| 00:41.58 | *** join/#brlcad dli (n=dli@adsl-75-33-240-234.dsl.chcgil.sbcglobal.net) |
| 01:25.20 | *** join/#brlcad haywood_giablomi (n=John_K@c-71-56-97-21.hsd1.ga.comcast.net) |
| 01:25.45 | *** join/#brlcad sebastienbailard (n=sbailard@CPE000d88ba7a3b-CM001371169626.cpe.net.cable.rogers.com) |
| 01:27.07 | sebastienbailard | Hello, anyone here today? |
| 01:28.31 | *** join/#brlcad IriX64 (n=IriX64@toronto-HSE-ppp4304415.sympatico.ca) |
| 01:28.39 | sebastienbailard | Hey. |
| 01:28.48 | IriX64 | hey |
| 01:28.52 | IriX64 | sups? |
| 01:29.00 | sebastienbailard | Don't know. Just came on. |
| 01:29.11 | sebastienbailard | I think everyone is busy. |
| 01:29.23 | IriX64 | usually you ask for help here.:) |
| 01:29.29 | IriX64 | and they reply given time. |
| 01:29.30 | sebastienbailard | Ah. I like to ask to ask, before I ask. |
| 01:29.46 | sebastienbailard | I've got a question about brlcad's UI. |
| 01:30.06 | IriX64 | i only type with two fingers, sorry im so slow. |
| 01:30.11 | sebastienbailard | Are people planning on integrating qcad for 2D input? |
| 01:30.18 | IriX64 | ask someone may answer. |
| 01:30.26 | IriX64 | whats qcad? |
| 01:30.41 | sebastienbailard | It's a really good linux 2d cad program. |
| 01:31.04 | sebastienbailard | http://www.nabble.com/-brlcad---Developers--QCAD-integration-inside-BRL-CAD--t2001250.html |
| 01:31.05 | IriX64 | brlcad is misnamed it should be brlmodelling. |
| 01:31.26 | sebastienbailard | People use it to design machinery, right? |
| 01:31.47 | IriX64 | model things, weapons systems and anything else. |
| 01:32.06 | sebastienbailard | So it's a modeler, it's not for design? |
| 01:32.12 | sebastienbailard | Huh. |
| 01:32.21 | IriX64 | design *to model. |
| 01:32.29 | sebastienbailard | Ah. |
| 01:32.48 | sebastienbailard | It's just there aren't any decent gpl 3D CAD programs. |
| 01:33.04 | IriX64 | gpl? |
| 01:33.20 | sebastienbailard | GNU public license. Like linux, or emacs. |
| 01:33.38 | IriX64 | so you want the cadillac for free? |
| 01:33.51 | sebastienbailard | Pretty much, yes. |
| 01:34.28 | IriX64 | well you have it in BrlCad :P |
| 01:34.32 | sebastienbailard | The way it is right now, people at the mit-type fab labs use blender to design the stuff they fabricate. |
| 01:34.58 | sebastienbailard | Right, but BrlCad's user interface is a little intimidating. |
| 01:35.06 | IriX64 | how? |
| 01:35.30 | sebastienbailard | Let's put it this way. I asked three experienced people what they thought of brl cad |
| 01:35.39 | sebastienbailard | should I learn to use it, and so on. |
| 01:35.46 | IriX64 | and? |
| 01:35.49 | sebastienbailard | Got three 'no's. |
| 01:36.03 | IriX64 | so why are you here? |
| 01:36.16 | sebastienbailard | I was coming to ask about the qcad stuff. |
| 01:36.39 | IriX64 | ill let someone who knows what qcad is answer that. |
| 01:37.05 | sebastienbailard | I figured if people had a good way to do 2D input using brl cad, they'd be more motivated to learn the 3d stuff. |
| 01:37.34 | IriX64 | why work in 2d when you can work in 3d.? |
| 01:37.52 | sebastienbailard | 1) Sometimes all people need is 2d. |
| 01:38.03 | IriX64 | Those experienced people couldn't have been experienced in brlcad. |
| 01:38.08 | sebastienbailard | 2) People can figure out 2d easier than 3d. |
| 01:38.34 | sebastienbailard | Ah yes. My point there was - they had all looked at brlcad, and were scared off. |
| 01:38.54 | IriX64 | if the tool doesn't suit you buy another tool why should brlcad regress? |
| 01:39.24 | sebastienbailard | I don't mean to be rude. My point is that brlcad might need a bit of candy-coating. |
| 01:39.46 | sebastienbailard | Assuming you guys want more people to use brlcad. |
| 01:39.47 | IriX64 | for those too lazy to take the tutorials? |
| 01:40.38 | sebastienbailard | Ah. I didn't mean to offend. |
| 01:41.58 | sebastienbailard | Is anyone else there? |
| 01:44.40 | sebastienbailard | Is anyone using brlcad to generate g-code then? |
| 06:33.46 | *** join/#brlcad clock_ (i=clock@84-72-93-198.dclient.hispeed.ch) |
| 07:44.18 | *** join/#brlcad SWAT (n=SWAT@a80-100-68-159.adsl.xs4all.nl) |
| 07:45.18 | SWAT | How do I start brlcad, do I need to compile it and where is the README file? |
| 08:06.18 | *** join/#brlcad clock_ (n=clock@zux221-122-143.adsl.green.ch) |
| 11:34.36 | brlcad | SWAT: brl-cad is a suite of many many tools .. what you "start" depends on what you want to do |
| 11:35.03 | brlcad | most start with the solid modeler mged, which has a gui and associated documentation |
| 11:36.23 | brlcad | SWAT: the readme is in both the source and binary distributions, in the binary distribution, it's embedded in the documentation directory in usr/brlcad/share/brlcad/7.8.2/README |
| 11:37.42 | brlcad | in the source distribution, the README is right there in the top level directory |
| 11:37.49 | brlcad | you can view it on-line here: http://brlcad.cvs.sourceforge.net/*checkout*/brlcad/brlcad/README |
| 11:44.54 | SWAT | thanks |
| 13:07.55 | ``Erik | in addition to the package inclusion... |
| 13:26.07 | SWAT | well, it 'normal' to put the README in the top dir |
| 13:26.18 | SWAT | just like the INSTALL (if applicable) |
| 13:40.45 | *** join/#brlcad rossberg (n=rossberg@bz.bzflag.bz) |
| 13:42.29 | CIA-9 | BRL-CAD: 03d_rossberg * 10brlcad/misc/win32-msvc/Dll/TclDummies.c: Tcl_SetErrno dummy function |
| 13:44.11 | CIA-9 | BRL-CAD: 03d_rossberg * 10brlcad/src/libbu/libbu.dsp: added bu_fgets.c |
| 13:46.06 | CIA-9 | BRL-CAD: 03d_rossberg * 10brlcad/src/librt/librt.dsp: added transform.c |
| 13:47.58 | brlcad | hello rossberg, how goes things? |
| 13:51.19 | *** join/#brlcad clock_ (n=clock@zux221-122-143.adsl.green.ch) |
| 13:58.56 | rossberg | brlcad: good, today is my first working day after vacation |
| 14:06.52 | *** join/#brlcad SWPadnos (n=Me@dsl245.esjtvtli.sover.net) |
| 14:28.02 | brlcad | rossberg: ahh, nice |
| 15:02.57 | *** part/#brlcad SWAT (n=SWAT@a80-100-68-159.adsl.xs4all.nl) |
| 16:56.40 | ``Erik | 2/det |
| 17:14.16 | CIA-9 | BRL-CAD: 03erikgreenwald * 10brlcad/ (include/rtgeom.h src/librt/g_metaball.c): Added "method" to metaball struct. Changed some language. |
| 17:30.19 | CIA-9 | BRL-CAD: 03erikgreenwald * 10brlcad/src/librt/g_metaball.c: fix a crash in the exporter... |
| 17:33.01 | CIA-9 | BRL-CAD: 03erikgreenwald * 10brlcad/src/mged/ (chgmodel.c typein.c): Added "method" to metaball struct. Changed some language. |
| 17:47.49 | CIA-9 | BRL-CAD: 03erikgreenwald * 10brlcad/src/mged/ (edsol.c sedit.h): Added "method" to metaball struct. Begin stubbing out ECMD stuff. Changed some language. |
| 18:02.50 | brlcad | "method"? |
| 18:03.02 | ``Erik | yeah |
| 18:03.27 | ``Erik | selectable formulas for evaluating field strength at a given point |
| 18:03.42 | brlcad | ahh |
| 18:03.54 | brlcad | you coulda just said that :) |
| 18:04.24 | brlcad | i'm thinking.. method for evaluation? method for rendering? c++ methods? method for deleting? method for creating? ?? :) |
| 18:04.51 | ``Erik | uh, ok, sed -i.bak 's/method/selectableformulaforevaluatingfieldstrengthatagivenpoint/g' `find . -type f` && cvs commit -m 'rephrasing' |
| 18:04.53 | ``Erik | :D |
| 18:04.58 | brlcad | heh |
| 18:05.47 | ``Erik | <-- ass |
| 18:06.02 | ``Erik | my run-on sentences rock |
| 18:11.43 | *** join/#brlcad ibot (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 18:11.43 | *** topic/#brlcad is http://brlcad.org/ || BRL-CAD is an open source solid modeling software suite || Developers needed! Read the HACKING file for details on getting involved |
| 19:34.42 | *** join/#brlcad clock_ (i=clock@84-72-90-174.dclient.hispeed.ch) |
| 20:07.06 | CIA-9 | BRL-CAD: 03bob1961 * 10brlcad/src/libpkg/pkg.c: It looks like the intent was to struct copy a "struct in_addr" in the _pkg_permserver_impl function. |
| 20:09.23 | CIA-9 | BRL-CAD: 03bob1961 * 10brlcad/src/tclscripts/archer/Archer.tcl: Added method to purge edit history. More mods to accomodate ogl being replaced by wgl. |
| 20:13.22 | CIA-9 | BRL-CAD: 03bob1961 * 10brlcad/src/mged/ (cmd.c ged.c): Mods to get things to natively compile on Windows. |
| 20:25.54 | *** part/#brlcad IriX64 (n=IriX64@toronto-HSE-ppp4304415.sympatico.ca) |
| 20:26.07 | *** join/#brlcad IriX64 (n=IriX64@toronto-HSE-ppp4304415.sympatico.ca) |
| 20:55.47 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/src/libpkg/pkg.c: minor, consolidate return |
| 21:11.40 | *** join/#brlcad digitalfredy (n=digitalf@200.71.62.161) |
| 21:15.36 | *** join/#brlcad clock_ (i=clock@84-72-90-174.dclient.hispeed.ch) |
| 21:27.57 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/src/libpkg/pkg.c: |
| 21:27.57 | CIA-9 | BRL-CAD: get rid of the _WIN32 sections for recv/send instead of read/write .. this |
| 21:27.58 | CIA-9 | BRL-CAD: really shouldn't be necessary if the compatibility macros in the config_win.h |
| 21:27.58 | CIA-9 | BRL-CAD: header are specified correctly, but for now keep it as it is just without all |
| 21:27.58 | CIA-9 | BRL-CAD: the repetition by using macros that call the corresponding routine. oh yeah, |
| 21:28.00 | CIA-9 | BRL-CAD: and M-x indent-region ws too. |
| 21:33.36 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/src/libpkg/pkg.c: forgot to enclose macro params in parens |
| 21:52.02 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/src/mged/cmd.c: no need to forward declare it if we just move the definition up |
| 21:54.27 | *** join/#brlcad IriX64_ (n=Who@toronto-HSE-ppp4304415.sympatico.ca) |
| 22:28.30 | brlcad | does anyone know if there's an equivalent call for sync() on win32? not finding anything of use in docs/search |
| 22:40.55 | ``Erik | grep headers for -i 'flush'? |
| 22:42.35 | brlcad | starting to find stuff, found what cygwin does |
| 22:42.52 | ``Erik | (does windows even buffer file descriptors? hahaha) |
| 22:43.02 | brlcad | or at least a proposal for dealing with it at one point, which basically involves iterating over the fixed drives and callling FlushFileBuffers |
| 22:43.42 | ``Erik | coo' |
| 22:48.18 | brlcad | or I could just remove our four bastard usages of sync() and replace em with fsync() |
| 22:51.04 | brlcad | though a bu_sync() would still be nice to have for bu_bomb() situations |
| 22:54.15 | ``Erik | uhhhhh |
| 22:54.27 | ``Erik | if sync() 'fails', your os is hosed and crashes... |
| 22:54.28 | ``Erik | :) |
| 22:54.47 | ``Erik | or locks |
| 22:54.57 | brlcad | "not my problem?" :) |
| 22:55.43 | ``Erik | well, ok |
| 22:55.45 | ``Erik | look at it this way |
| 22:55.50 | ``Erik | void sync(void); |
| 22:55.57 | ``Erik | how do you figure out if that failed? :) |
| 22:56.22 | brlcad | the standard doesn't actually require sync() to do anything either |
| 22:56.28 | ``Erik | { ask syncer to do it's thing; block until syncer has done it's thing. } |
| 22:56.37 | brlcad | which is why i'm just going to change it all to fsync which does |
| 22:56.45 | ``Erik | that's probably better |
| 22:57.05 | ``Erik | on a massively multi-user system, sync() could be a brutal pig of a function call, fsync is targetted |
| 22:58.01 | ``Erik | <-- tells the linux kernel on that altix to schedule sync once every 42 days to prove his point :) |
| 22:58.40 | brlcad | it can be a pig, but it should still return within a few seconds at worst |
| 22:58.57 | brlcad | just as can be calling the sync command |
| 22:59.23 | brlcad | kermit still issues sync after almost every command he types :) |
| 22:59.31 | ``Erik | on large fs's with brutal activity, I've watched sync take disturbing amounts of time |
| 22:59.34 | brlcad | due to old systems not doing it for you |
| 22:59.37 | ``Erik | <-- does that a lot, too |
| 22:59.41 | ``Erik | oh |
| 22:59.50 | ``Erik | and if you have a flakey disk, sync can take a *LONG* time |
| 23:00.01 | ``Erik | like a non-responsive nfs mount or something |
| 23:00.43 | ``Erik | heh, or a raid array in the process of crapping itself. :( |
| 23:02.09 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/src/librt/db_open.c: fsync the file before closing it just to be safe |
| 23:07.09 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/configure.ac: check for fsync(), bsd func that some older systems don't have |
| 23:16.07 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/src/mged/cmd.c: remove the sync() call. instead just fsync() the currently open geometry database file. doxygenify the comments while we're in here. |
| 23:21.13 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/src/librt/db_open.c: rework db_sync() so that it always flushs and tries fsync() if available and only sync() as last resort. |
| 23:26.40 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/src/librt/db_open.c: sync via db_sync in db_close instead of manually duplicating what is conceptually db_sync's job |
| 23:27.22 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/src/mged/cmd.c: sync via db_sync instead of fsync'ing directly. |
| 23:28.08 | brlcad | i'm just working down the list |
| 23:28.29 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/src/mged/dm_old/dm-mer.c: no protection for you |
| 23:29.43 | ``Erik | hm, so you got src/proc-db/pix2g.c and src/util/op-bw.c, cool |
| 23:29.45 | ``Erik | :) |
| 23:30.53 | brlcad | yep |
| 23:31.17 | brlcad | hence the alphabetical order of the fixes ;) |
| 23:31.22 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/src/proc-db/pix2g.c: stupid sync().. who wrote this? .. changed to db_sync() |
| 23:33.03 | brlcad | bleh.. op-bw is vax-specific |
| 23:33.23 | brlcad | i somehow doubt anyone has an optronics |
| 23:34.04 | brlcad | hmm |
| 23:34.23 | brlcad | and if they do, the sync() might actually be needed for that one, /me leaves it alone |
| 23:42.30 | *** join/#brlcad mlombardo (n=anonymou@6532233hfc181.tampabay.res.rr.com) |
| 23:45.23 | CIA-9 | BRL-CAD: 03brlcad * 10brlcad/ (TODO src/util/op-bw.c): op-bw had a long run but seeing as it is vax-specific, it is being marked obsolete. the sources are marked for removal at the next minor update (probably by 7.10). |