| 00:02.10 | CIA-128 | BRL-CAD: 03starseeker * r49839 10/brlcad/trunk/ (3 files in 2 dirs): Break the BRL-CAD target definition macros out into their own file, leave actual Utility macros in Util |
| 00:13.16 | CIA-128 | BRL-CAD: 03Bhinesley 07http://brlcad.org * r3404 10/wiki/User:Bhinesley/gsoc2012/search_proposal: Layout proposal outline for 'search -exec' idea... WIP |
| 00:37.50 | *** join/#brlcad Neil__ (~chatzilla@117.229.94.124) |
| 00:43.27 | *** join/#brlcad Neil__ (~chatzilla@117.229.94.124) |
| 01:40.03 | crdueck | browsing through some source code, i see the "register" keyword being used from time to time. I was under the impression that this was generally useless as the compiler will make this decision anyways. what is the general consensus on this? |
| 01:46.57 | brlcad | crdueck: "yes" |
| 01:47.34 | brlcad | crdueck: it does depend heavily on the compiler and "most" modern compilers do a reasonable job of register allocation |
| 01:49.54 | brlcad | a program that correctly marks registers in critical code sections will still beat out most compilers, but it's tricky to get it right |
| 01:50.56 | brlcad | there are a lot thrown all around brl-cad are are still there today just because the code dates back to when it used to make a huge difference |
| 01:51.37 | brlcad | that and the last time I categorically removed all of our register keywords, our performance measurably decreased |
| 01:52.17 | crdueck | interesting, thats reason enough to keep them in then. thanks |
| 01:52.54 | brlcad | yep, just not reason enough to add them to new code ;) |
| 01:54.04 | brlcad | nor reason enough to declare as a coding practice to eliminate or encourage use |
| 02:04.56 | CIA-128 | BRL-CAD: 03starseeker * r49840 10/brlcad/trunk/ (8 files in 6 dirs): (log message trimmed) |
| 02:04.56 | CIA-128 | BRL-CAD: Significant simplification of the distcheck logic - reduce it to two macros, one |
| 02:04.56 | CIA-128 | BRL-CAD: of which calls the other. Add documentation, improve behavior in some corner |
| 02:04.56 | CIA-128 | BRL-CAD: cases, enforce a convention we already pretty much followed by default of using |
| 02:04.56 | CIA-128 | BRL-CAD: relative paths to specify source files in the source tree (this is a robust way |
| 02:04.57 | CIA-128 | BRL-CAD: to spot generated files in build target definitions, since support for an out of |
| 02:04.58 | CIA-128 | BRL-CAD: directory build requires that they be specified with their full CMAKE_BINARY_DIR |
| 02:09.21 | CIA-128 | BRL-CAD: 03starseeker * r49841 10/brlcad/trunk/ (3 files in 2 dirs): CMakeFiles is specific to BRL-CAD, rename it accordingly |
| 02:10.21 | CIA-128 | BRL-CAD: 03starseeker * r49842 10/brlcad/trunk/misc/CMake/BRLCAD_CMakeFiles.cmake: Fix header |
| 02:16.11 | starseeker | "Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away." - Antoine de Saint-Exupery |
| 02:36.00 | brlcad | crdueck: excellent .. but does it still work too? :) |
| 02:36.41 | brlcad | suggest importing some BoT geometry to make sure it still renders |
| 02:36.42 | crdueck | brlcad: the patch i just submitted? |
| 02:36.52 | brlcad | nods |
| 02:37.01 | crdueck | will do |
| 02:37.45 | brlcad | we actually just discovered a bug fixed a month or two ago caused after exactly that kind of mod, had been in the code for nearly a year undetected |
| 02:38.08 | brlcad | caused .g corruption on 64-bit big endian platforms |
| 02:40.38 | brlcad | hehe, already have more than 11 applications |
| 02:40.52 | brlcad | but more than half are mistake/duplicates from one person :) |
| 02:43.06 | andrei_ | I have been struggling for the past days to get the selective-repeat protocol implementation to work. Hopefuly friday I can get back to writing the rest of the proposal |
| 02:43.24 | brlcad | selective-repeat? |
| 02:44.25 | andrei_ | it s a data link - level protocol that only resends from server to client the corrupted/missing packages instead of sending all after the corrupted one. |
| 02:44.42 | bhinesley | hey, I just learned about that last quarter :) |
| 02:44.54 | bhinesley | as opposed to go-back-n |
| 02:45.10 | andrei_ | I tried go-back-n on a large file with a high corruption rate |
| 02:45.17 | bhinesley | yeah it's terrible |
| 02:45.17 | andrei_ | I freezed my system :p |
| 02:45.33 | brlcad | andrei_: actually my question isn't what it is but whether it was related to a libpkg patch |
| 02:45.56 | brlcad | :) |
| 02:46.20 | andrei_ | package management system ? |
| 02:46.46 | andrei_ | speaking of patches |
| 02:46.56 | brlcad | that's brl-cad's old network 'package' library |
| 02:47.27 | brlcad | one of our suggested ideas was to enhance it, so ears perked up with selective-repeat ;) |
| 02:48.56 | andrei_ | I will have to decide upon a second project, so this might be interesting:) |
| 02:49.02 | andrei_ | anyhow , I wanted to ask |
| 02:49.32 | andrei_ | I did submit the test unit patch , but I wouldn't mind working on something else, perhaps a slight bit more difficult |
| 02:49.50 | andrei_ | there was a discussion a few days ago about splitting two libraries |
| 02:50.23 | andrei_ | has it been done so far ? is it still needed? |
| 02:50.27 | brlcad | not incredibly hard but the split would be too much for a patch |
| 02:51.11 | brlcad | more effective to have a patch that has some meat to it, requires comprehending and/or writing logic |
| 02:51.51 | brlcad | unit test is certainly good, but kicking it up a notch might be interesting |
| 02:52.08 | brlcad | first project is refactoring, right? |
| 02:52.13 | andrei_ | yes |
| 02:53.08 | brlcad | maybe see if you can figure out a minimal api-clean way to eliminate a single global var |
| 02:53.31 | brlcad | ls -la src/*/globals.c |
| 02:53.43 | brlcad | lots to choose from ;) |
| 02:54.16 | andrei_ | At first I was pretty intimidated by the size of the project/ source code in general |
| 02:54.53 | brlcad | that will require understanding what that variable actually does, where it's being used, and consideration of a variety of ways on how to eliminate it somehow |
| 02:54.54 | andrei_ | but I do discover things by reading / understanding the code |
| 02:55.02 | brlcad | that's great |
| 02:55.35 | andrei_ | I believe that there is a category of programming " tricks" that you only learn when you see them or need to apply them. |
| 02:56.10 | brlcad | "need to apply them" can be very subjective ;) |
| 02:56.29 | brlcad | that's like knowing when you "need" to add salt to a recipe |
| 02:56.59 | andrei_ | what I meant is that sometimes I have a very good idea about what I should do to solve the problem |
| 02:57.14 | andrei_ | but it takes time and several papers to decide upon the implementation |
| 03:08.23 | brlcad | you'll find a veritable buffet of such details throughout code |
| 03:08.45 | brlcad | but it is still rather subjective, knowing when to apply what |
| 03:09.13 | brlcad | some best practices but you will be naturally biased to what you've seen before or read about or recently learned, etc |
| 03:10.00 | brlcad | like I said, it's like knowing when to add salt to a recipe -- it can be very formulaic or it can be outright art or it can be outright wrong depending on the circumstances ;) |
| 03:10.06 | brlcad | everything is within a context |
| 03:12.09 | andrei_ | indeed :) |
| 03:12.30 | andrei_ | so is dynamic programming |
| 03:41.40 | *** part/#brlcad Neil__ (~chatzilla@117.229.94.124) |
| 03:52.39 | crdueck | i'm having trouble compiling brlcad from svn source, it compiles fine using the source pkg available from my linux distro. the only difference i can see is that the cmake option "Build static libraries" is set to off in the version that compiles fine. I'm still mostly unfamiliar with cmake, how can i change this setting? |
| 04:20.37 | starseeker | crdueck: what's the error? |
| 04:22.57 | crdueck | <PROTECTED> |
| 04:51.37 | starseeker | do you have the libXi-dev package installed? |
| 04:54.46 | crdueck | i do have libXi, but not dev. |
| 04:54.53 | crdueck | i'll try updating |
| 04:58.28 | crdueck | still getting errors from cmake with libxi-git |
| 07:57.36 | *** join/#brlcad bhinesley (~bhinesley@adsl-108-79-235-86.dsl.bkfd14.sbcglobal.net) |
| 08:28.46 | *** join/#brlcad jordisayol (~jordisayo@unaffiliated/jordisayol) |
| 08:31.38 | *** join/#brlcad stas (~stas@82.208.133.12) |
| 09:39.17 | brlcad | crdueck: have you tried using the vm image -- we pulled that together so students don't have to fuss with build issues until after they're selected |
| 09:39.29 | brlcad | https://sourceforge.net/projects/brlcad/files/BRL-CAD%20for%20Virtual%20Machines/ |
| 09:40.05 | brlcad | that's a one-stop place to get a checkout, all docs, preinstalled binaries, preconfigured consoles, etc |
| 11:08.25 | *** join/#brlcad d_rossberg (~rossberg@BZ.BZFLAG.BZ) |
| 11:25.42 | *** join/#brlcad simion314 (~quassel@109.96.61.168) |
| 11:38.24 | *** join/#brlcad Al_Da_Best (~Al_Da_Bes@027e71f6.bb.sky.com) |
| 12:16.47 | *** join/#brlcad witness123 (~witness@gateway/tor-sasl/witness123) |
| 12:34.06 | CIA-128 | BRL-CAD: 03starseeker * r49843 10/brlcad/trunk/CMakeLists.txt: The build type isn't enough to tell us whether we've got a build with debugging enabled - use the debug flags variable. |
| 12:39.32 | *** join/#brlcad Al_Da_Best (~Al_Da_Bes@027e71f6.bb.sky.com) |
| 13:16.08 | brlcad | hello Al_Da_Best |
| 13:16.18 | Al_Da_Best | Hiya |
| 14:21.58 | *** join/#brlcad Neil___ (~chatzilla@117.229.123.9) |
| 16:00.56 | *** join/#brlcad harmanpreet (~chatzilla@124.253.16.13) |
| 16:07.04 | brlcad | hello harmanpreet |
| 16:09.47 | harmanpreet | hi |
| 16:11.51 | harmanpreet | I am interested in the project: Material Database, under the category: "Web Development Projects" so I would like to know where can I find more information about this project. |
| 16:12.09 | brlcad | the title to each project links to a page with more information |
| 16:12.26 | crdueck | brlcad: i have brlcad working, the version of brlcad provided by my distro compiled from source fine. but i need to be able to compile the svn version to be able to test changes correct? |
| 16:13.17 | harmanpreet | I read it. :-) Now I need more information. |
| 16:14.26 | brlcad | harmanpreet: asking a more specific question will get you more specific information |
| 16:15.12 | brlcad | the projects are just staring point ideas -- you're actually suppoosed to decide what the project is that you'd like to propose |
| 16:16.01 | brlcad | i.e., you provide the details |
| 16:16.30 | brlcad | will gladly answer any questions, though, for what we had in mind |
| 16:16.37 | brlcad | crdueck: ideally, yes |
| 16:16.54 | harmanpreet | brlcad: thank you for your quick response. May I know where the details are supposed to be submitted? |
| 16:19.20 | brlcad | crdueck: the vm disk image has a recent svn checkout that should update and compile cleanly |
| 16:21.07 | brlcad | harmanpreet: I don't understand what you're asking? are you asking where/how to submit a gsoc application officially? |
| 16:21.29 | brlcad | or are you asking where to provide details to us for discussion? |
| 16:22.46 | *** join/#brlcad merzo (~merzo@82-251-132-95.pool.ukrtel.net) |
| 16:23.52 | harmanpreet | Yes, I wish to know where can I give details to brlcad about my project idea? |
| 16:25.18 | brlcad | harmanpreet: well, you can discuss your project here interactively as long as you like |
| 16:25.40 | harmanpreet | OK, thanks :-) |
| 16:26.01 | brlcad | there's also the mailing list and our wiki, but you still need specific questions if you're going to ask for help |
| 16:27.01 | Neil___ | brlcad: Hi! I've been reading up on caching, process queue, the type of database, file handling etc. but i have my exams going on at the moment. I'll put up the details in a couple of days if that's ok. |
| 16:29.04 | brlcad | harmanpreet: be sure to read http://brlcad.org/wiki/Summer_of_Code/Application_Guidelines particularly the Do's, Don'ts, and section on writing a successful application |
| 16:30.20 | harmanpreet | oh!! thank a lot :-D |
| 16:30.23 | brlcad | Neil___: that is fine but hopefully you can post something before monday so we have time to provide feedback |
| 16:31.12 | brlcad | harmanpreet: http://brlcad.org/wiki/Summer_of_Code/Checklist <-- that too, if you haven't read our step-by-step process |
| 16:35.49 | Neil___ | brlcad: I understand. I'll try to put it up by sunday. |
| 16:53.24 | crdueck | brlcad: it looks like i fixed it, i'm not sure exactly why, but the svn version didnt have a path set for X11_X1_LIB, i ran cmake -i and provided the path myself and it seems to be compiling fine now :) |
| 16:57.57 | brlcad | crdueck: glad to hear tit |
| 16:59.00 | brlcad | just keep in mind that there's like a month between the time students are accepted and summer coding begins to sort out your build environment |
| 17:00.29 | brlcad | there's only seven days remaining for writing up a proposal, going back and forth on discussions, working on a patch, distinguishing your abilities from that of the other students, etc so just make sure you budget your time wisely ;) |
| 17:03.27 | *** join/#brlcad diljot (caa43575@gateway/web/freenode/ip.202.164.53.117) |
| 17:27.32 | *** join/#brlcad diljot (caa43575@gateway/web/freenode/ip.202.164.53.117) |
| 19:00.41 | *** join/#brlcad merzo (~merzo@21-210-200-46.pool.ukrtel.net) |
| 19:01.40 | *** join/#brlcad Neil____ (~chatzilla@117.229.101.19) |
| 19:06.46 | *** join/#brlcad witness123 (~witness@14.139.228.211) |
| 19:45.51 | *** join/#brlcad atneik (~atneik@59.178.148.247) |
| 19:46.03 | *** join/#brlcad atneik_ (~atneik@59.178.148.247) |
| 19:48.54 | *** part/#brlcad atneik_ (~atneik@59.178.148.247) |
| 19:48.54 | *** part/#brlcad atneik (~atneik@59.178.148.247) |
| 20:29.20 | *** join/#brlcad witness123 (~witness@14.139.228.211) |
| 20:48.37 | CIA-128 | BRL-CAD: 03tbrowder2 * r49844 10/brlcad/trunk/src/vdeck/vdeck.1: correct typo |
| 21:08.35 | *** join/#brlcad witness123 (~witness@14.139.228.211) |
| 21:13.53 | *** join/#brlcad ksuzee (2e9551a6@gateway/web/freenode/ip.46.149.81.166) |
| 21:24.56 | CIA-128 | BRL-CAD: 03starseeker * r49845 10/brlcad/trunk/ (3 files in 2 dirs): Do a bit of rework on the BRLCAD_ADDDATA target - instead of WIN32 vs everything else, actually test whether the symlink mechanism works and base the macro behavior on those results. |
| 21:30.04 | CIA-128 | BRL-CAD: 03starseeker * r49846 10/brlcad/trunk/misc/CMake/BRLCAD_Targets.cmake: fix comment |
| 21:57.32 | *** join/#brlcad witness123 (~witness@14.139.228.211) |
| 22:15.35 | brlcad | ~ask |
| 22:15.35 | ibot | Questions 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. |
| 22:15.50 | brlcad | mm. |
| 22:15.52 | brlcad | need food. |
| 22:24.19 | crdueck | for my first patch i've decided to implement a volume function for a lin_pipe, since it's a simple solid of revolution. i've been browsing the source code to try to get familiar with how things work but i just need a little more context so i can get started. for example, how is a pipe "object" passed to functions? i see many functions using rt_db_internal as a parameter but i'm a little unclear on what it actually is. my guess would be to use the l |
| 22:37.08 | ``Erik | pipe is an aggregate primitive and might be a big ugly as an example, something like rcc or ell might be a better first primitive to help understand how the pieces fit together |
| 22:39.16 | ``Erik | if you are stuck on doing it for pipe, the lin_pipe segment is held in a BU_LIST container (using the magic to switch), which is held by the pipe primitive |
| 22:40.59 | ``Erik | (all interaction with lin_pep and bend_pipe is through the rt_pipe_*() set) |
| 22:44.18 | crdueck | when you say rcc, did you mean rec? |
| 22:44.51 | crdueck | oh nvm, right circular cylinder |
| 22:49.48 | crdueck | where can i find the rcc primitive? |
| 22:50.20 | ``Erik | um, tgc (rcc is a tgc with certain constraints), sorry :) |
| 22:55.56 | crdueck | looking at sph, it seems like it would also be a suitable first primitive no? |
| 23:00.43 | crdueck | again, it seems like the sph_specific struct holds all the relevant info for a sphere. so how does one pass a sph to a function? |
| 23:39.07 | *** join/#brlcad stas (~stas@188.24.36.145) |
| 23:57.23 | CIA-128 | BRL-CAD: 03tbrowder2 * r49847 10/brlcad/trunk/src/vdeck/vdeck.c: ws, style |