IRC log for #brlcad on 20130628

00:37.16brlcad[A
00:38.08brlcadzero_level: I replies regarding your wiki page, please make sure the patches themselves are clear
00:38.27brlcadALL of your patches -- if a newer patch replaces an older patch, close the old patch (or leave a comment)
04:02.20*** join/#brlcad Samooka (~sam@187.39.188.121)
04:02.59Samookahello
04:04.34SamookaI would like to edit a combination using oed, however I am missing something as the rotation is not applied to the combination, but to one of its parts only. Can anyone help me understand this?
04:04.40Samookathis is the tree:
04:05.00Samookaforno.s/
04:05.02Samookau domo_aberto.s/
04:05.02Samookau casca_domo.s/
04:05.02Samookau solido1_domo.s
04:05.02Samooka- vazio2_domo.s
04:05.02Samooka- vazio1_domo.s
04:05.03Samooka- vazio1_tunel.s/
04:05.05Samookau solido1_tunel.s
04:05.07Samookau solido2_tunel.s
04:05.09Samookau tunel.s/
04:05.11Samookau solido3_tunel.s/
04:05.13Samookau solido1_tunel.s
04:05.15Samookau solido2_tunel.s
04:05.17Samooka- vazio2_tunel.s/
04:05.21Samookau solido1_tunel.s
04:05.23Samookau solido2_tunel.s
04:06.06brlcadSamooka: what's our oed command look like?
04:06.12Samookausing "oed forno.s domo_aberto.s/casca_domo.s/solido1_domo.s" will enter edit mode, however tunel.s is ignored: it is not selected and transformations applied to forno.s do not apply to it.
04:06.18SamookaWhat am I missing?
04:07.09brlcadso that's "oed rightpath leftpath" and it puts the matrix over the leftpath
04:07.19brlcadso if you want the matrix over forno.s, you need it on the left
04:07.34brlcadbah, and I have my right and left REVERSED!
04:07.52brlcad"oed leftpath rightpath" .. puts it over rightpath
04:08.14brlcadso you realize, /this/is/the/full/path
04:08.57brlcadso what you wrote properly formatted would have been "oed /forno.s domo_aberto.s/casca_domo.s/solido1_domo.s"
04:09.01Samookabrlcad, wasn't leftpath supposed to be the path to a primitive?
04:09.15brlcadyes, it's also that
04:09.22brlcadwhat you probably are wanting is
04:09.39brlcadoed / forno.s/domo_aberto.s/casca_domo.s/solido1_domo.s
04:10.23Samookabrlcad, so this line you just gave me will apply the matrix to forno.s?
04:10.32brlcadlast comment, your naming convention there is suboptimal ... the .s suffix implies a primitive shape (like ell or tor) but clearly many of those are not primitives
04:10.45brlcadSamooka: yep
04:10.53Samookabrlcad, right, they are combinations
04:11.09brlcadcombs usually have a .c, .r, .g or no suffix
04:11.15SamookaI guess I should do away with the .s's...
04:11.17brlcad.c is most common
04:11.33brlcadbelow the region level at least, .c is common
04:11.41brlcadthen .r for your actual regions
04:11.47brlcadand no suffix above the region level
04:13.18Samookabrlcad, it worked, thanks a lot... let me re-read your posts; why is it that the left path has to be the root (/)?
04:15.31Samookaif I understand correctly, "oed /forno.s domo_aberto.s/casca_domo.s/solido1_domo.s" tells oed that I want to modify domo_aberto.s, and not forno.s?
04:16.17brlcadit says to modify the instance of domo_aberto.s referenced within forno.s
04:16.21brlcadthere can be other references
04:16.34brlcadsuch as domo_aberto.s by itself
04:16.50Samookagot it.
04:16.59brlcadif you wanted to update all referenes to domo_aberto.s, you'd run "oed / domo_aberto.s/...."
04:17.13brlcadthink of it as ONE path like a filesystem
04:17.19brlcad/path/to/file
04:17.41brlcadand depending on what you want to manipulate, you insert a space and put oed at the front ;)
04:18.27brlcadwe will eventually get rid of the need to specify the left and right hand sides, but that's the current usage for a variety of reasons
04:19.48Samookaso everytime I want to modify a comb or region in the top level, leftpath will be (/); if I have a region composed of several combs and want to modify just one instance, then I will have leftpath point to the region (/top_level_object)
04:20.36brlcadexactamente
04:20.36SamookaI've got to remember that!!
04:20.56brlcadif we can make that any more clear in the documentation or on the wiki, let us know what to say :)
04:24.20Samookabrlcad, I read through "Object editing: the oed command" but did not get this insight. If I can find another way of explaining it, I might submit a proposal...
04:24.30brlcadgreat
04:24.45brlcadespecially considering that's the document specifically written to help explain this :)
04:25.09Samookafor now, thank you so very much for the clarification
04:25.18Samooka:)
04:25.23brlcadany time
05:20.09*** join/#brlcad caen23 (~caen23@92.81.178.221)
06:44.19Notify03BRL-CAD:phoenixyjll * 55887 brlcad/trunk/src/libbrep/intersect.cpp: According to openNURBS's declaration, for ccx_point events, m_a[0] should be equal to m_a[1] (m_b[0] == m_b[1], m_A[0] == m_A[1], m_B[0] == m_B[1]...). And change the logic so that if it's not an overlap event, two ccx_point events may be generated.
06:48.50Notify03BRL-CAD:phoenixyjll * 55888 brlcad/trunk/src/libbrep/intersect.cpp: Used a wrong curve in point-curve intersection to test overlap.
07:11.51Notify03BRL-CAD:phoenixyjll * 55889 brlcad/trunk/src/libbrep/intersect.cpp: More comment to document the intersection approaches, and style conformance.
07:29.26Notify03BRL-CAD Wiki:Phoenix * 5531 /wiki/User:Phoenix/GSoc2013/Reports: /* Week 2 */
08:09.57*** part/#brlcad Mahi (~Mahi@ec2-54-226-39-211.compute-1.amazonaws.com)
08:22.58*** join/#brlcad d_rossberg (~rossberg@66-118-151-70.static.sagonet.net)
08:25.37d_rossbergbrlcad: i thought using -fPIC is common sense
08:25.47d_rossbergsee e.g. www.akkadia.org/drepper/dsohowto.pdf
08:26.38d_rossbergand that's the error i get w/o -fPIC: /usr/bin/ld: ../../../lib/libbn.a(mat.c.o): relocation R_X86_64_32 against `.rodata' can not be used when making a shared object; recompile with -fPIC
09:03.04*** join/#brlcad caen23 (~caen23@92.81.178.221)
09:15.05*** join/#brlcad infobot (~infobot@rikers.org)
09:15.06*** topic/#brlcad is BRL-CAD || http://brlcad.org || logs: http://ibot.rikers.org/%23brlcad/ || GSoC 2013! http://brlcad.org/wiki/Google_Summer_of_Code
10:27.22*** join/#brlcad kesha (~kesha@49.249.1.165)
10:48.01*** join/#brlcad kesha (~kesha@49.249.1.165)
12:04.02``Erikhttp://www.youtube.com/watch?v=WV4qXzM641o  raytracing in '78
12:04.22``Erikhttp://www.youtube.com/watch?v=lAYaX6NuI4M  MAGI synthavision in '80 :o
12:48.23*** join/#brlcad Yoshi47 (~jan@64.235.102.210)
13:25.32starseekersweet!  scintilla has some built-in Qt support now, not limited to the GPL 3rd party add-on
13:26.12starseekerdreams of replacing all the "use external editor" commands with a BRL-CAD ascii format aware Qt editing widget...
13:34.22starseekerhah, don't even need to wait for Qt:  http://sourceforge.net/projects/scintillatk/
13:55.36``Erikdid an ascii aware text widget long ago for a gtk+ mud client... handling the codes was easy, making it fast was not
14:02.33zero_level``Erik I found ap->a_color. but this get converted to uchar using two methods 1.(GAMMA CORRECTION) 2. Straight method with addition of random number(BW 0to 1).
14:03.19zero_levelwhen i write this to ICV image how do i procced with this
14:08.33``Eriksince some image formats cope with gamma correction, I'd imagine a good approach would be to add a gamma to the internal representation, then apply it when converting to the external format... as a very rough C like psuedocode with important bookkeeping bits missing: save_pix() { char *bytes; bytes=bu_alloc(img.w*img.h*3*sizeof(uchar); apply_gamma(img); img2rgb24(img, bytes); fwrite(fd, bytes, img.w*img.h*3*sizeof(uchar)); }
14:12.22*** join/#brlcad cstirk (~quassel@96.255.19.39)
14:12.22``Erik(and save_png would lack the apply_gamma() call, it would just set the gamma using the png api)
14:15.58``Erikdoes that make sense?
14:18.38zero_leveland i add gamma to internal structure of ICV_IMAGE struct ?
14:19.46``Erikyeh
14:19.53zero_levelok.
14:21.49``Erikicv_image should contain enough information to generate any format without any extra information... then the output functions will just need the icv_image struct and a place to put the output...
14:23.50``Erikdaydreams of an rt mode that uses icv to fill frames for ffmpeg to make animations directly :)
14:34.48*** join/#brlcad zero_level (~mohit@66-118-151-70.static.sagonet.net)
15:06.36Notify03BRL-CAD:carlmoore * 55890 brlcad/trunk/src/fb/fbfade.c: fix a comment because -h became -H
15:22.38Notify03BRL-CAD:carlmoore * 55891 brlcad/trunk/HACKING: make repairs to SoureForge and DocBook names, and fix a spelling
15:46.13Notify03BRL-CAD:carlmoore * 55892 (brlcad/trunk/CMakeLists.txt brlcad/trunk/ChangeLog): remove trailing blanks/tabs
16:30.16zero_level``Erik how does bif work in rt/view.c i dont see an icv_image_save_open does it use some extern function ?
16:39.30Notify03BRL-CAD Wiki:Level zero * 5532 /wiki/User:Level_zero/GSOC13/logs: /* WEEK 2 */
16:40.38Notify03BRL-CAD Wiki:Level zero * 5533 /wiki/User:Level_zero/GSOC13/logs: /* WEEK 2 */
16:41.54``Erikit's in do.c
16:48.03starseeker``Erik: I'm hoping that scintilla would help with the speed part
16:48.23zero_level``Erik see brlcad.org/wiki/User:Level_zero/GSOC13/logs I have put links to the current status of the code
17:28.03Notify03BRL-CAD Wiki:Harman052 * 5534 /wiki/User:Harman052/GSoc2013/Logs:
18:05.42*** join/#brlcad kanzure (~kanzure@131.252.130.248)
18:24.18*** join/#brlcad caen23 (~caen23@92.81.178.221)
18:46.28Notify03BRL-CAD Wiki:NyahCh3ck20 * 5535 /wiki/User:NyahCh3ck20/GSoc2013/Coding_Repor: /* 24 June - 30 June */
19:16.34Notify03BRL-CAD Wiki:NyahCh3ck20 * 5536 /wiki/User:NyahCh3ck20/GSoc2013/Coding_Repor: /* 24 June - 30 June */
19:50.00``Erikvery cool... (if alf aho and jeff ullman sound familiar, they're the dudes that did the "dragon book" on compilers) http://i.stanford.edu/~ullman/focs.html
19:57.34Notify03BRL-CAD:carlmoore * 55893 brlcad/trunk/doc/docbook/system/man1/en/fblabel.xml: new man page for fblabel
19:58.30*** join/#brlcad Ch3ck (295cd22a@gateway/web/freenode/ip.41.92.210.42)
20:00.26Ch3ckhey guys working on the code for adding orthogonal matrix support to the bn_mat_inverse() routine in /src/libbn/mat.
20:00.28Ch3ckmat.c
20:00.50Ch3ckand although the code compiles it don't still get why it fails 3 regression tests..
20:02.38Ch3ckneed some more eyes on this...
20:10.35zero_level``Erik the issue with write pixel is clear now. But viewedge.c and view.c also writesline in the image. They use scanline(struct in view.c and char buff in viewedge.c)
20:10.50zero_leveli am not getting hold of double data in these case.
20:17.45``Erikview.c uses a_color to construct the scanlines, view_edge.c keeps a 'scanline buffer' and compares neighbors to generate pixel values for the scanlines... (viewedge is also a bit weak in accomplishing it's task)
20:24.21Notify03BRL-CAD Wiki:41.92.210.42 * 5537 /wiki/User:Izak/GSOC_2013_logs: /* From June 24th to June 28th */
20:39.44brlcad``Erik: in issac's patch, he changed it to "hrt?" .. not "her?"
20:40.19brlcadso his latest does "match" .. just no longer the clever 'herz' value
20:53.10Notify03BRL-CAD:r_weiss * 55894 (brlcad/trunk/include/common.h brlcad/trunk/src/libbrep/libbrep_brep_tools.cpp brlcad/trunk/src/libbrep/libbrep_brep_tools.h): Fixed some DLL errors in the Windows build.
21:07.37``Erikyeh, I was more concerned that he change it and said "does this match?" instead of having some confidence in what he did... so I'm trying to push an "if you're not sure, try it and see"
21:10.54brlcadstarseeker: http://www.dune-project.org/
21:11.12brlcad``Erik: yeah, I noticed that as well
21:16.34brlcadnot license compatible, but interesting nonetheless
21:34.51*** join/#brlcad mpictor (~mark@2601:d:b280:9:d63d:7eff:fe2d:2505)
21:43.18Notify03BRL-CAD:carlmoore * 55895 brlcad/trunk/src/fb/fbline.c: remove -h (use -S 1024 instead); implement h,? help; notice that S is a new option
21:57.39zero_level``Erik: are there functions in different files which modify scanlines ?
22:00.42zero_levelbecause all i dont find the scanlines getting the data from double ap->a_color
22:18.52zero_levelalthough pixels are easy to trace from ap->a_color

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