| 01:31.29 | brlcad | starseeker: http://google-opensource.blogspot.com/2012/05/introducing-ceres-solver-nonlinear.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+GoogleOpenSourceBlog+%28Google+Open+Source+Blog%29 |
| 01:33.37 | brlcad | er, http://google-opensource.blogspot.com/2012/05/introducing-ceres-solver-nonlinear.html |
| 01:34.07 | brlcad | maybe useful for the polytonurbs |
| 01:40.06 | *** join/#brlcad Neil___ (~chatzilla@117.229.84.80) |
| 01:44.40 | CIA-124 | BRL-CAD: 03brlcad * r50395 10/brlcad/trunk/TODO: make sure we can call db_dirbuild() multiple times without creating duplicate entries. |
| 01:45.22 | CIA-124 | BRL-CAD: 03brlcad * r50396 10/brlcad/trunk/ (include/raytrace.h src/librt/db5_scan.c): move the doxygen comments from source to header |
| 01:49.00 | CIA-124 | BRL-CAD: 03brlcad * r50397 10/brlcad/trunk/src/rt/viewedge.c: note why this block is commented out for now until a fix can be made. should always annotate an #if0 block, otherwise it's a real pain in the ass to review later. |
| 01:59.42 | brlcad | starseeker: is r50388 (misc/CMake/FindX11.cmake multilib change) application of sf patch 3522852 from crdueck? |
| 02:00.11 | brlcad | and r50390 |
| 02:00.22 | brlcad | waves to Neil___ |
| 02:01.00 | brlcad | reverting the autogentrification for now ... soon enough all, soon enough it will be gone |
| 02:01.13 | Neil___ | hey Sean! |
| 02:03.23 | brlcad | it's important as a community that we agree to a deprecation policy and stick to it, otherwise our users cannot rely on us and that hinders adoption and growth |
| 02:05.01 | brlcad | the intent as the policy is currently written is that users are guaranteed to have at least two fully minor release sets to adjust to changes we make |
| 02:06.05 | CIA-124 | BRL-CAD: 03brlcad * r50398 10/brlcad/trunk/ (244 files in 132 dirs): |
| 02:06.06 | CIA-124 | BRL-CAD: revert r50386,50391-50393 as removal of prior build system is very premature. |
| 02:06.06 | CIA-124 | BRL-CAD: it was just deprecated in 7.20.0 and is supposed to exist deprecated across |
| 02:06.06 | CIA-124 | BRL-CAD: three minor releases. implies 7.24.0 would be the soonest (or 7.26.0 if |
| 02:06.06 | CIA-124 | BRL-CAD: interpreted strictly), and we're not even to 7.22.0 yet. |
| 02:06.13 | brlcad | the first minor is the introduction and suprise of the change, the second minor gives them time to schedule development time to review and accommodate/change, with the feature finally being removed sometime during/after the third minor |
| 02:22.09 | starseeker | brlcad: no - r50388 is related to his patch, but (should) solve the same problem in a non-OS specific way |
| 02:22.41 | starseeker | but props to him for digging in and working towards a solution |
| 02:23.45 | starseeker | brlcad: cool link - thanks! Will have to check that out |
| 02:23.53 | starseeker | compare it to http://joachimwuttke.de/lmfit/ |
| 02:25.11 | starseeker | like the sound of "build 3d models" for the ceres solver |
| 02:26.38 | starseeker | can't say he is surprised autotools is back |
| 02:31.22 | starseeker | I suppose I really should add a firing of the autotools distcheck to the CMake distcheck-full target - otherwise I'll just end up trying to fix it right before release again... |
| 02:31.36 | starseeker | ponders... |
| 02:42.09 | brlcad | starseeker: did something come up that prompted the purge or was erik just eager to do some easy slash'n'burning? |
| 02:42.21 | starseeker | the latter |
| 02:42.38 | brlcad | ah, hm |
| 02:43.10 | starseeker | he was planning on running for the hills ;-) |
| 02:43.16 | brlcad | i'll post some commentary to the list to see what others think, but I'm trying to crystalize an example so it's clear when things can go |
| 02:43.39 | brlcad | the autotools purge raises an interesting point |
| 02:43.50 | brlcad | most of the time, we don't *really* care when something goes because there's no cost |
| 02:44.06 | starseeker | can wait for the scheduled removal - just trying to figure out an easy way to integrate its testing into the distcheck-full test |
| 02:44.13 | brlcad | but then other times (autoconf, jove) it's a pain, but worth following a process for the sake of our community |
| 02:45.07 | starseeker | maybe flag those as "known active interest" ? |
| 02:46.29 | brlcad | firefox being a perfect example of giving no guarantees causing community distress |
| 02:46.42 | starseeker | <snort> no argument there |
| 02:46.43 | brlcad | what would that do though? |
| 02:47.04 | starseeker | you had mentioned cases where we don't *really* care vs. cases where we do |
| 02:47.17 | starseeker | with community interests being the distinction |
| 02:47.18 | brlcad | if we can arbitrarily mark something as "we don't want to follow the rules on this one", then that's the same as having no rule |
| 02:48.03 | brlcad | one thing that can be clarifies is the 2 minors vs 3 minors vs 4 minors |
| 02:48.11 | starseeker | ah, gotcha |
| 02:48.11 | brlcad | s/clarifies/clarified/ |
| 02:48.17 | starseeker | yeah, that would be nice |
| 02:48.56 | brlcad | the original intent was 3 minors (or at least a quarter of a year), but the wording strictly interpreted implies 4 minors |
| 02:49.10 | starseeker | does "two minor releases after 7.20.0" mean 7.24.0 releases without it or "complete 7.22.0 and 7.24.0, then release 7.26.0 without it" |
| 02:50.04 | brlcad | just thinking about the intent described above, it's one minor to announce, another to let them fix, and then it can be gone any time after .. that's probably about as realistically "quick" as we can make it without saying "tough shit sherlock, it's gone" |
| 02:51.10 | brlcad | as it's currently worded, it would mean 7.26 since it says specifically that three minors have to have the announcement |
| 02:51.26 | brlcad | so you could remove it at 7.24.2 .. except you cannot remove on a patch |
| 02:51.41 | brlcad | so 7.26 becomes soonest |
| 02:52.33 | brlcad | that said, I think 7.24.0 is fine for a minimum if 7.20 and 7.22 have the announcement, there have been three releases of any kind with the notice, and at least a quarter has passed |
| 02:53.38 | CIA-124 | BRL-CAD: 03starseeker * r50399 10/brlcad/trunk/Makefile.am: Well, that's one issue already spotted - let Makefile.am know there are no more .cmake toplevel doc files. |
| 02:55.02 | starseeker | brlcad: sounds reasonable |
| 02:56.28 | brlcad | so I'd asked about crdueck's patch just because it'd at least warrant special thanks in AUTHORS file (and the patch should probably be closed) |
| 03:11.46 | starseeker | nods - I always forget to update the AUTHORS file |
| 03:14.13 | CIA-124 | BRL-CAD: 03starseeker * r50400 10/brlcad/trunk/ (4 files in 3 dirs): Fix typos in deprecation messages, more Makefile.am changes (untested in the misc case) |
| 03:15.58 | starseeker | there is actually one (probable) remaining issue with archlinux, but it may be a CMake (as it Kitware CMake project) level fix - I doubt trying to do a 32 bit BRL-CAD compilation on archlinux would work currently |
| 03:16.47 | starseeker | there is a property that can be set to have CMake look in lib64 and friends for libraries, but so far as I know there is nothing equalivent that will make it look in lib32 |
| 03:17.24 | CIA-124 | BRL-CAD: 03brlcad * r50401 10/brlcad/trunk/doc/deprecation.txt: |
| 03:17.24 | CIA-124 | BRL-CAD: per IRC discussion in the aftermath of an autotools removal revert, tighten up |
| 03:17.24 | CIA-124 | BRL-CAD: the deprecation language to specify two minor release needing to include the |
| 03:17.24 | CIA-124 | BRL-CAD: deprecation notice. include an example so it's even more explicit and not |
| 03:17.24 | CIA-124 | BRL-CAD: subject to intepretation. strictly speaking, this reduces our deprecation |
| 03:17.24 | CIA-124 | BRL-CAD: window by a minor but makes the policy about as tight as it can get without |
| 03:17.25 | CIA-124 | BRL-CAD: being reckless or non-existent. |
| 03:27.20 | starseeker | the 64 bit global property for CMake is FIND_LIBRARY_USE_LIB64_PATHS |
| 03:28.45 | starseeker | and whadya know: http://www.cmake.org/Bug/view.php?id=11260&nbn=1 |
| 03:32.58 | CIA-124 | BRL-CAD: 03starseeker * r50402 10/brlcad/trunk/ (CMakeLists.txt misc/CMake/distcheck_autotools.cmake.in): |
| 03:32.58 | CIA-124 | BRL-CAD: This looks like it might be a viable way of regularly checking the Autotools |
| 03:32.59 | CIA-124 | BRL-CAD: distcheck target (at least to the extent of making sure it doesn't error out). |
| 03:32.59 | CIA-124 | BRL-CAD: Right now it looks like trunk Autotools distcheck is not happy, so this breaks |
| 03:32.59 | CIA-124 | BRL-CAD: distcheck-full until that gets sorted out. |
| 03:34.05 | CIA-124 | BRL-CAD: 03starseeker * r50403 10/brlcad/trunk/src/other/incrTcl/ (itcl/Makefile.am itk/Makefile.am): Just ignore all of CMake |
| 03:37.11 | CIA-124 | BRL-CAD: 03starseeker * r50404 10/brlcad/trunk/CMakeLists.txt: Add a note about the FIND_LIBRARY_USE_LIB32_PATHS bug, so we know where to look if/when the issue comes up. |
| 03:55.10 | CIA-124 | BRL-CAD: 03starseeker * r50405 10/brlcad/trunk/src/other/libz/Makefile.am: zconf.h.in is no more |
| 05:58.05 | *** join/#brlcad Jak_o_Shadows (~Fake@CPE-144-136-211-241.sa.bigpond.net.au) |
| 05:58.06 | *** join/#brlcad Jak_o_Shadows (~Fake@unaffiliated/jak-o-shadows/x-0479135) |
| 06:09.01 | *** join/#brlcad stas (~stas@188.24.35.114) |
| 06:18.50 | *** join/#brlcad cristina (~cristina@188.24.77.63) |
| 06:22.43 | kanzure | brlcad.org is back to borked? |
| 07:02.34 | *** join/#brlcad Neil___ (~chatzilla@117.229.111.54) |
| 07:08.41 | *** join/#brlcad pawleeq (~pawleeq@212-96-188-229.cust.selfnet.cz) |
| 07:13.42 | *** join/#brlcad pawleeq_ (~pawleeq@212-96-188-229.cust.selfnet.cz) |
| 07:27.07 | *** join/#brlcad merzo (~merzo@71-86-200-46.pool.ukrtel.net) |
| 07:32.26 | *** join/#brlcad d_rossberg (~rossberg@BZ.BZFLAG.BZ) |
| 07:32.35 | *** join/#brlcad ksuzee (~ksuzee91@46.149.82.166) |
| 07:46.52 | *** join/#brlcad stas (~stas@82.208.133.12) |
| 08:08.38 | *** part/#brlcad ksuzee (~ksuzee91@46.149.82.166) |
| 08:38.32 | *** join/#brlcad ksuzee (~ksuzee91@46.149.82.166) |
| 08:48.25 | brlcad | kanzure: looking fine here |
| 08:50.47 | brlcad | fix was rather instantaneous, there was an index file overriding the usual and only for the main page |
| 08:51.03 | brlcad | maybe you're having browser cache problems |
| 09:03.37 | CIA-124 | BRL-CAD: 03Ksuzee 07http://brlcad.org * r3528 10/wiki/User:Ksuzee: |
| 09:03.56 | CIA-124 | BRL-CAD: 03Ksuzee 07http://brlcad.org * r3529 10/wiki/User:Ksuzee: |
| 09:10.43 | CIA-124 | BRL-CAD: 03Ksuzee 07http://brlcad.org * r3530 10/wiki/User:Ksuzee/Reports: New page: = GSoC 2012 = == 0 week == *2/09/2012 - installing Ubuntu-12.04 TODO: 1) downloading sources 2) run BRL-CAD from a Subversion |
| 09:11.10 | CIA-124 | BRL-CAD: 03Ksuzee 07http://brlcad.org * r3531 10/wiki/User:Ksuzee/Reports: |
| 09:23.38 | *** join/#brlcad kane_ (~Mesut@g226123008.adsl.alicedsl.de) |
| 09:23.50 | *** join/#brlcad stas (~stas@82.208.133.12) |
| 10:02.55 | *** join/#brlcad stas (~stas@82.208.133.12) |
| 10:20.32 | *** join/#brlcad stas (~stas@82.208.133.12) |
| 10:52.42 | *** part/#brlcad ksuzee (~ksuzee91@46.149.82.166) |
| 11:07.44 | ``Erik | poop, reverted O.o section b in deprecation.txt was self-contradicting with it's poor grammar and technically, the cmake thing satisfies c with s/configure/configure.cmake.sh/ (starseeker mentioned automake not building the right stuff or the right way or something, to boot). Ah well, was worth a shot :D |
| 11:09.44 | *** join/#brlcad jordisayol (~jordisayo@unaffiliated/jordisayol) |
| 11:39.36 | *** join/#brlcad stas (~stas@82.208.133.12) |
| 13:11.45 | *** join/#brlcad ibot (~ibot@rikers.org) |
| 13:11.45 | *** topic/#brlcad is BRL-CAD Open Source Solid Modeling || http://brlcad.org || logs: http://ibot.rikers.org/%23brlcad/ || Congratulations to our 11 accepted GSoC students!! |
| 13:14.21 | *** join/#brlcad kane_ (~Mesut@g226123008.adsl.alicedsl.de) |
| 13:54.55 | *** join/#brlcad Neil___ (~chatzilla@117.229.98.73) |
| 14:54.31 | brlcad | ``Erik: it's not minimally impacting by the nature that you have to have cmake installed for the replacement to work (and all the command-line option names are subtlely different) |
| 14:54.40 | brlcad | considered whether it was a long while back |
| 14:55.26 | brlcad | there wasn't poor grammar, read exactly it required 4 minor releases iirc |
| 14:56.11 | brlcad | plus autotools build has more/different steps so any scripted build system invocation was certainly not regexable |
| 14:56.52 | brlcad | either way, it sparked some clarity in the policy and should be minimal now |
| 14:58.30 | brlcad | what that does mean is that jove can go away, though |
| 14:58.47 | brlcad | maybe get your purge rocks off there ;) |
| 15:18.32 | CIA-124 | BRL-CAD: 03r_weiss * r50406 10/brlcad/trunk/src/tclscripts/mged/lc.tcl: Added tcl script 'lc.tcl' for a new mged command 'lc' to list the region codes (ie attributes) within a group. |
| 15:20.06 | jordisayol | brlcad: about this error I told you yesterday http://paste.debian.net/166899/ it happens only in ubuntu 10.04, but not in 11.04 and newer |
| 15:26.53 | CIA-124 | BRL-CAD: 03r_weiss * r50407 10/brlcad/trunk/src/tclscripts/mged/ (CMakeLists.txt Makefile.am): Updated files 'CMakeLists.txt' and 'Makefile.am' to add the tcl script 'lc.tcl' to mged. This adds a new command called 'lc' to mged which lists the codes (ie attributes) of regions within a group. |
| 15:54.13 | CIA-124 | BRL-CAD: 03n_reed * r50408 10/brlcad/trunk/src/other/step/src/ (3 files in 2 dirs): apply cleanup changes from SCL git 9f83ece, 5e6ebb0, and ca50d53 |
| 16:40.46 | kanzure | brlcad: ah perhaps you're right. brlcad.org?3941 loads your usual site |
| 17:37.54 | *** join/#brlcad kane__ (~Mesut@e181167076.adsl.alicedsl.de) |
| 18:05.22 | *** join/#brlcad merzo (~merzo@71-86-200-46.pool.ukrtel.net) |
| 18:48.38 | *** join/#brlcad Stattrav_ (~Stattrav@61.12.114.82) |
| 18:59.58 | *** join/#brlcad cristina (~cristina@188.24.77.63) |
| 19:13.18 | *** join/#brlcad andrei_ (~andrei@188.25.170.86) |
| 19:13.22 | andrei_ | hello! |
| 19:15.51 | *** join/#brlcad stas (~stas@188.24.35.114) |
| 19:34.58 | cristina | hi andrei_ |
| 19:48.22 | *** join/#brlcad Neil__ (~chatzilla@117.229.98.73) |
| 19:49.47 | CIA-124 | BRL-CAD: 03starseeker * r50409 10/brlcad/trunk/src/other/ (5 files in 5 dirs): Switch other src/other Makefile.am files to just straight-up ignore CMake dir |
| 20:05.32 | CIA-124 | BRL-CAD: 03starseeker * r50410 10/brlcad/trunk/configure.ac: No wfobj with autotools |
| 20:17.09 | CIA-124 | BRL-CAD: 03starseeker * r50411 10/brlcad/trunk/src/libgcv/ (Makefile.am wfobj/Makefile.am): finish commenting out wfobj for autotools |
| 20:29.19 | CIA-124 | BRL-CAD: 03starseeker * r50412 10/brlcad/trunk/src/libgcv/wfobj/CMakeLists.txt: update wfobj CMakeLists.txt |
| 20:57.21 | *** join/#brlcad stas (~stas@188.24.35.114) |
| 21:03.54 | CIA-124 | BRL-CAD: 03bob1961 * r50413 10/brlcad/trunk/src/mged/chgview.c: This fixes an issue in MGED where matrix edit was broken. |
| 21:23.49 | starseeker | brlcad: if you get a chance, would you mind seeing what happened to "make distcheck" in trunk's autotools build? Something seems to be buggered with enable/disable logic (png isn't detected but the local copy isn't enabled, for example) - not sure if I broke something or what's going on... |
| 21:25.35 | CIA-124 | BRL-CAD: 03starseeker * r50414 10/brlcad/trunk/ (configure.ac src/conv/Makefile.am src/other/Makefile.am): Turn off step - Nick's improvements have now made the new step-g build process too different from the original to be worth re-implementing in autotools |
| 22:55.21 | brlcad | jordisayol: hm, not enough information in that error snippet to diagnose |
| 22:57.07 | jordisayol | hmmm, it's all the log error from rtwizard. anyway, it only happen in a 2year old ubuntu release, so I don't care about it |
| 22:57.20 | brlcad | hm, okay |
| 22:57.35 | brlcad | all it's saying is that it cannot talk to the framebuffer, but not why |
| 22:57.43 | brlcad | would have to try some tests |
| 22:58.22 | jordisayol | is up to you. I can do some test |
| 22:58.24 | brlcad | a recent change to rtwizard may fix that issue if it's related to X vs ogl |
| 22:58.44 | brlcad | if you run "fbhelp" what does it list for the first two device types? |
| 22:59.14 | jordisayol | give me a second. i've to start up in a virtualbox machine |
| 23:00.02 | jordisayol | btw, this happen with last svn too |
| 23:02.06 | brlcad | the rtwizard change I'm thinking of happened yesterday |
| 23:05.17 | jordisayol | with 7.20.6 release |
| 23:05.17 | jordisayol | /dev/ogl Silicon Graphics OpenGL |
| 23:05.17 | jordisayol | /dev/X 24 bit X Window System (X11) |
| 23:08.18 | brlcad | hm, could be related |
| 23:08.57 | brlcad | make sure rtwizard is not running then run this: fbserv 1 /dev/X & |
| 23:10.21 | jordisayol | [1] 1950 |
| 23:10.28 | brlcad | k, good |
| 23:10.56 | brlcad | then try: rt -F1 /usr/brlcad/rel-7.20.6/share/brlcad/7.20.6/db/moss.g all.g |
| 23:11.12 | brlcad | should see a simple rendering |
| 23:14.08 | jordisayol | yes, it properly render |
| 23:14.24 | brlcad | kill -9 %1 |
| 23:15.18 | jordisayol | bash: kill: %1: no such job |
| 23:15.34 | brlcad | er, kill 1950 |
| 23:15.37 | brlcad | ? |
| 23:16.18 | jordisayol | now it worked |
| 23:16.49 | brlcad | rtwizard /usr/brlcad/rel-7.20.6/share/brlcad/7.20.6/db/moss.g |
| 23:19.48 | jordisayol | same result, rtwizard open it, after selecting "all" and when try to render, error raise |
| 23:30.54 | brlcad | hm, you completed all of the "steps" in the menu? |
| 23:31.04 | brlcad | does preview work? |
| 23:31.39 | jordisayol | preview on right side? yes |
| 23:33.32 | brlcad | no, on the menu, it's render -> preview |
| 23:34.37 | jordisayol | no, it's there when error rise |
| 23:35.27 | jordisayol | when i do this, prints this no terminal: |
| 23:35.27 | jordisayol | ogl_getmem: shmget failed, errno=22 |
| 23:35.27 | jordisayol | ogl_getmem: Unable to attach to shared memory. |
| 23:35.27 | jordisayol | ogl_getmem: malloc failure |
| 23:35.27 | jordisayol | fb_open: can't open device "/dev/ogl", ret=-1. |