| 00:56.19 | ``Erik | new breakage in opennurbs, huzzah |
| 01:01.22 | CIA-38 | BRL-CAD: 03erikgreenwald * r37011 10/brlcad/trunk/src/other/openNURBS/opennurbs_memory.c: try not to reference things that don't exist |
| 02:19.00 | starseeker | blinks - wonder how that compiled on my Mac then... |
| 02:22.15 | ``Erik | got me, mebbe a timestamp issue left the old object file? |
| 02:22.44 | starseeker | looks like that should be p != p, rather than memblock... wonder why they renamed it |
| 02:22.47 | starseeker | ah well |
| 02:23.06 | ``Erik | either way, it's a punt |
| 02:23.34 | starseeker | hmm? I thought Sean threw that into the old code base |
| 02:23.42 | ``Erik | probably an 'unused variable' hack |
| 02:24.00 | starseeker | will it cause you breakage either way? |
| 02:24.12 | starseeker | yeah, it was warning quellage |
| 02:25.09 | ``Erik | if it were return p != p; it woulda been fine... did you upgade the existing code with a patch? |
| 02:25.30 | starseeker | nah, just didn't check carefully enough whether that variable name had changed |
| 02:25.35 | starseeker | my bad |
| 02:26.15 | ``Erik | okie, allz I saw was an error about an undefined variable, didn't turn on strict flags |
| 02:26.18 | CIA-38 | BRL-CAD: 03starseeker * r37012 10/brlcad/trunk/src/other/openNURBS/ (BRL-CAD_changes.txt opennurbs_memory.c): Er, whoops - thanks Erik. p != p might work better when the variable name changed to p... |
| 02:26.38 | ``Erik | kinda care more about mged getting the BadAlloc... rt -F/dev/Xl works fine |
| 02:26.41 | starseeker | ``Erik: actually a lot of our changes disappeared |
| 02:27.00 | starseeker | was more concerned about minimizing our changes to openNURBS without breaking the raytracer |
| 02:27.06 | starseeker | yeah |
| 02:27.20 | starseeker | pulls the libdm logs for the last few weeks... |
| 02:27.35 | starseeker | mged is kinda foobared at the moment |
| 02:28.12 | ``Erik | actually, at the moment, I care more about reading a webcomic and being a cat matress, but *shrug* :) |
| 02:28.19 | starseeker | heh |
| 02:28.31 | starseeker | hopes like hell it wasn't his libdm changes |
| 02:29.04 | ``Erik | it's been doing it for a bit, I figured one of ya'll messing with that tsuff woulda noticed it, all my work has been lower level |
| 02:29.08 | starseeker | oh, it might be r36844 |
| 02:29.23 | ``Erik | g2asc/asc2g, adrt, and librt crud |
| 02:29.27 | starseeker | come to think of it I remember something about that |
| 02:29.49 | ``Erik | which is about as close to mged as I care to mess... still have flashbacks about edsol.c |
| 03:01.09 | CIA-38 | BRL-CAD: 03starseeker * r37013 10/brlcad/trunk/src/libdm/dm-X.c: Back out some dm-X changes causing breakage on Mac and Linux platforms - trying to preserve some of the more recent changes to drawLine3d functions. |
| 03:12.05 | ``Erik | whuddya know, mged starts up now :D |
| 03:31.00 | CIA-38 | BRL-CAD: 03starseeker * r37014 10/brlcad/trunk/src/librt/ (3 files in 3 dirs): |
| 03:31.00 | CIA-38 | BRL-CAD: Compile gotchas on gentoo - FALSE returning as not defined in a couple files - |
| 03:31.00 | CIA-38 | BRL-CAD: looks like this was coming in from opennurbs and is gone with the latest update |
| 03:31.00 | CIA-38 | BRL-CAD: - just go with 0; also include assert.h in brep.cpp since it's not getting |
| 03:31.00 | CIA-38 | BRL-CAD: pulled in through opennurbs any more. |
| 03:58.59 | CIA-38 | BRL-CAD: 03starseeker * r37015 10/brlcad/trunk/src/proc-db/ (5 files): FALSE and assert.h tweaks for proc-db |
| 04:04.46 | *** join/#brlcad talcite (n=matthew@69-196-132-129.dsl.teksavvy.com) |
| 04:27.37 | *** join/#brlcad PrezKennedy (i=Matthew@whitecalf.net) |
| 04:58.53 | *** join/#brlcad Nohla (n=jesica@201.255.220.253) |
| 05:25.43 | *** join/#brlcad talcite_ (n=matthew@69-196-180-185.dsl.teksavvy.com) |
| 06:08.22 | starseeker | blinks - well, rtgl -> ogl -> rtgl and rtgl -> X -> rtgl both just succeeded on gentoo |
| 06:11.07 | starseeker | and we've got a nurbs raytrace |
| 06:11.30 | starseeker | aaaand csgbrep barfs |
| 10:06.12 | *** join/#brlcad docelic (n=docelic@78-2-80-119.adsl.net.t-com.hr) |
| 11:24.09 | *** join/#brlcad csanyipal (n=csanyipa@132-142-85-95.dynamic.stcable.net) |
| 11:24.15 | csanyipal | Hi, |
| 11:25.19 | csanyipal | When I install brlcad from SVN must I delete first the /usr/brlcad directory? |
| 11:25.31 | csanyipal | the system is Debian GNU/Linux Lenny. |
| 11:42.43 | csanyipal | When run 'make' after './configure --enable-optimized', |
| 11:42.50 | csanyipal | I get error messages: |
| 11:43.01 | csanyipal | make[2]: *** [region_end.lo] Error 1 |
| 11:43.01 | csanyipal | make[2]: Leaving directory `/debian-adatok-1/CAD/BRL-CAD/SVN/brlcad/src/libgcv' |
| 11:43.01 | csanyipal | make[1]: *** [all-recursive] Error 1 |
| 11:43.01 | csanyipal | make[1]: Leaving directory `/debian-adatok-1/CAD/BRL-CAD/SVN/brlcad/src' |
| 11:43.02 | csanyipal | make: *** [all-recursive] Error 1 |
| 11:43.11 | csanyipal | Any solutions? |
| 11:44.30 | docelic | csanyipal, you didn't paste the actuall errors |
| 11:44.48 | csanyipal | docelic: ? |
| 11:45.07 | csanyipal | region_end.c: In function âgcv_region_endâ: |
| 11:45.07 | csanyipal | region_end.c:38: error: argument âcurtreeâ might be clobbered by âlongjmpâ or âvforkâ |
| 11:45.28 | docelic | that is the real error. |
| 11:45.52 | csanyipal | docelic: thanks. How can I resolve it? |
| 11:47.54 | docelic | don't know about this one, see Google for "might be clobbered by". Also, (my guess), I think this is usually a warning, which is an error here because of some compile options which say to treat all warnings as errors. So if you can't solve it, maybe you can tell ./configure to not treat warnings as errors, and it'll go forwarsd |
| 11:48.29 | csanyipal | docelic: thanks! :) |
| 11:48.31 | *** join/#brlcad mafm (n=mafm@162.Red-81-32-97.dynamicIP.rima-tde.net) |
| 13:15.18 | *** join/#brlcad docelic_ (n=docelic@78-2-80-2.adsl.net.t-com.hr) |
| 14:22.35 | ``Erik | that's from strict flags, turn'em off and it'll compile |
| 14:38.40 | csanyipal | ``Erik: how can I turn'em off? |
| 14:41.38 | csanyipal | Can I do this with --disable-warnings ? |
| 14:41.45 | starseeker | I think it's --disable-strict |
| 14:41.58 | starseeker | although brlcad might have changed it - give that a try |
| 14:42.13 | csanyipal | starseeker: ok. |
| 14:47.08 | CIA-38 | BRL-CAD: 03erikgreenwald * r37016 10/brlcad/trunk/src/libgcv/Makefile.am: forcibly disable strict flags until the jump clobber issue can be sorted out. |
| 14:50.11 | csanyipal | CIA-38: does this line: 'r37016...' mean that I can checkout again SVN and try to compile?? |
| 14:51.11 | starseeker | yes |
| 14:51.12 | ``Erik | CIA is a bot that reports commits |
| 14:51.24 | csanyipal | ok |
| 15:30.45 | csanyipal | I run make successfully, now I run make benchmark. |
| 15:32.25 | csanyipal | make benchmark is comlpete now. |
| 15:32.59 | csanyipal | I'm send now the benchmakr results to benchmark@brlcad.org. |
| 15:33.58 | ``Erik | you did use --enable-optimized on yoru configure, correct? |
| 15:34.35 | csanyipal | I did the following: |
| 15:34.37 | csanyipal | ./configure --enable-optimized |
| 15:34.42 | csanyipal | make -j2 |
| 15:34.48 | csanyipal | make benchmark |
| 15:34.52 | csanyipal | so far. |
| 15:34.55 | ``Erik | ok, good, then you have good numbers |
| 15:35.15 | csanyipal | yes: Benchmark results indicate an approximate VGR performance metric of 4614 |
| 15:35.24 | csanyipal | Logarithmic VGR metric is 3,66 (natural logarithm is 8,44) |
| 15:37.33 | csanyipal | I have send the benchmark results to benchmark@brlcad.org. |
| 15:38.21 | csanyipal | Before sudo make install should I delete the /usr/brlcad directory? |
| 15:39.43 | csanyipal | or it will be overwritten during installation? |
| 15:40.00 | csanyipal | I have an old brlcad installation here. |
| 15:40.43 | csanyipal | I think it's better to delete this directory, isn't? |
| 15:51.31 | csanyipal | I have installed now brlcad. Just to find out how to setup PATH so I can run on xterm (urxvt) the 'mged' command. |
| 15:52.02 | csanyipal | I'm using Window Maker window manager with gdm. |
| 15:52.24 | csanyipal | thank you for help! |
| 15:53.14 | *** join/#brlcad talcite_ (n=matthew@69-196-180-185.dsl.teksavvy.com) |
| 16:20.26 | *** join/#brlcad csanyipal (n=csanyipa@132-142-85-95.dynamic.stcable.net) |
| 16:27.42 | *** join/#brlcad csanyipal (n=csanyipa@132-142-85-95.dynamic.stcable.net) |
| 17:41.31 | brlcad | volks, can disable strict with: make STRICT_FLAGS= |
| 17:41.39 | brlcad | don't have to reconfigure |
| 17:41.57 | brlcad | otehrwise, reconfigure with --disable-strict should shut it off too |
| 17:42.27 | csanyipal | oh |
| 17:43.59 | brlcad | csanyipal: no worries for you, you have a build :) |
| 17:44.14 | brlcad | turning it off in the Makefile.am works too, which is what that commit did |
| 17:44.14 | csanyipal | of course.. :) |
| 17:44.52 | brlcad | you don't have to set path if you don't want to, just have to write the whole path: /usr/brlcad/bin/mged |
| 17:45.15 | brlcad | otherwise for bash: export PATH=/usr/brlcad/bin:$PATH |
| 17:45.24 | csanyipal | i have set ip up now in .bashrc |
| 17:45.30 | brlcad | nods |
| 17:45.33 | csanyipal | set it up |
| 17:45.53 | brlcad | merry christmas and thanks :) |
| 17:46.05 | csanyipal | ok :) |
| 17:46.52 | csanyipal | merry christmas for all of you here too! |
| 19:28.24 | *** join/#brlcad csanyipal (n=csanyipa@132-142-85-95.dynamic.stcable.net) |
| 20:13.05 | starseeker | hmm. csgbrep works on the mac |
| 20:13.29 | starseeker | boy gentoo is unforgiving when it comes to this stuff... |
| 20:13.34 | starseeker | which is a good thing I guess |
| 20:16.11 | *** join/#brlcad csanyipal (n=csanyipa@132-142-85-95.dynamic.stcable.net) |
| 20:27.50 | yukonbob | Merry [almost] Christmas, #brlcad. |
| 20:30.19 | starseeker | brlcad: wonder if it would be worth checking if Archer works with the itcl 4.0b3 code? looks like that version of itcl has been crafted to play nice with the new TclOO stuff |
| 20:35.04 | yukonbob | starseeker: iirc, itcl-ng (i.e. 4.x) is _not_ really ready. I'd have to surf around to find reference... |
| 20:35.32 | starseeker | http://wiki.tcl.tk/19873 is what I was looking at... |
| 20:36.42 | yukonbob | I'm reviewing same, atm... |
| 20:40.12 | yukonbob | I forget where I saw it (avoid itcl-ng), but it stuck in my mind. The choice of BRL-CAD to dig into Tcl8.5b, and the itcl-ng, etc., caused a lot of churn on my system, which I still haven't actually recovered from. Picking proper itcl was on my List of Things, and like I said, iirc (I should have kept better notes) itcl-ng is -not- the way to go (atm). |
| 20:41.26 | starseeker | what churn did it cause? |
| 20:42.15 | starseeker | we're going to want tcl/tk 8.6 - among other things, their Aqua support on the mac will be targeting Cocoa itself |
| 20:42.31 | starseeker | dunno if itcl 3.x can work on 8.6 |
| 20:51.47 | yukonbob | starseeker: at the time (a while ago), brl-cad required a beta tcl/tk; on NetBSD, I had setup the brl-cad distribution to build only the minimal requirements and use pre-installed pkgs for rest... |
| 20:52.08 | yukonbob | (i.e. use pre-existing tcl, tk, utah, etc., etc) |
| 20:52.35 | yukonbob | not having the beta tcl/tk threw that into a tailspin, until I got everything working w/ 8.4 again... |
| 20:53.13 | yukonbob | anyway, long story short, I didn't stay caught-up, and now don't have a working brl-cad using in-system pkgs, which is the way I'm trying to pursue the install... |
| 20:53.38 | yukonbob | I haven't had time to dedicate to sovling as I did in past either :P |
| 20:54.02 | yukonbob | 8.6 is still a way out, from my understanding talking w/ coreteam. |
| 20:57.10 | starseeker | ah, yes - on-system packaging is a problem |
| 20:57.40 | yukonbob | I had it solved, but that was some time ago. |
| 20:57.50 | starseeker | or can be rather |
| 20:57.57 | yukonbob | It's my holy grail to reattain that |
| 20:58.04 | yukonbob | then try to maintain. |
| 20:58.32 | starseeker | archer is now using the new ttk widgets, so the minimum system tk will need to support those |
| 20:58.59 | yukonbob | 8.5 == ttk, and indeed, I believe 8.4 has ttk as a package.... |
| 20:59.10 | yukonbob | "tile" |
| 20:59.29 | starseeker | yes, but I can't remember if tile uses the ttk prefix |
| 20:59.39 | yukonbob | neither |
| 21:00.08 | starseeker | the ttk widgets are a considerable improvement in the visual appearance department |
| 21:00.09 | yukonbob | tcl/tk == the stuff I'm pretty familiar with -- indeed, it was the vector that led me to brl-cad ;) |
| 21:00.17 | yukonbob | starseeker: yup |
| 21:00.44 | yukonbob | and native OS X will be really nice, too. 8.6 is certainly going to bring even more goodness to tcl |
| 21:00.49 | starseeker | must become familiar - the time has come for a working Tk display manager and frame buffer |
| 21:01.15 | starseeker | yukonbob: do you happen to be familiar with togl? |
| 21:02.32 | yukonbob | starseeker: short answer: no |
| 21:02.55 | starseeker | bemusedly watches the tk display manager crash in X11 Tk on the mac where it (sort) worked under Aqua |
| 21:04.20 | starseeker | yukonbob: I have a tendency to wander down the "ooo this is an interesting Tcl/Tk library" path |
| 21:05.04 | yukonbob | starseeker: there are a fscking tonne of them out there... |
| 21:05.18 | starseeker | yukonbob: my current "ooo shiny" fixations are on togl and RamDebugger |
| 21:05.19 | yukonbob | it's an interesting path, and I can relate to you in that regard ;) |
| 21:05.36 | yukonbob | ?ramdebugger |
| 21:05.51 | starseeker | http://www.compassis.com/ramdebugger |
| 21:07.16 | yukonbob | http://wiki.tcl.tk/17771 |
| 21:08.01 | starseeker | I've yet to get and BRL-CAD running successfully in the same bwish session, but friendly Tcl debugging would be nice... |
| 21:08.12 | starseeker | ah yes, tcl3d :-) |
| 21:08.23 | starseeker | they even have an osg binding, iirc |
| 21:08.39 | yukonbob | ramdebugger looks neat. |
| 21:09.09 | starseeker | main concern I had with tcl3d is how portable it is |
| 21:09.12 | yukonbob | usually uses some kind of introspection, [puts], and gdb, all in XEmacs |
| 21:09.28 | starseeker | I think it tcl3d sucks in togl, actually... |
| 21:09.43 | yukonbob | re: portability: /me nods -- no idea. |
| 21:10.14 | starseeker | yukonbob: my ideal would be to have RamDebugger fire up on an Archer Tcl/Tk error, politely pointing me to the offending source code line :-0 |
| 21:10.41 | starseeker | dunno if it's any good, but non-commercial solutions in that space seem to be rather thin on the ground |
| 21:12.17 | starseeker | hmm - tcl3d sucks in gl2ps as well |
| 21:12.21 | starseeker | interesting |
| 21:12.37 | yukonbob | I guess it depends on what kind of errors are tripping you up, but my experience has led me to [puts], gdb, xemacs |
| 21:13.03 | yukonbob | that, and familiarity w/ Tcl and good practices, which only come w/ use. |
| 21:13.18 | starseeker | ah, that's what made me wary of tcl3d - their swig files |
| 21:13.30 | yukonbob | you don't like swig? |
| 21:13.55 | starseeker | wary of depending on it for key functionality - adds a core dependency on a translation tool |
| 21:14.28 | starseeker | might be the best/only way to do some things though |
| 21:15.03 | starseeker | yukonbob: with mged though, we have the option of talking directly to C apis insteadl of tcl scripting |
| 21:15.44 | starseeker | Archer is tcl scripting at the moment on top of wish, classic Tcl/Tk app - we're planning to make it run on top of its own C core the way mged does |
| 21:15.54 | yukonbob | nods |
| 21:16.58 | yukonbob | I'll try to make another push to get brl-cad running on my system (7.16.2) and start helping out again. |
| 21:17.05 | starseeker | obviously you can do tcl scripts in MGED as well - there are many examples - but it lets us build the core levels at the C level (see libdm and libfb :-) |
| 21:17.20 | starseeker | yukonbob: cool! sounds great |
| 21:17.30 | yukonbob | used tcl extensively with mged. |
| 21:17.31 | starseeker | is the stable Tcl/Tk up to 8.5 now? |
| 21:17.36 | yukonbob | yup |
| 21:17.47 | starseeker | that should make life simpler |
| 21:18.31 | yukonbob | 8.6 Real Soon Now, but like I said, the understanding is it's a way in the future still... for a while, they were hoping to have it out of beta before Tcl Conference in Oregon in Sept; obviously didn't happen, and it's being pushed further and further out... |
| 21:18.36 | yukonbob | which I actually don't mind too much. |
| 21:19.14 | starseeker | as far as I know there are no "must have today!" features in 8.6 for BRL-CAD the way there were for 8.5 with ttk |
| 21:19.35 | yukonbob | starseeker: re: simple life --- indeed. incr tcl/tk on my NetBSD were (are) an issue that I need to solve, and then push through the ./configure and getting brl-cad to recognize that stuff I'm offering to it so it doesn't build own utah, tcl, tk, etc., etc. |
| 21:20.39 | yukonbob | starseeker: nice -- that'll help me push if I know it's results will last a while before having to rework for 8.6 ;) |
| 21:21.46 | yukonbob | I'll see what i can muster, and start bugging #brlcad for issues I run into to get this up/running. |
| 21:21.53 | yukonbob | misses modelling :P |
| 21:22.14 | starseeker | <snort> given how long it looks to be before 8.6 is out, we'd better hope there are no necessary features... |
| 21:22.35 | yukonbob | "necessary" is a state of mind |
| 21:22.40 | yukonbob | there are some neat things. |
| 21:23.05 | yukonbob | in-core OO, in-core database abstraction, NRE (non-recursive engine) |
| 21:23.12 | yukonbob | co-routines, |
| 21:23.24 | yukonbob | it's getting more lisp-y all the time. |
| 21:23.43 | starseeker | yukonbob: if 8.6 lets us run MGED and Archer in Aqua mode on the Mac and 8.5 doesn't (for whatever reason) I'd call that "necessary", but we could probably conditionalize for an Aqua build to demand 8.6 in that case |
| 21:24.32 | starseeker | in core OO would be nice, but that would require reworking our EXTENSIVE use of incrTcl throughout Archer to target TclOO |
| 21:24.40 | starseeker | I believe the word for that is "eek" |
| 21:25.38 | yukonbob | starseeker: I'm not clear on the implications of Tk rendering on OS X; look/feel, I guess, and not having to hack together look-alikes, because instead one simply uses native will reduce work on Tcl core team, and maybe(?) enhance end-users' experience, but otherwise, tcl/tk still work on OS X... |
| 21:26.01 | yukonbob | re: in-core OO |
| 21:26.18 | yukonbob | <PROTECTED> |
| 21:26.38 | yukonbob | they are more low-level primitives that a user-facing OO system can use. |
| 21:26.50 | starseeker | Tk rendering on OSX let's us run MGED and friends without needing an X11 server. X11 on the Mac has been a little tempremental |
| 21:27.13 | yukonbob | and they'll be common and standardized, so incr tcl can use, as well as snit, or any of other myriad OO systems can take advantage of. |
| 21:27.40 | starseeker | if we can get native libdm and libfb code in Tk, we can ditch X11 on the Mac completely and sidestep all the bugs :-) |
| 21:27.42 | yukonbob | does one currently need X11 to get a Tk window up on a mac? |
| 21:27.49 | starseeker | yes |
| 21:27.53 | yukonbob | hrm. |
| 21:28.14 | yukonbob | well, that's be a very nice side effect. |
| 21:28.26 | yukonbob | (getting rid of X11 on Mac) |
| 21:28.31 | starseeker | at least, if you want properly working display manager and any sort of raytracing ;-) |
| 21:28.39 | yukonbob | anyway -- I should get back to my own coding. |
| 21:28.59 | starseeker | 8.5 does indeed build aqua, only questions are stability, performance, etc. |
| 21:29.01 | starseeker | cool |
| 21:29.02 | yukonbob | ...and I'll take another stab at brl-cad on my system and see if I can get it up/running agin. |
| 21:29.15 | starseeker | Merry Christmas, and good luck! |
| 21:29.28 | starseeker | needs to get outta here... |
| 21:29.32 | yukonbob | thx :) Merry Christmas to you too. |
| 21:29.42 | yukonbob | ttyl starseeker |
| 21:33.56 | starseeker | ah yes, tkpath... http://tclbitprint.sourceforge.net/tkpath/quartz/index.html :-) |
| 21:34.06 | starseeker | gotta be some fun to be had there... |
| 21:37.54 | starseeker | or tkzinc |
| 21:38.14 | starseeker | stops himself and runs |
| 22:09.30 | *** join/#brlcad talcite (n=matthew@CPE00131078af68-CM001225ddf578.cpe.net.cable.rogers.com) |
| 22:11.47 | *** join/#brlcad R0b0t1 (n=Enigma@unaffiliated/r0b0t1) |
| 22:40.15 | talcite | brlcad: do we have a mailing list? |
| 22:49.03 | *** join/#brlcad mafm (n=mafm@162.Red-81-32-97.dynamicIP.rima-tde.net) |