| 01:17.11 | Notify | 03BRL-CAD:starseeker * 62221 brlcad/trunk/src/conv/step/STEPWrapper.cpp: Making a bit of progress towards hierarchy-first conversion, but something is going wrong with the matricies (see weldment_asm_solid.stp from the NIST DPPA repo for an example.) |
| 01:17.14 | starseeker | grrr |
| 01:34.42 | *** join/#brlcad Zhao_Anqing (clouddrift@222.205.15.157) |
| 03:11.20 | Notify | 03BRL-CAD:ejno * 62222 (brlcad/trunk/src/conv/3dm/conv3dm-g.cpp brlcad/trunk/src/conv/3dm/conv3dm-g.hpp): minor refactoring |
| 03:57.17 | Notify | 03BRL-CAD:zhaoanqing * 62223 brlcad/branches/nmgreorg/src/libnmg/CMakeLists.txt: rename the files in libnmg from nmg_XXX.c to XXX.c. move them from a dictionary to the root of libnmg. |
| 06:26.41 | Notify | 03BRL-CAD Wiki:ChaseKayet * 0 /wiki/User:ChaseKaye: |
| 06:37.37 | *** join/#brlcad mihaineacsu (~mihaineac@92.85.29.79) |
| 06:40.27 | *** join/#brlcad Zhao_Anqing (~clouddrif@183.157.160.22) |
| 06:51.29 | *** join/#brlcad d_rossberg (~rossberg@66-118-151-70.static.sagonet.net) |
| 07:42.55 | *** join/#brlcad Ch3ck_ (~localhost@195.24.220.134) |
| 07:52.47 | *** join/#brlcad teepee (~teepee@gateway/tor-sasl/teepee) |
| 08:25.09 | *** join/#brlcad clock (~clock@77-58-143-135.dclient.hispeed.ch) |
| 09:11.46 | *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51) |
| 09:42.41 | *** join/#brlcad ries (~ries@D979EA84.cm-3-2d.dynamic.ziggo.nl) |
| 10:07.47 | *** join/#brlcad Zhao_Anqing (~clouddrif@183.157.160.24) |
| 10:54.08 | *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51) |
| 11:39.55 | *** join/#brlcad andrei_ (~IceChat77@188.26.59.156) |
| 11:40.02 | andrei_ | Hello ! |
| 11:40.55 | andrei_ | Daniel, I've seen your e-mail related to sketch. I thought memcpy copies the pointers and values(integers and so forth) from the array, and that sketch_copy copies the arrays and the "whole structure" |
| 11:41.01 | andrei_ | I'm not sure what is wrong |
| 11:46.12 | *** join/#brlcad infobot (ibot@rikers.org) |
| 11:46.12 | *** topic/#brlcad is BRL-CAD || http://brlcad.org || logs: http://ibot.rikers.org/%23brlcad/ || GCI winners: Jacob Burroughs and Peter Amidon! || GSoC 2014 selections are announced! Thank you to all we got to work with. Remember that SOCIS is coming up right around the corner and you don't need a summer of code to get involved with open source. |
| 11:47.31 | *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51) |
| 12:18.56 | d_rossberg | andrei_: what does "="? |
| 12:22.15 | Notify | 03BRL-CAD:starseeker * 62224 brlcad/trunk/src/conv/step/STEPWrapper.cpp: Revert hierarchy-first change, since it breaks matrix building |
| 12:28.49 | d_rossberg | in "rtInternal = rt_copy_sketch(sketch->Internal());"? |
| 12:50.53 | Notify | 03BRL-CAD:brlcad * 62225 brlcad/trunk/CHANGES: bu_kill_parallel() doesn't kill the current process nor terminate a current thread set like the name implies, so make it go away. it basically only worked from a forked process. |
| 12:54.42 | Notify | 03BRL-CAD:brlcad * 62226 brlcad/trunk/src/libbu/bomb.c: remove the bu_kill_parallel() call. this is pretty much a pointless no-op for our existing use cases, and the very next step is abort(). |
| 12:55.41 | Notify | 03BRL-CAD:brlcad * 62227 brlcad/trunk/src/libbu/parallel.c: first steps towards allowing bu_parallel() from within a bu_parallel() context. |
| 12:58.09 | Notify | 03BRL-CAD:brlcad * 62228 brlcad/trunk/src/libbu/parallel.c: remove bu_kill_parallel() along with the fairly pointless and unlikely warning about the pid changing. |
| 12:59.02 | Notify | 03BRL-CAD:brlcad * 62229 brlcad/trunk/include/bu/parallel.h: no longer need the bu_kill_parallel decl, it be gone |
| 13:06.28 | *** join/#brlcad vladbogo (~vlad@86.121.98.7) |
| 13:16.43 | *** join/#brlcad infobot (ibot@rikers.org) |
| 13:16.43 | *** topic/#brlcad is BRL-CAD || http://brlcad.org || logs: http://ibot.rikers.org/%23brlcad/ || GCI winners: Jacob Burroughs and Peter Amidon! || GSoC 2014 selections are announced! Thank you to all we got to work with. Remember that SOCIS is coming up right around the corner and you don't need a summer of code to get involved with open source. |
| 13:21.05 | *** join/#brlcad vladbogo (~vlad@86.121.98.7) |
| 13:21.32 | Notify | 03BRL-CAD:brlcad * 62230 brlcad/trunk/src/libbn/poly.c: remove the statefulness. the fpe handling isn't even consistently used since it was only enabled for non-parallel app contexts, but this also lets us eliminate all of the globals and stateful bu_is_parallel() calls. if there are fpe issues, they should be dealt with here via logic checks (check for division by zero, etc). |
| 13:24.36 | *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51) |
| 13:27.54 | *** join/#brlcad Ch3ck_ (~localhost@195.24.220.134) |
| 14:08.52 | Notify | 03BRL-CAD:indianlarry * 62231 (brlcad/trunk/src/libbrep/PullbackCurve.cpp brlcad/trunk/src/librt/primitives/brep/brep.cpp): Added work around for case where OpenNURBS routines not returning normals for valid points along a trim. |
| 14:36.47 | Notify | 03BRL-CAD:carlmoore * 62232 brlcad/trunk/doc/docbook/system/man1/en/cad_boundp.xml: use <command> construct for all references to the command name; remove most underscores from the EXAMPLE (where I am concerned about proper implemtation of line breaks), and remove period from SEE ALSO |
| 14:36.53 | andrei_ | I ll check the sources again, but as far as I can remember |
| 14:37.27 | andrei_ | rt_copy_sketch returns a pointer to the new sketch( copied sketch) |
| 14:38.53 | andrei_ | Also, Daniel, is there any way to see details about what previous work has been done on coreinterface, I need it for my bachelor thesis and |
| 14:39.00 | andrei_ | wiki is empty |
| 14:50.13 | Notify | 03BRL-CAD:brlcad * 62233 brlcad/trunk/include/bu/log.h: mention BU_SETJUMP() more explicitly |
| 14:53.26 | d_rossberg | "rt_copy_sketch returns a pointer to the new sketch( copied sketch)" exactly, and assigning this pointer to rtInternal means to throw away the memory it pointed to before, giving a memory leak, etc. |
| 14:54.23 | d_rossberg | do you know http://brlcad.org/wiki/BRL-CAD%27s_core_C%2B%2B_interface ? |
| 15:06.50 | Notify | 03BRL-CAD:brlcad * 62234 brlcad/trunk/TODO: leave a note about the next bit of try/catch bu_setjump cleanup |
| 15:08.55 | d_rossberg | or http://brlcad.org/wiki/Object-oriented_interfaces ? |
| 15:10.03 | Notify | 03BRL-CAD:brlcad * 62235 brlcad/trunk/TODO: sort for next release, move non release stoppers |
| 15:14.05 | Notify | 03BRL-CAD:brlcad * 62236 brlcad/trunk/src/libbu/parallel.c: let ncpu=0 begin to mean something. intent is to mean 'use any and all available within parent context' which is all cpus/cores if there is no parent context (which this commit does) or something more intelligent, keeping track of available slots as threads complete. |
| 15:18.58 | Notify | 03BRL-CAD:starseeker * 62237 brlcad/trunk/src/conv/step/STEPWrapper.cpp: Re-enable the hierarchy-first approach. The difficulty was with LocalUnits::length, which is used for scaling in the hierarchy matrix build - the hierarchy logic was relying on the brep loop(s) to set the length and assuming that the local length is uniform for all objects. This behavior is restored and hierarchy-first conversion now appears |
| 15:19.00 | Notify | to be working, but we should look into doing a units lookup on a per-object basis during the hierarchy build so we can handle more complex situtations. |
| 15:21.03 | Notify | 03BRL-CAD:carlmoore * 62238 brlcad/trunk/doc/docbook/system/man1/en/cad_parea.xml: same comments as for cad_boundp |
| 15:38.53 | *** join/#brlcad raj12lnm (uid35020@gateway/web/irccloud.com/x-kuquwjliairegtzm) |
| 16:12.52 | Notify | 03BRL-CAD:brlcad * 62239 (brlcad/trunk/src/lgt/error.c brlcad/trunk/src/remrt/rtsrv.c): will have to test if this is cross-platform-true, but asserting that application codes no longer need to manually define bu_setjump_valid and bu_jmpbuf. next step is to remove the override altogether with a registered bomb callback. |
| 16:14.17 | mihaineacsu | brlcad: hi Sean! did you have a chance to take a look over the last patch I submitted? |
| 16:17.27 | Notify | 03BRL-CAD:carlmoore * 62240 brlcad/trunk/doc/docbook/system/man1/en/cell-fb.xml: remove non-printing characters; use <command> for references to the subject command; remove underscores of punctuation marks |
| 16:23.45 | *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51) |
| 16:24.13 | *** join/#brlcad clock (~clock@77-58-143-135.dclient.hispeed.ch) |
| 16:25.02 | Notify | 03BRL-CAD:carlmoore * 62241 brlcad/trunk/doc/docbook/system/man1/en/cell-fb.xml: remove -h, since it is gone from the cell-fb.c program |
| 16:35.43 | Notify | 03BRL-CAD Wiki:Jewel25Vxhy * 0 /wiki/User:Jewel25Vxhy: |
| 17:05.50 | *** join/#brlcad clock (~clock@77.58.143.135) |
| 17:14.08 | *** join/#brlcad javampire (4f811bce@gateway/web/freenode/ip.79.129.27.206) |
| 17:17.34 | *** join/#brlcad javampire (4f811bce@gateway/web/freenode/ip.79.129.27.206) |
| 17:39.41 | Notify | 03BRL-CAD Wiki:Mkenweaver * 0 /wiki/User:Mkenweaver: |
| 18:02.37 | Notify | 03BRL-CAD:carlmoore * 62242 brlcad/trunk/doc/docbook/system/man1/en/chan_mult.xml: underscore 'infile', and correct the command names in DESCRIPTION, along with providing <command> construct |
| 18:10.06 | Notify | 03BRL-CAD:carlmoore * 62243 brlcad/trunk/doc/docbook/system/man1/en/bwthresh.xml: did I commit this yesterday??? I had to implement the changes again, and they are saved temporarily in a home-directory copy |
| 18:19.31 | Notify | 03BRL-CAD:carlmoore * 62244 brlcad/trunk/doc/docbook/system/man1/en/chan_permute.xml: use <command> for command names (except in EXAMPLES), and use underscores for the arguments |
| 18:20.56 | *** part/#brlcad javampire (4f811bce@gateway/web/freenode/ip.79.129.27.206) |
| 18:21.48 | Notify | 03BRL-CAD:starseeker * 62245 (brlcad/trunk/src/conv/step/BRLCADWrapper.cpp brlcad/trunk/src/conv/step/BRLCADWrapper.h and 5 others): Add support for a -D option on step-g to do dry runs that go through the motions of the conversion without doing the STEP->ON_Brep or write-to-.g-file parts - useful for debugging. |
| 18:21.59 | *** join/#brlcad javampire (4f811bce@gateway/web/freenode/ip.79.129.27.206) |
| 18:23.28 | *** part/#brlcad javampire (4f811bce@gateway/web/freenode/ip.79.129.27.206) |
| 18:26.17 | Notify | 03BRL-CAD:carlmoore * 62246 brlcad/trunk/doc/docbook/system/man1/en/pixhalve.xml: apparently forgot to remove -h from the SYNOPSIS |
| 18:31.34 | Notify | 03BRL-CAD:carlmoore * 62247 brlcad/trunk/doc/docbook/system/man1/en/cmap-fb.xml: use <command> in the DESCRIPTION; reword remark about -H, which stays in because I found no other setting of height and width |
| 18:40.42 | Notify | 03BRL-CAD:carlmoore * 62248 brlcad/trunk/doc/docbook/system/man1/en/comgeom-g.xml: remove non-printing character; remove some underscores from EXAMPLE |
| 18:43.35 | Notify | 03BRL-CAD:brlcad * 62250 brlcad/trunk/src/libbu/tests/bu_semaphore.c: fix the bu_semaphore unit test. the test was trying to ensure that both threads actually made it into the parallel function, but the first thread was getting through too fast before the second thread could even test+set. it also wasn't setting thread-safe so there was potential both would skip setting parallel=1. now it spin loops the threads |
| 18:43.37 | Notify | until a second process can get there to set parallel=1. |
| 18:43.43 | Notify | 03BRL-CAD:carlmoore * 62249 brlcad/trunk/doc/docbook/system/man1/en/conv-vg2g.xml: use <command> in the DESCRIPTION; other stuff looked okay |
| 18:51.39 | Notify | 03BRL-CAD:brlcad * 62251 (brlcad/trunk/TODO brlcad/trunk/src/libbu/tests/CMakeLists.txt): now possible to run the bu_semaphore test, problem fixed (test threads were completing too quickly). |
| 18:56.54 | Notify | 03BRL-CAD:carlmoore * 62252 brlcad/trunk/doc/docbook/system/man1/en/cv.xml: redo the punctuation in SEE ALSO; because I remember 'twos complement' from long ago, I fixed its spelling here |
| 19:00.39 | Notify | 03BRL-CAD:carlmoore * 62253 brlcad/trunk/doc/docbook/system/man1/en/cy-g.xml: remove underscore from command name in EXAMPLE |
| 19:13.30 | Notify | 03BRL-CAD:carlmoore * 62254 (brlcad/trunk/include/bu/cv.h brlcad/trunk/src/fb/plot3-fb.c): fix 2 cases of 'twos-compliment' I found -- a 3rd was sent in in a separate commit of a man page |
| 19:25.42 | Notify | 03BRL-CAD:starseeker * 62255 (brlcad/trunk/src/conv/step/step-g/AdvancedBrepShapeRepresentation.cpp brlcad/trunk/src/conv/step/step-g/AdvancedFace.cpp and 41 others): Add tags for successful or unsuccessful loading in some of the step entities. Need to be systematic about this - these are just from testing. |
| 19:31.54 | andrei_ | how do I svn checkout https://svn.code.sf.net/p/brlcad/code/rt^3/trunk rt^3 |
| 19:31.56 | andrei_ | on windows |
| 19:32.00 | andrei_ | ^ is a reserved char |
| 19:35.31 | Notify | 03BRL-CAD:starseeker * 62256 brlcad/trunk/src/conv/step/STEPWrapper.cpp: Additinal validity checks on the reps from aCDSR before we try to access members - this avoids a crash with the biggest of the Neo FreeRunner models, but it's not immediately clear what led to this point - maybe Manifold_Surface_Shape_Representation not being processed, maybe something else. |
| 19:40.20 | Notify | 03BRL-CAD:carlmoore * 62257 brlcad/trunk/doc/docbook/system/man1/en/db.xml: revise the xml stuff so that we don't underscore brackets or '...' |
| 19:44.35 | ``Erik | change the ^ to something else |
| 19:44.45 | ``Erik | svn checkout https://svn.code.sf.net/p/brlcad/code/rt^3/trunk rt_3 |
| 19:45.37 | andrei_ | it didn't work, I always checkout with rt3 instead of rt^3 because writing a caret each time I cd to it is unpleasant |
| 19:45.48 | andrei_ | but TortoiseSVN gui did |
| 19:56.37 | andrei_ | how would one define a "Combination", in BRL-CAD context |
| 19:57.12 | andrei_ | from what I've seen it s a class/structure that contains data/pieces from multiple objects and the means to traverse it via trees? Does this seem correct? |
| 19:59.33 | Notify | 03BRL-CAD Wiki:EricTonnerre * 0 /wiki/User:EricTonnerre: |
| 19:59.51 | Notify | 03BRL-CAD:brlcad * 62258 (brlcad/trunk/src/libbu/tests/CMakeLists.txt brlcad/trunk/src/libbu/tests/bu_semaphore.c): prevent single-cpu systems from taking absurd amounts of time waiting for nobody. also support a cpu count option to specify how many cpus to use, and add a few tests to exercise single-threaded/p2/p3. |
| 20:02.09 | Notify | 03BRL-CAD Wiki:EricTonnerre * 7716 /wiki/ARL_Technical_Reports: Coffre |
| 20:37.08 | Notify | 03BRL-CAD:brlcad * 62259 brlcad/trunk/src/libbu/backtrace.c: backtrace megalogs .. include backtrace info for all running threads (specifically in addition to bt full as some versions of gdb don't/didn't support this feature) |
| 20:38.11 | Notify | 03BRL-CAD:brlcad * 62260 brlcad/trunk/src/libbu/crashreport.c: include the thread/cpu ID number in crash reports |
| 20:44.37 | Notify | 03BRL-CAD:starseeker * 62261 brlcad/trunk/doc/docbook/system/man1/en/db.xml: fix docbook build |
| 20:45.02 | Notify | 03BRL-CAD:brlcad * 62262 (brlcad/trunk/include/bu/parallel.h brlcad/trunk/src/libbu/bomb.c brlcad/trunk/src/libbu/globals.c): make BU_SETJUMP long jump contexts be per-cpu so that we don't have to worry about multiple BU_SETJUMPs from different threads clobbering each other. |
| 20:47.50 | Notify | 03BRL-CAD:brlcad * 62263 brlcad/trunk/CHANGES: the jump bufs are now per-cpu arrays |
| 20:56.40 | Notify | 03BRL-CAD:brlcad * 62264 brlcad/trunk/src/libbu/parallel.c: slight decrease in library statefulness affecting recursive bu_parallel invocations. make pid_of_initiating_thread global go away as it no longer serves any use. |
| 20:58.41 | Notify | 03BRL-CAD:brlcad * 62265 brlcad/trunk/src/librt/db_tree.c: check should have no effect as its still not yet possible to invoke recursively and none of our code is attempting this that I'm aware of. |
| 20:59.20 | Notify | 03BRL-CAD:brlcad * 62266 brlcad/trunk/src/librt/primitives/submodel/submodel.c: nobody else checks either, so why are we bothering here? |
| 21:04.24 | Notify | 03BRL-CAD:carlmoore * 62267 (brlcad/trunk/doc/docbook/system/man1/en/dbclean.xml brlcad/trunk/src/conv/dbclean.sh): do touchup AND SOME REWRITING, which included 2 comment lines in dbclean.sh |
| 21:19.48 | Notify | 03BRL-CAD Wiki:Sean * 7717 /wiki/ARL_Technical_Reports: Reverted edits by [[Special:Contributions/EricTonnerre|EricTonnerre]] ([[User talk:EricTonnerre|talk]]) to last revision by [[User:Sean|Sean]] |
| 21:20.15 | Notify | 03BRL-CAD Wiki:Sean * 0 /wiki/User:EricTonnerre: Spamming links to external sites |
| 21:54.19 | Notify | 03BRL-CAD:starseeker * 62268 (brlcad/trunk/src/conv/step/step-g/AmountOfSubstanceContextDependentUnit.cpp brlcad/trunk/src/conv/step/step-g/AmountOfSubstanceConversionBasedUnit.cpp and 154 others): Add entity_status sets to the other supported object types. Need to look at how to flag the failed factory calls as well. |