55BRL-CAD: 03tbrowder2 * r50958 10/brlcad/trunk/src/conv/asc/pix2asc.c: make the 'test' valid
21:54.54brlcadi goes from 0 to 2 so that's either (pix[0] + 1 * 0) or (pix[0] + 1 * 1) or (pix[1] + 1 * 2)
21:55.23CIA-55BRL-CAD: 03tbrowder2 * r50959 10/brlcad/trunk/src/conv/asc/pix2asc.c: use simpler print statement
21:55.45brlcadwith order of operations, that further reduces to (pix[0] + 0) or (pix[0] + 1) or (pix[0] + 2)
21:55.53CIA-55BRL-CAD: 03tbrowder2 * r50960 10/brlcad/trunk/src/conv/asc/pix2asc.c: delete unused var
21:56.12brlcadso basically you were taking the red component, adding 0, 1, or 2 to it and seeing if it was between 0 and 255
21:56.34brlcadnaturally fails if red component is 255, 254, 253 .. apparently about 20 of them in his test case
21:56.42starseekernods
21:56.50CIA-55BRL-CAD: 03tbrowder2 * r50961 10/brlcad/trunk/src/conv/asc/pix2asc.c: localize i
21:56.59brlcadway too complicated :)
21:57.08starseekergranting that's the incorrect way to do it - what would have been the correct way to get the right size to add to *pix?
21:57.21CIA-55BRL-CAD: 03tbrowder2 * r50962 10/brlcad/trunk/src/conv/asc/pix2asc.c: delete unused var
21:57.37CIA-55BRL-CAD: 03tbrowder2 * r50963 10/brlcad/trunk/src/conv/asc/pix2asc.c: delete unused var
21:58.11CIA-55BRL-CAD: 03tbrowder2 * r50964 10/brlcad/trunk/src/conv/asc/pix2asc.c: localize var pix
21:58.21brlcad*pix is the red component, *(pix+1) is the green, *(pix+2) the blue .. but then that's exactly why there's a bracket operator
21:58.26brlcadpix[0], pix[1], pix[2]
21:58.50starseekerah
21:59.04brlcadand for that, we have RED/GRN/BLU macros to include extra semantic information
21:59.09starseekertries to remember what he was thinking... and fails miserably
21:59.14brlcadpix[RED], pix[GRN], pix[BLU]
21:59.26brlcadyou were trying to index into the array using pointer arithmetic
21:59.53starseekerright... I guess that must have been the "getting frustrated with Coverity not liking any of our bounds checking attempts" phase
22:02.27starseekerwell, Tom's mods pass a round-robin test with moss.pix
22:02.39starseekerwe'll see if Coverity likes it
22:04.49brlcadit should, it really does just care that you check lower and upper bounds
22:05.01brlcadapplied that same method in dozens of other places
22:06.00brlcadcoincidentally ran into that pix2asc bug just a couple days ago, it halted reading input (presumably on a white pixel) but didn't have time to investigate
22:06.25starseekerseems to remember a number of attempts by both ``Erik and myself to do that early on that kept coming back... should probably go back at some point and see what we were doing wrong
22:06.40starseekersurprised the regression tests didn't catch it
22:07.10starseekerguess we don't go back to ascii anywhere
22:07.50starseekerwonders if it's worth adding...
22:09.52brlcadworth having a test for every tool but not worth actively spending time on it this month...
22:10.04brlcadbesides, maybe tom will
22:10.18starseekernods
22:10.22brlcadhe's awesome with the testing ;)
22:10.40starseekernow if only I can get him to add new files to the build logic :-)
22:10.51starseekerstill, those are easy fixes
22:15.50anrgmrtybrlcad: hi
22:29.41CIA-55BRL-CAD: 03n_reed * r50965 10/brlcad/trunk/src/other/step/include/exppp/exppp.h: more symbols for MSVC dll import/export; SCL git f853b99
22:37.14CIA-55BRL-CAD: 03anrgmrty * r50966 10/brlcad/trunk/src/conv/g-voxel.c: multiple rays per voxel allowed and some bugs fixed
22:45.17CIA-55BRL-CAD: 03Anuragmurty 07http://brlcad.org * r3825 10/wiki/User:Anuragmurty: /* Development Log */
22:55.46brlcadanrgmrty: what's with the += 0 ?
22:57.03anrgmrtyactually i thought it will have to be replaced later with adding to some float value.. which will be non-zero
22:57.41brlcadk
22:59.11anrgmrtybrlcad: i have a question with multiple rays.. i am shooting them at the same point for now since daniel says the location of the points of shooting is to be discussed...
23:00.30anrgmrtyassuming we have ascertained the locations whr the rays have to be shot on the voxel, how are we supposed to determine the volume?
23:00.53anrgmrtyuse something like simpson's rule?
23:03.44CIA-55BRL-CAD: 03starseeker * r50967 10/brlcad/trunk/src/brlman/brlman.tcl: Base brlman's behavior (Tk graphical or man command) on what's actually available on the system.
23:40.12brlcadanrgmrty: technically, you're not determining the volume
23:40.25brlcadyou're deriving a voxel representation
23:41.33brlcadyou could certainly add up the volumes of the voxel cells and have your volume estimate (which is more of a rectangle method)
23:41.51anrgmrtyhmm
23:42.06brlcadbut computing volume isn't the goal, having a new representation of the geometry is
23:42.20brlcadrtweight and gqa both already compute volume
23:42.36brlcadthe latter does it *very* well
23:43.07anrgmrtyyes.. so when multiple rays are involved i got confused about how to proceed.. we see the proportion of those rays that pass a certain threshold?
23:43.23anrgmrtyfor voxels i mean
23:43.30brlcadgqa fires a grid of rays from three axes, progressively refining until the solution converges or defined limits are reached, and runs in parallel
23:43.47brlcadbut it doesn't retain the representation, it just computes
23:43.47anrgmrtyhmm..
23:44.04anrgmrtyok
23:44.42brlcadthat's why looking at what rtweight and gqa during the bonding period was so important and repeated
23:45.30brlcadit does 90% of what you need, but isn't easy to understand if you don't know anything about geometry, librt's API, ray tracing, etc
23:47.54brlcadso yeah, don't worry about volume -- you're trying to generate a 3d volumetric bitmap for geometry
23:48.31brlcadanrgmrty: have you successfully run any of the tools mentioned yet?
23:49.10anrgmrtyyes
23:49.44brlcadwhich?
23:49.59anrgmrtyviewweight.c
23:50.05anrgmrtyi mean rtweight
23:50.19anrgmrtyand gqa
23:50.23anrgmrtyboth
23:50.55anrgmrtybut only run them.. not understood the code well
23:50.59brlcadyou ran gqa?  how'd you provide densities?
23:51.17anrgmrty.density file
23:51.38anrgmrtyi put in random values as mentioned
23:51.48anrgmrtynot any specific ones
23:52.23brlcaddid you import that into a _DENSITIES object?
23:53.02anrgmrtyno. i do not understand what this means. i just modified the .density file
23:53.27brlcadgqa doesn't use a .density file
23:54.11anrgmrtyok..
23:55.26anrgmrtyis that only for rtweight ?
23:55.34brlcadsort of
23:55.39brlcadrtweight directly uses it
23:55.46brlcadbut it must be imported for gqa to use it
23:56.20brlcadyou may have ran gqa in some other mode, not computing mass
23:56.33brlcaddo you remember the command you ran?
23:57.05anrgmrtyhmm.. then i have not run GQA.. the other thing i did was rtexample.c only
23:57.27anrgmrtywhich i have been modifying for the present g-voxel

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