IRC log for #brlcad on 20130801

02:29.27 *** join/#brlcad yiyus (1242712427@je.je.je)
02:48.47 *** join/#brlcad yiyus (1242712427@je.je.je)
03:32.50 *** join/#brlcad yiyus (1242712427@je.je.je)
03:33.22 Notify 03BRL-CAD:starseeker * 56400 (brlcad/trunk/include/raytrace.h brlcad/trunk/src/libged/comb.c brlcad/trunk/src/librt/search.c): string search plans are likely to be a very common input - provide functions to handle them automatically.
03:47.10 *** join/#brlcad jass (~jass@202.164.53.117)
03:48.48 jass exit
03:48.57 jass exit
03:53.07 Notify 03BRL-CAD:brlcad * 56401 (brlcad/trunk/misc/CMake/BRLCAD_Targets.cmake brlcad/trunk/src/fb/tests/CMakeLists.txt and 6 others): simplify. reviewing all of the remaining NO_INSTALL applications that lacked LOCAL, they all needed it. this combines them. NO_INSTALL now means leave/put the binary in the build directory.
04:00.51 *** join/#brlcad jass (~jass@202.164.53.117)
04:15.55 Notify 03BRL-CAD:phoenixyjll * 56402 brlcad/trunk/src/libbrep/intersect.cpp: It seems that J.invert() doesn't work (it cause brep arb8.brep intersect ehy.brep 0 1 on csgbrep.g to fail). Use explicit pseudo-inverse.
04:35.30 Notify 03BRL-CAD:brlcad * 56403 (brlcad/trunk/CMakeLists.txt brlcad/trunk/misc/CMake/BRLCAD_Targets.cmake and 10 others): consistency with NO_INSTALL and other var/opts. make NOSTRICT/NOSTRICTCXX and var friends be NO_STRICT/NO_STRICT_CXX.
04:44.19 zero_level clear
04:45.02 zero_level oops! nevermind :-
05:11.02 Notify 03BRL-CAD:phoenixyjll * 56404 brlcad/trunk/src/libbrep/intersect.cpp: Code clean up. Move the functions used by multiple ON_Intersect()s to the beginning.
05:21.15 *** join/#brlcad caen23 (~caen23@92.81.175.190)
05:50.34 Notify 03BRL-CAD:phoenixyjll * 56405 brlcad/trunk/src/libbrep/intersect.cpp: Separate the Newton iterations of PCI out to a function.
06:19.16 *** join/#brlcad caen23_ (~caen23@92.81.199.45)
07:52.28 *** join/#brlcad merzo (~merzo@user-94-45-58-138-1.skif.com.ua)
08:47.14 *** join/#brlcad d_rossberg (~rossberg@66-118-151-70.static.sagonet.net)
08:50.45 Notify 03BRL-CAD:d_rossberg * 56406 brlcad/trunk/misc/win32-msvc/Dll/TclDummies.c: added some flesh to the function stubs, they will comile with the strict flag now
09:57.13 Notify 03BRL-CAD:phoenixyjll * 56407 brlcad/trunk/src/libbrep/intersect.cpp: Try to implement PSI in a similar fashion of other intersections, so that further the surface tree structures can be easily reused during different intersections.
10:03.15 Notify 03BRL-CAD:phoenixyjll * 56408 brlcad/trunk/src/libbrep/intersect.cpp: Mark tree as unused.
10:47.08 *** join/#brlcad Ch3ck_ (~Ch3ck@195.24.220.16)
10:48.53 *** join/#brlcad Ch3ck__ (~Ch3ck@66-118-151-70.static.sagonet.net)
12:23.41 zero_level hi ``Erik : I take those mails as an assurance of things being set right.
12:23.49 zero_level So the benchmark works.
12:23.52 zero_level ?
12:24.08 zero_level ``Erik : Moving on with dev work.
12:24.52 zero_level Thanks GSOC students Ch3ck_ , Izak__ and ``Erik for making benchmark and testing on your machines.
12:28.46 Ch3ck_ zero_level:its ok
12:37.53 Notify 03BRL-CAD:tbrowder2 * 56409 (brlcad/trunk/src/burst/burst.c brlcad/trunk/src/burst/extern.h brlcad/trunk/src/burst/prnt.c): move usage string into file with main; eliminate unnecessary function
12:43.18 Notify 03BRL-CAD Wiki:QWERTYUIOP * 0 /wiki/User:QWERTYUIOP:
12:46.14 Notify 03BRL-CAD:tbrowder2 * 56410 brlcad/trunk/src/burst/burst.c: make new string to hold bu_getopt option char input; ws
12:48.42 Notify 03BRL-CAD:mohitdaga * 56411 (brlcad/trunk/src/libicv/CMakeLists.txt brlcad/trunk/src/libicv/fileformat.c): Segregate functions Supporting bw images to a new file bw.c.
12:58.25 *** join/#brlcad merzo (~merzo@user-94-45-58-138-1.skif.com.ua)
13:03.17 ``Erik bw.c fails to compile
13:04.55 zero_level msg ?
13:05.02 ``Erik http://paste.lisp.org/display/138293
13:05.59 brlcad zero_level: I get RIGHT answers again now too
13:06.02 brlcad just fyi
13:06.17 Notify 03BRL-CAD Wiki:Phoenix * 5893 /wiki/User:Phoenix/GSoc2013/Reports: /* Week 7 */
13:07.53 zero_level brlcad : Thanks :-)
13:10.25 Notify 03BRL-CAD:mohitdaga * 56412 brlcad/trunk/src/libicv/bw.c: Add sys/stat header for FLAGS used in read/write.
13:10.28 zero_level ``Erik : worked on my machine. Checked bu removing each header.
13:10.54 zero_level I believe after r56412 must compile cleanly on your machine.
13:11.03 zero_level Pls do ack. Thanks.
13:11.14 ``Erik zero_level: I like to test a couple different platforms just to verify things like... I saw the error on both fbsd (brlcad.org) and a rhel machine
13:12.34 ``Erik yup, works now and you chose the right header (stat.h also has those defines, but your purpose would want fcntl.h)
13:12.58 zero_level ok. Thanks.
13:25.57 zero_level ``Erik I am poor at naming functions/file names.
13:26.26 zero_level Can u suggest me name for for a file where I keep uchar2double double2uchar.
13:26.38 zero_level Also the scope of this files will extend in future.
13:27.10 zero_level Where we will keep routines to convert to other data types for high resolution bmp, png images.
13:27.21 zero_level my suggestion is do.c
13:28.44 zero_level opps double.c
13:33.29 ``Erik uh, util.c ?
13:34.19 ``Erik libicv_private.h and do 'em as macro or static inline? :D
13:35.24 ``Erik seems to be some precidence for doing something like icv_util.c
13:35.32 ``Erik precedence
13:40.03 zero_level hi. getting this error in benchmark now.
13:40.05 zero_level http://paste.kde.org/p18486ad6/
13:40.18 zero_level related to Permissions of .g file
13:41.14 Notify 03BRL-CAD:mohitdaga * 56413 (brlcad/trunk/src/libicv/CMakeLists.txt brlcad/trunk/src/libicv/fileformat.c): Segrgating routines related to pix format
13:41.58 ``Erik might have to manually delete m35.g (or change it's permissions)
13:42.06 brlcad zero_level: perhaps type2type.c or encoding.c or conversion.c
13:50.56 Notify 03BRL-CAD Wiki:IIIzzzaaakkk * 5894 /wiki/User:Izak/GSOC_2013_logs: /* Mid-term Evaluation week */
13:51.28 Notify 03BRL-CAD:starseeker * 56414 brlcad/trunk/doc/docbook/system/mann/en/comb.xml: reorganize comb man page - break options out into list.
13:57.03 Notify 03BRL-CAD:mohitdaga * 56415 brlcad/trunk/src/libicv/CMakeLists.txt: Trailing ws
14:02.02 *** join/#brlcad caen23 (~caen23@92.81.199.45)
14:02.53 zero_level I guess encoding.c is nice.
14:16.22 Notify 03BRL-CAD:mohitdaga * 56416 (brlcad/trunk/src/libicv/CMakeLists.txt brlcad/trunk/src/libicv/bw.c brlcad/trunk/src/libicv/pix.c): Add a new file encoding.c. This file will contain data conversion for saving and loading from different formats of images. These routines will act as a bridge. For eg. file.someformat ---> Use routine from this file --> Load to icv_image struct --> do processing --> Use routines from this
14:16.25 Notify file for desired formats --> Save in the desired format.
14:16.42 Ch3ck_ working on the unit test for bn_poly_quadratic_roots and I wish to know what to do when a quadratic equation has two roots how does bn_poly_quadratic_roots() handle that case.
14:17.21 zero_level brlcad : Although encoding.c is fine as a name. But this will create a bunch of very small files. (I guess that is fine) else ``Erik's idea of a utility shld be followed
14:22.35 Ch3ck_ its ok guys seen a way to handle it.. :)
14:24.11 Notify 03BRL-CAD:mohitdaga * 56417 (brlcad/trunk/src/libicv/encoding.c brlcad/trunk/src/libicv/pix.c): Add file descriptions.
14:25.42 Notify 03BRL-CAD:mohitdaga * 56418 brlcad/trunk/src/libicv/fileformat.c: Adding file information for extern functions.
14:26.30 zero_level ``Erik what was the tag flags in "icv_image_file" struct used for?
14:26.40 zero_level I added that without change.
14:26.53 zero_level I see there are two flas used.
14:27.00 zero_level for stream and buffer.
14:27.29 zero_level *flags.
14:35.06 *** join/#brlcad cogitokat (~kat@ip70-171-0-190.ga.at.cox.net)
14:37.58 ``Erik I don't think the flags field was ever actually used, looked like I was setting up things to allow a bif to be streamed or buffered but that capability never became important
14:38.27 ``Erik way back in '07 O.o
14:44.31 *** join/#brlcad Izak_ (~Izak@195.24.220.16)
14:53.48 Notify 03BRL-CAD:mohitdaga * 56419 (brlcad/trunk/include/icv.h brlcad/trunk/src/libicv/fileformat.c): Remove Flags from 'icv_image' struct.
14:56.31 Notify 03BRL-CAD:carlmoore * 56420 (brlcad/trunk/doc/docbook/system/mann/en/comb.xml brlcad/trunk/include/raytrace.h): remove trailing blanks/tabs, and fix a spelling
15:02.46 Ch3ck_ hey guys what's happening to brlcad source these days compiling code and i'm getting this opennurbs error : http://pastebin.com/8a6FbJzs
15:05.33 Ch3ck_ compilation even failed. Trying a fresh compile to see
15:14.32 Ch3ck_ brlcad: how communicative am I during this as concerning my writing of the unit tests for poly.c functions
15:14.42 Ch3ck_ brlcad: I mean this weeki
15:17.23 Ch3ck_ brlcad:~:-P
15:31.08 Notify 03BRL-CAD:erikgreenwald * 56421 brlcad/trunk/src/adrt/isst_tcltk.c: check that the dm was successfully opened before setting the default bg color
15:41.20 cogitokat When I run make tests after compiling, there are 4 failing tests. http://pastebin.com/DZkvF9BY
15:42.10 cogitokat Why?
15:43.38 ``Erik the NURBS- and test_ssi ones are supposed to fail (not implemented yet), which other two are you seeing? (i'm seeing 3 others on some platforms)
15:45.46 zero_level ``Erk here is the paste kde link of his errors
15:45.52 zero_level http://paste.kde.org/p33524d6e/
15:46.56 Notify 03BRL-CAD Wiki:NyahCh3ck20 * 5895 /wiki/User:NyahCh3ck20/GSoc2013/Coding_Repor: /* 29 July - 4 August */
15:47.14 ``Erik progname is due to a recent modification in the test suite, it seems to work on mac but not anywhere else... and bottess_test is build system related, ctest is trying bin/test_bottess but it's being build in src/libgcv/test_bottess
15:47.35 ``Erik they'll be fixed soonish, before the next release *shrug*
15:47.48 cogitokat Okay, thank you.
16:08.34 zero_level wq
16:31.36 *** join/#brlcad Izak_ (~Izak@195.24.220.16)
16:31.39 cogitokat I'm trying to use swig to make python bindings, and when I do the swig command it says this http://paste.kde.org/p3775a896/ This is the #if it's taliking about in db.h http://paste.kde.org/p15615e93/
16:32.42 cogitokat Why is it an error and how can I fix it?
16:42.16 *** join/#brlcad Izak_ (~Izak@195.24.220.16)
16:44.39 *** join/#brlcad Izak__ (~Izak@66-118-151-70.static.sagonet.net)
16:52.03 Notify 03BRL-CAD Wiki:IIIzzzaaakkk * 5896 /wiki/User:Izak/GSOC_2013_logs: /* Mid-term Evaluation week */
17:01.59 *** join/#brlcad kesha (~kesha@14.139.122.114)
17:04.06 Notify 03BRL-CAD:erikgreenwald * 56422 brlcad/trunk/src/rt/main.c: free the rti after raytracing is done
17:04.12 *** join/#brlcad Izak_ (~Izak@195.24.220.16)
17:04.25 *** join/#brlcad Izak___ (c318dc10@gateway/web/freenode/ip.195.24.220.16)
17:17.34 cogitokat Has anyone succesfully used swig? It seems like there are all sorts of errors in the header files that it does not like.
17:27.37 Izak__ I have not used swig before
17:41.31 Notify 03BRL-CAD:starseeker * 56423 brlcad/trunk/doc/docbook/system/mann/en/comb.xml: add an example search for flattenable objects to the comb man page.
18:18.42 Notify 03BRL-CAD:mohitdaga * 56424 brlcad/trunk/include/icv.h: Add flags in icv_image struct
18:19.20 Notify 03BRL-CAD Wiki:KeshaSShah * 5897 /wiki/User:KeshaSShah/GSoC13/Reports: /* Week 7 */
18:19.33 *** join/#brlcad kesha (~kesha@14.139.122.114)
18:19.43 Notify 03BRL-CAD Wiki:KeshaSShah * 5898 /wiki/User:KeshaSShah/GSoC13/Reports: /* August 1 */
18:30.11 Notify 03BRL-CAD:mohitdaga * 56425 (brlcad/trunk/include/icv.h brlcad/trunk/src/libicv/CMakeLists.txt): Add operations file. This will contain all the airthmatic and related routines.
18:32.34 brlcad cogitokat: they're not errors in the header, it's just we used some advanced pre-processor constructs that swig has trouble with
18:32.43 brlcad that said, "yes" is the answer to your question ;)
18:32.58 brlcad you don't want to feed swig all the headers, just ones for a particular library
18:34.34 brlcad and actually, your error in http://paste.kde.org/p3775a896/ is just a compiler-related issue
18:36.42 Notify 03BRL-CAD:brlcad * 56426 brlcad/trunk/include/db.h: we're no longer support k&r, assume stdc
18:36.45 Notify 03BRL-CAD:mohitdaga * 56427 (brlcad/trunk/include/icv.h brlcad/trunk/src/libicv/operations.c): Adding icv_add_val function. This adds a constant value to all the pixels of an image.
18:37.26 cogitokat I did a thing that fixed that error, but ended up with Error: Syntax error in input(1) for something in render.h... so should I follow the example in misc/perl for whatever header I might want?
18:40.41 brlcad your timing is ironic
18:40.57 brlcad one of the other devs (tom) has been working in misc/perl just the past few days on a swig interface
18:41.14 brlcad so yeah, you could certainly start there though that is a work in progress
18:41.21 Notify 03BRL-CAD:brlcad * 56428 brlcad/trunk/include/orle.h: assume stdc
18:41.28 cogitokat Ah, I see
18:41.57 Notify 03BRL-CAD:mohitdaga * 56429 (brlcad/trunk/include/icv.h brlcad/trunk/src/libicv/operations.c): Add routines which performs operations with contant.
18:42.44 cogitokat brlcad, is there anywhere else I can see bindings being made successfully?
18:44.28 Notify 03BRL-CAD:mohitdaga * 56430 brlcad/trunk/src/libicv/operations.c: trailing ws
18:45.04 brlcad cogitokat: I'm afraid only in Tcl and not via SWIG
18:45.29 brlcad cogitokat: do you have a goal in mind? there might be another way
18:46.56 cogitokat brlcad, the goal is to make nurbs things happen in python, using brlcad
18:47.39 brlcad of course you reach into the most advanced and active area of devleopment... :)
18:48.44 brlcad cogitokat: three thoughts come to mind
18:49.08 brlcad 1) you could try to swigify just openNURBS (src/other/openNURBS, C++)
18:50.06 brlcad 2) you could swigify our LIBGED library, which gets you access to most commands used for our scripting layer and mged
18:51.21 brlcad 3) our geometry service project (which is not yet usable) is designed to bridge language bindings through a net-centric protocol
18:52.09 Notify 03BRL-CAD:tbrowder2 * 56431 (brlcad/trunk/doc/burst/Funcs =================================================================== and 33 others): add burst doc source
18:56.11 cogitokat brlcad, I think I'll attempt to do #2.
18:57.47 Notify 03BRL-CAD:mohitdaga * 56432 (brlcad/trunk/include/icv.h brlcad/trunk/src/libicv/operations.c): Add functions which performs operations between two images.
19:00.03 cogitokat brlcad, thanks for the info and suggestions
19:07.42 Notify 03BRL-CAD:mohitdaga * 56433 (brlcad/trunk/include/icv.h brlcad/trunk/src/libicv/operations.c): Add icv_saturate function. This changes the saturation of image pixels.
19:11.14 Notify 03BRL-CAD:mohitdaga * 56434 brlcad/trunk/include/icv.h: Add ICV_EXPORT before the functions.
19:20.04 brlcad cogitokat: let me or any one of the other guys know if you need anything
19:20.27 brlcad really looking forward to someone taking that to the next level, setting up a useful interface
19:24.47 Notify 03BRL-CAD:starseeker * 56435 (brlcad/trunk/include/raytrace.h brlcad/trunk/src/libged/move_all.c): Extract the bit from mvall that does the object reference renaming for a comb tree and make it available as a function call - will need this for a comb command feature.
19:24.59 cogitokat brlcad, I sure will, does tom use irc?
19:25.49 brlcad sometimes, but he's more active on the mailing list
19:26.07 cogitokat ah
19:26.39 brlcad you're welcome to work with him there or, if his progress stagnates, take over the initiative
19:27.28 zero_level time for some stats functions after the break. And it is raining cats and dogs here :-)
19:27.50 cogitokat sounds good
19:28.42 zero_level brlcad : pixsaturate is converted to icv_satureate.
19:28.52 zero_level *icv_saturate.
19:29.05 zero_level pix_satureate contains author info
19:29.31 zero_level do you think I should bring that name too icv_saturate
19:29.41 zero_level c/pix_saturate/pixsaturate
19:29.55 zero_level c/too/to
19:33.40 Notify 03BRL-CAD:brlcad * 56436 brlcad/trunk/src/util/pixsaturate.c: clean up the description, remove duplicate info and authorship commentary
19:33.42 brlcad zero_level: there's your answer
19:35.38 zero_level alright. Thanks.
19:42.28 ``Erik what's the issue with the current swig stuff? it was able to compile generated interfaces for a couple languages when I committed it
19:47.54 cogitokat <PROTECTED>
19:55.32 Notify 03BRL-CAD:brlcad * 56437 brlcad/trunk/include/bu.h: simplify temporal comment
19:56.44 Notify 03BRL-CAD:mohitdaga * 56438 brlcad/trunk/include/icv.h: The way flags are used. They must be powers of 2.
20:07.27 Notify 03BRL-CAD:mohitdaga * 56439 (brlcad/trunk/include/icv.h brlcad/trunk/src/libicv/CMakeLists.txt): Add stat.c This will contain statistics and histogram routines. Also adding icv_hist function
20:08.34 Notify 03BRL-CAD:starseeker * 56440 brlcad/trunk/src/librt/comb/db_comb.c: Oops, helps to actually commit the definition.
20:13.00 Notify 03BRL-CAD:mohitdaga * 56441 (brlcad/trunk/include/icv.h brlcad/trunk/src/libicv/stat.c): Change the name of the function
20:14.16 Notify 03BRL-CAD:mohitdaga * 56442 brlcad/trunk/src/libicv/stat.c: Change the name of the function
20:16.45 Notify 03BRL-CAD:brlcad * 56443 brlcad/trunk/src/external/Cubit/g-sat.cpp: don't forget to update the cpp files, s/rt_g/RTG/
20:16.56 Notify 03BRL-CAD:mohitdaga * 56444 brlcad/trunk/src/libicv/stat.c: Add icv_min function.
20:18.10 Notify 03BRL-CAD Wiki:Vladbogolin * 5899 /wiki/User:Vladbogolin/GSoC2013/Logs: /* Week 7 */
20:18.31 Notify 03BRL-CAD:mohitdaga * 56445 brlcad/trunk/include/icv.h: Declare functions in the header.
20:18.58 Notify 03BRL-CAD:brlcad * 56446 (brlcad/trunk/include/raytrace.h brlcad/trunk/src/adrt/load_g.c and 93 others): rename the rt_g global to RTG so that it's more distinct that it's a global, less likely to shadow, and so that it doesn't match struct rt_g in name.
20:19.38 zero_level brlcad : what is your say about the api now ?
20:23.46 Notify 03BRL-CAD Wiki:Level zero * 5900 /wiki/User:Level_zero/GSOC13/logs: /* Week 7 */
20:24.37 Notify 03BRL-CAD:brlcad * 56447 brlcad/trunk/src/librt/globals.c: rename the actual global, rt_g is now RTG.
20:24.38 brlcad zero_level: I'll have a look later this afternoon
20:24.45 zero_level alright
20:24.52 zero_level goes to bed.
20:25.24 Notify 03BRL-CAD:mohitdaga * 56448 brlcad/trunk/src/libicv/stat.c: trailing ws
20:25.25 brlcad watching the commits, it's been looking good
20:25.37 brlcad might want to do a pass over the header file to make sure it's neat/organized
20:25.41 brlcad so it's easy to digest
20:25.45 brlcad a quick pass
20:25.46 zero_level alright.
20:26.06 zero_level Its 1:55 am here. :-)
20:26.30 zero_level stat requires comments and operations a modifcations.
20:26.45 zero_level I will try to catch them early morning tommorow.
20:27.23 brlcad sounds good
20:28.53 Notify 03BRL-CAD:brlcad * 56449 brlcad/trunk/CHANGES: rt_g is now RTG
20:37.34 *** join/#brlcad mpictor (~mark@2601:d:b280:b5:d63d:7eff:fe2d:2505)
20:44.50 Notify 03BRL-CAD:brlcad * 56450 brlcad/trunk/src/libicv/operations.c: remove debug statement (should have used %f and/or bu_log anyways).
20:57.37 Notify 03BRL-CAD:starseeker * 56451 (brlcad/trunk/doc/docbook/system/mann/en/comb.xml brlcad/trunk/src/libged/comb.c): Add initial stab at implementation of the 'lift region' option to comb (-l) that clears out all region flags below the specified combination and sets the region flag on the specified combination. If a region is used in another tree, this option uses the wrapping functionality to swap regions for combs in
20:57.39 Notify the local tree while preserving the region's properites elsewhere, except when doing so would yank the region flag out from a comb that is defining something in another tree. (basic option is described in man page, examples will be added to clarify.)
21:01.24 Notify 03BRL-CAD:brlcad * 56452 brlcad/trunk/src/conv/step/g-step.cpp: sfile is unused, removed
21:13.49 Notify 03BRL-CAD:n_reed * 56453 (brlcad/trunk/src/tclscripts/archer/Archer.tcl brlcad/trunk/src/tclscripts/archer/CMakeLists.txt): add empty brep editing frame to archer
21:20.47 Notify 03BRL-CAD:starseeker * 56454 brlcad/trunk/src/libged/comb.c: No need to search every time, just add and remove the pointers as needed.
21:26.00 Notify 03BRL-CAD:starseeker * 56455 brlcad/trunk/doc/docbook/system/mann/en/comb.xml: Add -l option to man page usage.
21:27.14 brlcad cogitokat: try commenting out that render.h line (it shouldn't even be looking at that file)
21:27.42 brlcad that might be the difference, you're running on an install tree instead of the published headers in our source include/ directory
22:12.33 ``Erik brlcad: any chance of adding %%ircname%% to new-committer.template? :D
22:13.17 ``Erik searches AUTHORS O.o
22:51.03 brlcad ``Erik: sure
22:51.12 brlcad he's ejno ;)
22:51.55 Notify 03BRL-CAD:brlcad * 56456 (brlcad/trunk/CHANGES brlcad/trunk/doc/docbook/system/man3/en/librt.xml and 136 others): the rt_functab global is now OBJ. this will eventually not matter because we'll hide the table behind an API (and then make it go away with self-registered types).

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.