IRC log for #brlcad on 20100622

00:18.58CIA-42BRL-CAD: 03brlcad * r39637 10/brlcad/trunk/src/other/tcl/generic/tclDecls.h: quell strict compilation failures reintroduced with the update to 8.5.8 without the r38389 quellage.
00:20.41brlcadstarseeker: did you update any other external deps recently, or just tcl/tk?
00:26.10CIA-42BRL-CAD: 03brlcad * r39638 10/brlcad/trunk/NEWS: document the updates to tcl/tk since 8.5.1 including the intermediate update to 8.5.6 on 2009-02-13, and now the update to 8.5.8; update supports mac platform support and new archer gui developments.
02:13.37starseekerbrlcad: I believe just tcl/tk, not counting the openNURBS thing a while back and adding tktable
02:14.19starseekernuts, sorry - thought I had forwarded ported all the necessary changes
02:31.41starseekerOOoooo.  Category Theory in Isabelle/HOL - http://afp.sourceforge.net/entries/Category2.shtml
02:32.53*** join/#brlcad yukonbob (~svs@S0106001125477e9c.ok.shawcable.net)
02:42.16*** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1)
03:10.44*** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1)
03:26.20CIA-42BRL-CAD: 03brlcad * r39639 10/brlcad/trunk/src/libdm/dm-ogl.c: USE_PROTOTYPES should not be used any more
03:34.38CIA-42BRL-CAD: 03brlcad * r39640 10/brlcad/trunk/include/ (fb.h fbio.h): ws cleanup
03:53.47brlcadyukonbob: that could be some custom ray-shooter, but it sounds a bit like you're wanting something like the UV coordinates of a model printed in 2D
03:55.06brlcador just not understanding what it is you're trying to accomplish
03:56.02brlcadif you just want a 2D ouline of the model, you'd just rtedge it and add some labels to the image
03:58.04Raliththe UV thing sounds closest to what he described
04:11.13CIA-42BRL-CAD: 03brlcad * r39641 10/brlcad/trunk/src/mged/ (mged.c mged.h set.c setup.c): remove mged_variable_setup() as it's not used. quell warnings while trampling through.
04:13.24brlcadyeah
04:16.26CIA-42BRL-CAD: 03brlcad * r39642 10/brlcad/trunk/src/mged/setup.c: only delete it if the original pointer is non-null
04:17.52brlcadstarseeker: on a quick test of head, 'red' still doesn't seem to do anything
04:19.10CIA-42BRL-CAD: 03brlcad * r39643 10/brlcad/trunk/TODO: red tested. doesn't work. must be unbusted before release.
05:48.42*** join/#brlcad Ralith (~ralith@216.162.199.202)
06:11.21*** join/#brlcad Ralith (~ralith@216.162.199.202)
09:45.40*** join/#brlcad CIA-40 (cia@208.69.182.149)
10:05.58*** join/#brlcad mafm (~mafm@83.45.73.17)
11:43.23*** join/#brlcad SWPadnos_ (~Me@dsl107.esjtvtli.sover.net)
12:16.24*** join/#brlcad Stattrav (~Stattrav@117.192.146.39)
12:19.18starseekerbrlcad: yeah, I'm getting reports of that - I plan to tackle it today
12:20.15starseekermutters under his breath... red has been busted since v5 was introduced, wonder that it didn't wipe out data...
13:38.09*** join/#brlcad brlcad (~sean@BZ.BZFLAG.BZ)
13:38.24brlcad~seen kanzure
13:38.26ibotkanzure <bryan@dhcp-84-36.me.utexas.edu> was last seen on IRC in channel #brlcad, 125d 16h 38m 29s ago, saying: 'ah maybe http://brlcad.org/xref/source/src/librt/primitives/'.
13:53.09brlcadwonders if d-lo is going to do anything else with mkbuilding .. start of something interesting there!
14:14.43CIA-40BRL-CAD: 03brlcad * r39644 10/brlcad/trunk/src/proc-db/ (Makefile.am metaballs.pl):
14:14.44CIA-40BRL-CAD: include an example procedural geometry generator perl script from tom browder
14:14.44CIA-40BRL-CAD: that helps create metaballs. script takes a simple text input and produces an
14:14.45CIA-40BRL-CAD: mged tcl script. I modified the script slightly to catch and quiet db_lookup
14:14.45CIA-40BRL-CAD: kill failures.
14:31.49CIA-40BRL-CAD: 03brlcad * r39645 10/brlcad/trunk/src/librt/db_tree.c: make sure ma_color is valid before printing it
15:33.15CIA-40BRL-CAD: 03brlcad * r39646 10/brlcad/trunk/src/proc-db/metaballs.pl: fix the HERE ws destruction. specify warnings with a use statement instead of via the -w argument for implementations of env that don't support arguments.
16:13.20CIA-40BRL-CAD: 03brlcad * r39647 10/brlcad/trunk/src/proc-db/ (Makefile.am spiral.pl):
16:13.20CIA-40BRL-CAD: add another contributed perl script that generates geometry. this script, from
16:13.21CIA-40BRL-CAD: bryan bishop (aka kanzure) generates a 'gear spiral' with teeth. presently
16:13.31CIA-40BRL-CAD: generates overlapping regions, but the basic shape is there. script was
16:13.31CIA-40BRL-CAD: provided from bryan and put into the public domain.
16:15.35CIA-40BRL-CAD: 03brlcad * r39648 10/brlcad/trunk/AUTHORS: special thanks to bryan bishop for his example spirals.pl procedural geometry perl script. not a code contribution to a completed feature or existing code, hence special thanks categorization.
16:43.05starseekerer... if ged_red is now the red command functionality, what is red.c doing in src/mged?
16:43.48CIA-40BRL-CAD: 03brlcad * r39649 10/brlcad/trunk/src/proc-db/spiral.pl:
16:43.49CIA-40BRL-CAD: cleanup. enable perl warnings, remove unnecessary comments, cleanup formatting,
16:43.50CIA-40BRL-CAD: test for objects before creating them (probably should just delete the file or
16:43.50CIA-40BRL-CAD: abort early). create combs instead of regions since they overlap.
16:46.50starseekerinvestigates...
16:56.30CIA-40BRL-CAD: 03brlcad * r39650 10/brlcad/trunk/src/proc-db/spiral.pl: test if spiral.g exists so we don't have to test for each object existing.
17:01.28CIA-40BRL-CAD: 03brlcad * r39651 10/brlcad/trunk/src/proc-db/spiral.pl: create a proper region
17:09.57*** join/#brlcad packrat (~packrator@99-67-225-40.lightspeed.livnmi.sbcglobal.net)
17:51.01*** join/#brlcad mafm (~mafm@83.45.73.17)
17:54.12brlcadhello packrat
17:54.20brlcadand mafm :)
17:54.21*** join/#brlcad yukonbob (~svs@S0106001125477e9c.ok.shawcable.net)
17:54.46packratsup
17:57.24CIA-40BRL-CAD: 03brlcad * r39652 10/brlcad/trunk/BUGS: edcomb command also seems to be busted, similar to red -- doesn't seem to do anything. at least, testing showed no changes applied to geometry and it would crash if you provided one less arg than expected.
17:58.30*** join/#brlcad kanzure (bryan@dhcp-84-252.me.utexas.edu)
17:59.39kanzurebrlcad: btw, i didn't mean to imply that you had an affiliation with OpenCASCADE
17:59.46kanzureobviously it was a typo and i meant to say BRL-CAD
17:59.53kanzure"Btw, if you want to integrate STEP into OpenCASCADE, you should" <-- where the typo occured
18:00.11kanzuredo you have a copy of NIST SCL that actually compiles? :)
18:01.21mafmhi
18:08.33*** join/#brlcad dtidrow (~dtidrow@c-71-238-51-148.hsd1.mi.comcast.net)
18:29.32brlcadkanzure: no problem, wasn't sure if it was a typo or not
18:30.04brlcadwe have a copy in src/other/step that should compile
18:30.46brlcadit's now a required part of our build, that's part of the reason why we took over its maintenance (we need it, nist is done with it)
18:39.39kanzureneato
18:39.43kanzurewell, props to you guys for doing that
18:39.55kanzuredid the config/make file need to be rewritten?
18:46.06brlcadthat was a long time ago, but I believe it did.  plus, that was the easiest means to integrate it with our build
18:46.39brlcadprevious was a pretty quick build system, not very portable without editing files
18:47.40kanzurethanks for the email
18:47.56kanzurei'm surprised that you're allowed to distribute the STEP docs (even for BRL-CAD purposes or mutual development efforts)
18:48.01kanzurethat's amazing and great news
18:49.21brlcadthat's because it wouldn't technically be distributing the docs, it's still to "the same group for the same purpose"
18:49.32brlcadthat'
18:49.45kanzureright
18:51.11brlcadhence the need for it to be a brl-cad purpose, and we'd probably have to craft a simple agreement that makes that explicit just in case some dev posted the specs up on limewire and iso came hunting us down with a legal attack
18:51.21kanzuresure
18:51.41kanzurei have the EXPRESS files already
18:51.47kanzureis there "Other Stuff"?
18:51.54kanzureyou mentioned pdf files and ps files? do they have anything useful/interesting in them
18:52.01brlcadmost of the open source community has been unaware/ignorant of STEP until recent years
18:52.32kanzurebah, most people don't even know the difference between CSG/constraint-based-modeling/solids and mesh is
18:52.42brlcadpartly because of the licensing cost (I mean, *damn*) ..
18:52.53kanzureiso.org sells 10303 specs for $350/ea
18:53.13brlcadbut also just complexity .. it's meant to solve everything and most people care about their tiny niche problem of the moment
18:53.22brlcadI call it "the union of all cad formats"
18:54.10brlcadyeah, ARL spent a couple grand on the specs when we originally started working on STEP
18:54.40kanzureahh
18:54.45kanzurethat's also kinda sad- the military didn't have them already?
18:55.00brlcadhad they purchased all of 10303, I think I remember them saying it would have cost something like $20k
18:55.05kanzurehahah
18:55.11brlcadand would have taken up about 50' of bookshelf space if printed
18:55.12kanzure(btw, this is retarded)
18:55.20kanzureoh well
18:55.31kanzurefor a while my only option for STEP-related stuff was reading the OpenCASCADE code base
18:55.35kanzuresince they have an implementation of STEP
18:55.43kanzurewhich is not the best way to learn a standard ;)
18:55.48brlcadsomeone int he military might have already had a copy
18:55.50kanzurei dunno if you've ever looked into their source code
18:55.58kanzurebut it's pretty terrible
18:56.03brlcadbut then you get into the whole gray licensing area of a "group" and a purpose
18:56.23kanzurewell, all of this is making me pretty happy today
18:56.44brlcadintentionally have not ever looked at the opencascade source code as their license is incompatible
18:57.17brlcadi've pretty consistently heard others say it's terrible, though
18:58.18kanzurei have no idea how anyone does development at Matra Datavision / or whatever they call themselves these days
18:59.33kanzurebrlcad: ok. i'd like to get the STEP docs, and implement some stuff if i can. but i already have some of the documents, as you know
18:59.46kanzureso i need to make sure this is worthwhile- especially if you need to draft up a document for me to sign or something
18:59.56kanzureif i already have the same files, then it's pointless to draft up a document for me to sign
19:09.26brlcadquite
19:09.50brlcadI'll take a look at what's on archive and compare them to ours, see how close it is
19:10.49starseekeryeah, here's the opencascade license - iirc the having to send modifications back to the original developer is a no go, and possibly the requirement to acknowledge use of the software
19:11.03starseekerhttp://www.opencascade.org/getocc/license/
19:11.14starseekerwe'll let FreeCAD play with them
19:12.00kanzurefreecad has a pretty terrible interface
19:12.04kanzurei suggest http://heekscad.org/ instead
19:12.39starseekerI've never successfully compiled HeeksCAD
19:13.35starseekerFreeCAD definitely has its issues, but I've never been sure how many of those were just due to a non-mature build process
19:14.52starseekereither way, glad to see activity in the open source CAD arena - if openCASCADE has features people can use it's nice to have projects making them available
19:15.08starseeker(we'll eventually crush them all of course :-P)
19:15.25kanzurestarseeker: really? HeeksCAD has never given me trouble compiling
19:15.31kanzureare you on osx or something bizarre like that?
19:15.36starseekerGentoo Linux
19:15.40kanzurehuh
19:15.44starseekerit's been a while since I looked at it
19:15.49starseekerthey may have improved
19:15.54kanzurei wouldn't expect any problems on gentoo. that's weird.
19:16.02starseekerfor a while, just getting openCASCADE working was an adventure
19:16.42starseekergentoo has been kinda slow to the open source CAD game, in some ways - they've completely ditched QCAD now since it's not been ported to QT4 on the open source side
19:17.06kanzurethere should be a package for heekscad now, btw
19:17.10kanzurethere's definitely a debian package somewhere
19:17.12starseekerah, sweeet
19:17.15starseekerwill look
19:17.16kanzuredebian has a tool called 'alien' to convert foreign packaging formats
19:17.21kanzurei dunno if gentoo has something equivalent
19:17.40starseekernot typically - unless there's no alternative they'll compile things from source
19:17.53starseekerso binary rpms and debs are fairly useless
19:18.40starseekergentoo HATED our use of external libs in the tree, and Fedora is the same way - they want everything broken out
19:19.39starseekergenerally we can use external libs OK, but there are some (opennurbs, step, utahrle) where we're pretty much it and there isn't a workable upstream (at least for the use we're making)
19:20.54starseekerI can see their point in some ways, since the worry about security fixes and what not, but it ends up being a real pain all around
19:21.49starseekerBRL-CAD is intended to "just work" when you build it from the tarball
19:25.13CIA-40BRL-CAD: 03bob1961 * r39653 10/brlcad/trunk/src/libged/putmat.c: Modify putmat to "get" the matrix if a matrix is not specified.
19:29.45*** join/#brlcad jam555 (~on_Chatzi@adsl-99-114-165-115.dsl.okcyok.sbcglobal.net)
19:29.57*** part/#brlcad jam555 (~on_Chatzi@adsl-99-114-165-115.dsl.okcyok.sbcglobal.net)
19:30.31CIA-40BRL-CAD: 03bob1961 * r39654 10/brlcad/trunk/src/libged/ (Makefile.am combmem.c): Added the combmem command for setting/getting a combinations members.
19:31.17CIA-40BRL-CAD: 03bob1961 * r39655 10/brlcad/trunk/src/libtclcad/ged_obj.c: Added combmem to the command table.
19:33.17CIA-40BRL-CAD: 03bob1961 * r39656 10/brlcad/trunk/src/tclscripts/lib/Ged.tcl: Added a method for combmem.
19:36.44CIA-40BRL-CAD: 03brlcad * r39657 10/brlcad/trunk/src/libged/wdb_obj.c: remove dead code. old style and at least some of the functions referenced don't exist.
19:38.29CIA-40BRL-CAD: 03brlcad * r39658 10/brlcad/trunk/src/libged/wdb_obj.c: quell size_t warnings
19:39.40``Erikinclude/ged.h not committed?
19:41.10brlcadprolly not
19:41.17CIA-40BRL-CAD: 03brlcad * r39659 10/brlcad/trunk/src/libged/putmat.c: restructure to see ged_getmat() before using it. clean up indendation and comments, fix constness.
19:41.22``Erikah, that was bob
19:43.48CIA-40BRL-CAD: 03bob1961 * r39660 10/brlcad/trunk/include/ged.h: Added a declaration for combmem.
19:48.06CIA-40BRL-CAD: 03brlcad * r39661 10/brlcad/trunk/src/libged/combmem.c: static/HIDDEN functions should not have a ged_ prefix. instead, use the name of the command/group that they belong to or leave them without prefix. use HIDDEN instead of static when declaring private library functions.
19:52.55*** join/#brlcad Stattrav (~Stattrav@117.192.128.175)
19:53.01CIA-40BRL-CAD: 03brlcad * r39662 10/brlcad/trunk/src/libged/combmem.c: quell all compilation warnings. shadowing vars, unused vars, and size_t comparisons.
19:53.43CIA-40BRL-CAD: 03brlcad * r39663 10/brlcad/trunk/src/libged/combmem.c: ws consistency indent cleanup
19:54.29CIA-40BRL-CAD: 03brlcad * r39664 10/brlcad/trunk/src/libged/combmem.c: s/GED_GETCOMBTREE/COMBMEM_GETCOMBTREE/g
20:00.41kanzurebrlcad: should i send you a directory listing of files that i have re: ISO 10303 from archive.org?
20:11.20brlcadnot necessary
20:11.35kanzureokie dokie
20:13.32brlcadwoot, http://brlcad.org/tmp/spirot.png
20:13.39brlcad(rotations)
20:17.32*** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1)
20:23.15brlcadand http://brlcad.org/tmp/spirot2.png
20:27.09CIA-40BRL-CAD: 03brlcad * r39665 10/brlcad/trunk/src/proc-db/spiral.pl: rotate the boxes as we spiral outward creating a box-ish tunnel. angle of rotation is 90 minus arctan(y / x).
20:30.44kanzureawesome.
20:31.03kanzurealso, i think one of the things on the todo list for that was a z-axis thing. shouldn't be too hard.
20:31.14kanzurebut it's also kinda unnecessary ;)
20:37.22CIA-40BRL-CAD: 03brlcad * r39666 10/brlcad/trunk/src/util/pixhalve.c: fix memory corruption on exit. we weren't freeing the right pointers.
20:39.36brlcadwho cares about necessary, it's fun ;)
20:40.06CIA-40BRL-CAD: 03brlcad * r39667 10/brlcad/trunk/NEWS: fixed a minor bug in pixhalve where we weren't releasing the memory we allocated properly. now frees the memory and avoids the noisy Deallocation warning on Mac OS X.
20:44.19CIA-40BRL-CAD: 03brlcad * r39668 10/brlcad/trunk/src/util/pixhalve.c: cleanup. remove forward decls, quell all warnings, upgrade to size_t.
20:45.56*** join/#brlcad Ralith (~ralith@216.162.199.202)
20:47.08CIA-40BRL-CAD: 03brlcad * r39669 10/brlcad/trunk/src/util/pixhalve.c: plug lil memory leak on exit, free out in/out bufs.
20:47.09*** join/#brlcad Ralith (~ralith@216.162.199.202)
20:49.41brlcadneat-o:  http://brlcad.org/tmp/spirot2.png
20:50.08brlcadcalcs aren't quite right, as can be seen in that picture
20:50.16brlcadkeypoint is wrong
20:50.33brlcadit's off half the box width
22:04.29RalithI have no idea what the significance of that is but it's neat looking.
22:50.52``Erikyowza, that's a lot of rain O.o
22:51.18``Erikralith: spiral rotation script written to generate geometry procedurally using perl O.o :)
22:55.39``Erikdamn, now it's hail
22:59.46kanzure:)
23:52.50dtidrow``Erik: hailing outside there?
23:54.27dtidrowoh, nm - almost an hour ago
23:55.21dtidrowlooks like the mid-atlantic region is getting clobbered

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