| 00:00.46 | CIA-53 | BRL-CAD: 03starseeker * r42671 10/brlcad/branches/cmake/ (6 files in 3 dirs): Update cmake to trunk r42670 |
| 00:28.22 | CIA-53 | BRL-CAD: 03starseeker * r42672 10/brlcad/branches/cmake/ (3 files in 3 dirs): sigh. jove is explicitly referenced as the editor of last resort in other code, so go ahead and build it. wonder if this actually works on Windows... |
| 00:40.23 | CIA-53 | BRL-CAD: 03starseeker * r42673 10/brlcad/trunk/src/other/libz/zconf.h.cmakein: add the fix from 41902 to zconf.h.cmakein too |
| 00:48.45 | yukonbob_ | starseeker: re: jove -- in case of ted, it says it'll run "ed" by default. |
| 00:48.59 | yukonbob_ | <PROTECTED> |
| 00:49.21 | starseeker | nods |
| 00:52.57 | CIA-53 | BRL-CAD: 03starseeker * r42674 10/brlcad/trunk/src/other/step/src/ (7 files in 3 dirs): Hmm - warning flags got passed to step, which isn't happy - maybe could ignore, but since we're maintaining the code anyway might as well... |
| 01:00.19 | CIA-53 | BRL-CAD: 03starseeker * r42675 10/brlcad/branches/cmake/ (11 files in 6 dirs): |
| 01:00.19 | CIA-53 | BRL-CAD: OK, this should bring the compiler flags logic pretty darn close to the latest |
| 01:00.19 | CIA-53 | BRL-CAD: autotools, with this possible exception of not isolating the src/other dir quite |
| 01:00.19 | CIA-53 | BRL-CAD: well enough - on the other hand, may be OK since it built successfully on gentoo |
| 01:00.19 | CIA-53 | BRL-CAD: with these settings. |
| 01:22.56 | CIA-53 | BRL-CAD: 03starseeker * r42676 10/brlcad/trunk/src/mged/dm-rtgl.c: interp->INTERP in rtgl |
| 01:23.35 | CIA-53 | BRL-CAD: 03starseeker * r42677 10/brlcad/branches/cmake/ (CMakeLists.txt src/mged/CMakeLists.txt src/mged/dm-rtgl.c): Allow the enabling of RTGL, although things look to be a tad broken at the moment - need to check trunk. |
| 01:49.11 | CIA-53 | BRL-CAD: 03starseeker * r42678 10/brlcad/trunk/src/libdm/dm-rtgl.c: (log message trimmed) |
| 01:49.11 | CIA-53 | BRL-CAD: This gets past the initial bu_malloc error, but we've got some kind of weirdness |
| 01:49.11 | CIA-53 | BRL-CAD: going on here. The RT_CK_SEGS test in recordHit fails with what looks like a |
| 01:49.11 | CIA-53 | BRL-CAD: bad magic error - printing out the seg list in debug shows a lot of strange |
| 01:49.11 | CIA-53 | BRL-CAD: looking numbers, and commenting out the check shows a lot of one hit reports and |
| 01:49.12 | CIA-53 | BRL-CAD: a visible rtgl raytrace (e.g. aside from the errors it largely succeeds.) rtgl |
| 01:49.13 | CIA-53 | BRL-CAD: code hasn't changed in quite a while, so something must have changed out from |
| 01:50.47 | CIA-53 | BRL-CAD: 03starseeker * r42679 10/brlcad/branches/cmake/src/libdm/dm-rtgl.c: Might as well sync this from trunk... |
| 01:52.55 | CIA-53 | BRL-CAD: 03starseeker * r42680 10/brlcad/branches/cmake/TODO.cmake: Getting down there - add note on opensolaris |
| 01:59.08 | CIA-53 | BRL-CAD: 03starseeker * r42681 10/brlcad/branches/cmake/TODO.cmake: add note about library versions in src/other builds |
| 02:02.13 | CIA-53 | BRL-CAD: 03starseeker * r42682 10/brlcad/branches/cmake/TODO.cmake: note wish exe on Windows needs work |
| 02:07.20 | brlcad | starseeker: you threw in a new logic scope, but didn't fix the indentation in dm-rtgl.c |
| 02:07.31 | starseeker | oh, sorry |
| 02:07.33 | starseeker | fixes |
| 02:07.43 | brlcad | and if( isn't the right style |
| 02:08.24 | brlcad | otherwise, that looks like a good fix for a category of alloc failures |
| 02:12.34 | CIA-53 | BRL-CAD: 03starseeker * r42683 10/brlcad/trunk/src/libdm/dm-rtgl.c: ws, indent |
| 02:13.13 | brlcad | wow, that hit a lot more than expected |
| 02:14.31 | brlcad | er, yeah.. that's not right |
| 02:15.15 | brlcad | starseeker: how'd you auto-indent that? |
| 02:15.24 | starseeker | indent.sh and ws.sh |
| 02:15.35 | brlcad | huh, it looks like it's using 2-char indent |
| 02:16.19 | brlcad | odd, indent.sh here reverts your changes |
| 02:16.29 | starseeker | blinks |
| 02:16.33 | brlcad | you might have something in your .emacs |
| 02:16.40 | starseeker | or .vimrc |
| 02:16.52 | starseeker | go ahead and commit - I won't argue |
| 02:17.10 | starseeker | is more concerned about what's happening with the rtgl raytrace |
| 02:18.11 | brlcad | try running indent.sh on vers.c |
| 02:18.16 | brlcad | does it modify the file? |
| 02:18.52 | starseeker | yes - just the return brlcad_ident line |
| 02:19.10 | starseeker | which is the only one there, of course... |
| 02:19.18 | brlcad | svn revert vers.c, move your .emacs out of the way, and retry |
| 02:19.40 | brlcad | it shouldn't modify |
| 02:20.12 | starseeker | still did |
| 02:20.29 | brlcad | what platform are you on? |
| 02:20.34 | starseeker | gentoo |
| 02:20.47 | brlcad | emacs --version |
| 02:20.54 | CIA-53 | BRL-CAD: 03brlcad * r42684 10/brlcad/trunk/src/libdm/dm-rtgl.c: revert and fix formatting. |
| 02:20.55 | starseeker | GNU Emacs 23.2.1 |
| 02:22.34 | brlcad | hm, okay -- so the indent.sh is unusable by you for some reason -- is there any clue in the output when you run it? some warning or error? |
| 02:26.10 | starseeker | no - just "Loading vc-svn..." |
| 02:26.20 | starseeker | Indenting region... |
| 02:26.23 | starseeker | Indenting region... done |
| 02:26.30 | brlcad | hm, darn |
| 02:26.31 | starseeker | saving and wrote lines |
| 02:26.35 | brlcad | suspect something in misc/batch-indent-region.el is getting interpreted differently causing the variable block to be ignored |
| 02:26.51 | brlcad | well, so it's just unusable for you until debugged |
| 02:26.57 | starseeker | nods |
| 02:27.16 | brlcad | you'll have to stick with manually indenting or vim (the vi-line mode should be respected if you turn it on) |
| 02:27.42 | starseeker | that's the mode name? vi-line? |
| 02:27.47 | brlcad | only an issue because that last edit is a prime example that can lead to a logic bug later |
| 02:28.19 | brlcad | no, it's not |
| 02:31.24 | starseeker | hmm - my default vim setup doesn't work gg=G indents everything wrong |
| 02:36.43 | brlcad | :help 'modeline' |
| 02:36.59 | brlcad | turning that on should make it respect the modeline at the bottom of every file |
| 02:37.32 | brlcad | otherwise, you can set the style in your .vimrc with rules similar to this: http://drupal.org/node/29325 |
| 02:37.52 | brlcad | except it's shiftwidth=4 tabstop=8 |
| 02:48.42 | starseeker | the modeline gets close |
| 03:13.49 | brlcad | there's probably more variables that could/should be set for vim users |
| 03:14.02 | brlcad | but the two there are the critical ones for consistent indentation |
| 03:14.10 | starseeker | nods |
| 03:14.18 | brlcad | the stylistic ones were left as an exercise to the reader :) |
| 03:14.26 | starseeker | heh |
| 03:14.33 | brlcad | continues working on rt_db_corrupt() |
| 03:16.19 | CIA-53 | BRL-CAD: 03starseeker * r42685 10/brlcad/branches/cmake/ (3 files in 3 dirs): Pick up sunmath if it's there for tcl/tk |
| 03:20.35 | starseeker | yow |
| 03:20.44 | starseeker | opennurbs and sun studio don't get along at all |
| 03:22.03 | starseeker | hah - actually, that's the same issue clang saw |
| 03:22.04 | starseeker | cool |
| 03:45.39 | *** join/#brlcad yukonbob (~bch@S0106002129e399fc.ok.shawcable.net) |
| 03:45.42 | yukonbob | oh hai. |
| 03:46.38 | yukonbob | q: re problems w/ sed/ted in 7.18.0 (apparently fixed in 7.18.1) -- does anybody remember roughly (or better, acutely and accurately) what the fix was? |
| 04:03.05 | starseeker | needed to test for TCL_OK, as opposed to just using the return val for the if statement (IIRC) |
| 04:03.26 | starseeker | return val of editit in tedit.c I believe, but not sure |
| 04:04.10 | starseeker | yukonbob: recommend setting up gdb, breaking just before the edit is to be applied to disk (as a start) and tracing back where the failure is |
| 04:04.20 | starseeker | grrrr |
| 04:04.52 | starseeker | sun stuido gives this error: "The function "_finite" must have a prototype" |
| 04:05.00 | starseeker | what on earth... |
| 04:05.04 | yukonbob | starseeker: thanks for the clues... I was going to try to get away w/ least-invasive patch possible. |
| 04:14.55 | CIA-53 | BRL-CAD: 03starseeker * r42686 10/brlcad/branches/cmake/src/libdm/dm-rtgl.c: grab dm-rtgl ws/indent fixes |
| 04:37.40 | CIA-53 | BRL-CAD: 03starseeker * r42687 10/brlcad/branches/cmake/src/other/openNURBS/ (4 files): These are minimal 'get opennurbs compiling on sun studio' hacks - commiting them because I accidently erased one last go-around, but need cleanup and conditional wrappers |
| 04:53.11 | brlcad | starseeker: sun studio is notorious for having drastically different compilation behavior when you change standard compliance levels |
| 04:54.00 | brlcad | the default does not match gcc behavior but there is a mode that does match pretty closely iirc |
| 05:06.11 | yukonbob | sees some interesting header changes in bwish -- stripping itk.h and replacing w/ tk.h alone? |
| 05:07.46 | yukonbob | crosses fingers, configs 7.18.1 |
| 05:08.15 | *** join/#brlcad Stattrav (~Stattrav@122.172.46.55) |
| 05:08.16 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 05:08.43 | yukonbob | feh -- immediate 'make' failure :P |
| 05:09.16 | yukonbob | missing rtprivate.h |
| 05:26.41 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) |
| 05:36.45 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 06:00.19 | *** join/#brlcad R0b0t1 (~Enigma@64-136-219-55.dyn.everestkc.net) |
| 06:00.19 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) |
| 06:00.52 | *** join/#brlcad yukonbob (~bch@S0106001cf044d085.ok.shawcable.net) |
| 06:49.23 | yukonbob | starseeker: that TCL_OK on editit() was the problem/fix. Thx. |
| 07:15.24 | yukonbob | is there such a process as a lathe yet, like povray? |
| 07:17.17 | brlcad | yukonbob: yes, though much more general (and much less tested) :) |
| 07:17.24 | brlcad | 'revolve' primitive |
| 07:17.32 | yukonbob | brlcad: hello :) |
| 07:17.37 | brlcad | takes a 2D 'sketch' object, and an axis of rotation |
| 07:17.47 | yukonbob | how old is revolve? |
| 07:17.54 | brlcad | few years |
| 07:17.55 | yukonbob | <2years |
| 07:17.57 | yukonbob | ? |
| 07:18.08 | yukonbob | cool |
| 07:18.32 | yukonbob | brlcad: how're things? |
| 07:18.33 | brlcad | r31599 | pacman87 | 2008-06-24 13:34:05 -0400 (Tue, 24 Jun 2008) | 1 line |
| 07:18.33 | brlcad | beginning of revolve primitive, with shot() algorithm for straigh line sketches (untested) |
| 07:19.17 | yukonbob | that makes sense -- probabably ~the time I became less involved... |
| 07:19.29 | brlcad | http://brlcad.org/wiki/Revolve_Primitive |
| 07:19.43 | yukonbob | povray "lathe" operation is quite nice -- looking forward to playing w/ revolve... |
| 07:20.52 | brlcad | http://brlcad.org/tmp/revolve/ has a sample |
| 07:21.01 | yukonbob | brlcad: you see the ted bug that shipped w/ 7.18.0 :P |
| 07:21.11 | brlcad | yep |
| 07:21.16 | brlcad | what about it? |
| 07:21.42 | yukonbob | ted doesn't work :P |
| 07:21.45 | brlcad | ted isn't best practice |
| 07:21.54 | brlcad | more of a crutch |
| 07:22.03 | yukonbob | the return code is incorrectly checked, and the edits are effectively thrown away. |
| 07:22.30 | brlcad | I'm aware of what the bug is and how it was fixed -- I helped diagnose and fix when it was reported |
| 07:22.32 | yukonbob | what do the cool kids use instead of ted? |
| 07:23.09 | brlcad | it depends on the type of edit, there are different commands for different operations |
| 07:23.22 | brlcad | most of the common ones are listed on the quick ref card |
| 07:23.49 | yukonbob | will have to get to a printer and print all these again... I used to carry them with me ;) |
| 07:23.56 | brlcad | e.g., sca, rot, tra |
| 07:24.45 | yukonbob | scale, rotate, translate, sure... will those let one redefine the second radius of a torus, though? |
| 07:24.55 | brlcad | the only tricky edits are the refined parameter edits displaed on the Edit menu when you go into edit mode, but even those are selectable on the command line using the "press" command |
| 07:25.10 | yukonbob | ah |
| 07:26.24 | yukonbob | why is ted considered !Best Practice ? |
| 07:26.52 | brlcad | press "Set Radius 2" |
| 07:26.58 | brlcad | for that specific action |
| 07:27.04 | brlcad | p 10 |
| 07:27.24 | brlcad | ted basically punts on providing a user interface, kicking off a dump to a text file |
| 07:27.54 | yukonbob | so is just non-elegant, in at least one way... |
| 07:28.01 | brlcad | that's fine for some things, like diagnostics, but a really crappy and supremely error-prone way to go about things |
| 07:28.30 | brlcad | the errors it causes tend to far outweigh the warm fuzzy feeling some people get working in their favorite text editor |
| 07:28.46 | brlcad | it's a crutch |
| 07:29.03 | yukonbob | nods |
| 07:29.39 | yukonbob | anyway, is broken crutch in latest .tgz; I've got it working here, now, but will train self to use alternatives. |
| 07:34.15 | brlcad | wow, my corruption detection routine is actually working... woot! |
| 07:59.48 | *** join/#brlcad packrat (~packrator@99-67-225-40.lightspeed.livnmi.sbcglobal.net) |
| 08:51.07 | CIA-53 | BRL-CAD: 03brlcad * r42688 10/brlcad/trunk/src/librt/ (Makefile.am db_corrupt.c): |
| 08:51.07 | CIA-53 | BRL-CAD: add an initial working implementation of a routine to help detect when a v4 |
| 08:51.07 | CIA-53 | BRL-CAD: geometry file seems corrupt due to endianness. this walks over all matrices of |
| 08:51.07 | CIA-53 | BRL-CAD: combinatino members and tests whether they preserve axis perpendicularity and |
| 08:51.07 | CIA-53 | BRL-CAD: that scaling/rotation elements are unitized. if a matrix fails, it flips it and |
| 08:51.08 | CIA-53 | BRL-CAD: tries again to see if flipping fixes the problem. if ALL failures were |
| 08:51.09 | CIA-53 | BRL-CAD: successfully fixed by flipping, then true is returned. |
| 09:00.12 | CIA-53 | BRL-CAD: 03brlcad * r42689 10/brlcad/trunk/include/raytrace.h: provide declaration and documentation for rt_db_flip_endian(), to be used for detecting binary-incompatible v4 geometry database files. |
| 10:38.42 | *** join/#brlcad epileg (~epileg@unaffiliated/epileg) |
| 12:18.19 | epileg | brlcad: hello. I've made the needed changes in /misc/debian folder and in sh/make_deb.sh file, to make it working. Now I've to commit |
| 13:31.05 | CIA-53 | BRL-CAD: 03starseeker * r42690 10/brlcad/branches/cmake/src/other/openNURBS/ (4 files): Wrap sun studio stuff in conditionals |
| 13:41.47 | CIA-53 | BRL-CAD: 03starseeker * r42691 10/brlcad/branches/cmake/src/other/openNURBS/ (CMakeLists.txt opennurbs_system.h): Whoops - opennurbs build isn't using the CONFIG_H_FILE at the moment, so add definitions explicitly. Perhaps this is why ieeefp.h wasn't getting picked up, so comment out the define - if this works remove it. |
| 13:44.50 | CIA-53 | BRL-CAD: 03starseeker * r42692 10/brlcad/branches/cmake/src/other/openNURBS/opennurbs_system.h: Nope, HAVE_IEEEFP_H wasn't enough |
| 14:52.45 | brlcad | epileg: okay, go for it :) |
| 14:53.13 | brlcad | you didn't have to wait until you were done, you could/can commit incrementally |
| 14:54.00 | epileg | brlcad: Do have I commit acces to trunk/sh/ folder? |
| 15:34.43 | *** join/#brlcad epileg (~epileg@unaffiliated/epileg) |
| 15:39.47 | CIA-53 | BRL-CAD: 03brlcad * r42693 10/brlcad/trunk/src/libbu/mappedfile.c: checking the wrong debug flag |
| 15:51.07 | CIA-53 | BRL-CAD: 03brlcad * r42694 10/brlcad/trunk/include/bu.h: BU_DEBUG_DB is no longer used, mark as such to free up the slot. |
| 15:52.22 | CIA-53 | BRL-CAD: 03brlcad * r42695 10/brlcad/trunk/include/raytrace.h: mark the private db_i members as such more explicitly, ws consistency migration to keep names associated with their type for db_i and directory structs. |
| 15:57.15 | CIA-53 | BRL-CAD: 03starseeker * r42696 10/brlcad/branches/cmake/src/other/CMakeLists.txt: Don't mess with jove on Windows |
| 15:59.02 | CIA-53 | BRL-CAD: 03starseeker * r42697 10/brlcad/branches/cmake/src/other/CMakeLists.txt: Say something about it instead of turning off silently |
| 16:06.35 | *** join/#brlcad epileg (~epileg@unaffiliated/epileg) |
| 16:06.59 | brlcad | epileg: the only way to know that for sure is to try, right? |
| 16:07.12 | epileg | right :-) |
| 16:07.40 | epileg | about the commit comment. some rule to follow? |
| 16:08.09 | brlcad | sure, make it something useful to the other developers and to yourself 5 years from now |
| 16:09.01 | brlcad | briefly describing what *and* why/impact/reasoning |
| 16:22.02 | CIA-53 | BRL-CAD: 03starseeker * r42698 10/brlcad/branches/cmake/src/bwish/CMakeLists.txt: tweak bwish build logic |
| 16:24.13 | CIA-53 | BRL-CAD: 03brlcad * r42699 10/brlcad/trunk/src/librt/namegen.c: lots of dead code, mostly dead code |
| 16:26.43 | CIA-53 | BRL-CAD: 03brlcad * r42700 10/brlcad/trunk/src/librt/db5_scan.c: |
| 16:26.43 | CIA-53 | BRL-CAD: change db_dirbuild() to no longer directly check the file header and calculate |
| 16:26.43 | CIA-53 | BRL-CAD: the version. let db_version() be the (only) place that happens. should reduce |
| 16:26.43 | CIA-53 | BRL-CAD: a few petty file i/o operations but more importantly avoid overriding |
| 16:26.44 | CIA-53 | BRL-CAD: dbi_version since it needs to be negative for flipped v4 files. |
| 16:28.19 | CIA-53 | BRL-CAD: 03brlcad * r42701 10/brlcad/trunk/ (3 files in 2 dirs): enable db_corrupt for librt compilation |
| 16:31.31 | CIA-53 | BRL-CAD: 03brlcad * r42702 10/brlcad/trunk/src/librt/db5_scan.c: right, 'header' is no longer used. |
| 16:37.11 | CIA-53 | BRL-CAD: 03starseeker * r42703 10/brlcad/branches/cmake/ (10 files in 10 dirs): Add version number logic for some of the src/other libs |
| 16:47.31 | CIA-53 | BRL-CAD: 03brlcad * r42704 10/brlcad/trunk/src/fbserv/fbserv.c: return from main instead of exiting |
| 16:49.20 | CIA-53 | BRL-CAD: 03brlcad * r42705 10/brlcad/trunk/src/other/tnt/tnt_fortran_array3d.h: quell warning about having no space on the empty for loop before the semicolon. |
| 17:09.51 | *** join/#brlcad Yoshi47 (~jan@64.235.102.210) |
| 18:21.39 | CIA-53 | BRL-CAD: 03starseeker * r42706 10/brlcad/branches/cmake/src/other/ (tcl/CMakeLists.txt tk/CMakeLists.txt): Hmm endian test isn't happy with VC2010 |
| 18:27.59 | CIA-53 | BRL-CAD: 03starseeker * r42707 10/brlcad/branches/cmake/src/other/incrTcl/ (itcl/CMakeLists.txt itk/CMakeLists.txt): itcl/itk too |
| 18:28.30 | CIA-53 | BRL-CAD: 03erikgreenwald * r42708 10/brlcad/branches/cmake/src/libpc/CMakeLists.txt: add TCL_INCLUDE_DIRS to make bu.h happy |
| 18:36.05 | CIA-53 | BRL-CAD: 03starseeker * r42709 10/brlcad/branches/cmake/misc/CMake/CompilerFlags.cmake: Hmm - was this the issue with the endian test? |
| 18:47.04 | starseeker | hmm. nope |
| 18:47.10 | starseeker | does fix another problem though |
| 18:48.09 | starseeker | sees other indications that TestBigEndian may be quirky with VC2010... well, not likely to be needed anyhow with MSVC |
| 19:18.20 | CIA-53 | BRL-CAD: 03starseeker * r42710 10/brlcad/branches/cmake/misc/CMake/CompilerFlags.cmake: More flag tweaking... |
| 19:55.49 | CIA-53 | BRL-CAD: 03starseeker * r42711 10/brlcad/branches/cmake/misc/CMake/test_srcs/report_username.c.in: Tweak WIN32 conditional define to _WIN32 |
| 20:06.53 | CIA-53 | BRL-CAD: 03bob1961 * r42712 10/brlcad/trunk/ (include/tclcad.h src/libtclcad/ged_obj.c): Added a callback for when the view changes to libtclcad. |
| 20:13.13 | CIA-53 | BRL-CAD: 03starseeker * r42713 10/brlcad/branches/cmake/src/other/step/ (11 files in 6 dirs): Simplify the SCL CMake logic, grab some of the updates from newer BRL-CAD logic |
| 20:14.43 | CIA-53 | BRL-CAD: 03bob1961 * r42714 10/brlcad/trunk/src/tclscripts/lib/Ged.tcl: Added methods to expose libtclcad's view_callback function. |
| 20:18.38 | CIA-53 | BRL-CAD: 03starseeker * r42715 10/brlcad/branches/cmake/include/config_win_cmake.h: This is defined in newer MSVC compilers - conditionalize to quell warnings |
| 20:29.05 | CIA-53 | BRL-CAD: 03starseeker * r42716 10/brlcad/trunk/src/libbu/backtrace.c: Let's not do the assignment in the if statement - quell MSVC warning |
| 20:29.38 | CIA-53 | BRL-CAD: 03starseeker * r42717 10/brlcad/trunk/src/libbu/backtrace.c: typo |
| 20:32.20 | CIA-53 | BRL-CAD: 03brlcad * r42718 10/brlcad/trunk/src/tclscripts/mged/text.tcl: can't just directly call 'bind' since procedures might be called without ::tk:: namespace loaded. catch any error. |
| 20:34.11 | CIA-53 | BRL-CAD: 03brlcad * r42719 10/brlcad/trunk/src/tclscripts/mged/dbupgrade.tcl: make dbupgrade work once again in classic console mode where there is no ::tk |
| 20:36.10 | CIA-53 | BRL-CAD: 03starseeker * r42720 10/brlcad/branches/cmake/src/other/CMakeLists.txt: try adding xlib to TCL_INCLUDE_DIRS (untested) |
| 20:38.19 | CIA-53 | BRL-CAD: 03starseeker * r42721 10/brlcad/branches/cmake/src/other/CMakeLists.txt: typo |
| 20:44.52 | CIA-53 | BRL-CAD: 03bob1961 * r42722 10/brlcad/trunk/src/tclscripts/archer/CombEditFrame.tcl: Fixed CombEditFrame::updateGeometry. It's now able to turn off inherit. |
| 20:52.06 | CIA-53 | BRL-CAD: 03brlcad * r42723 10/brlcad/trunk/src/conv/asc/g2asc.c: check ==0 instead of ==DB5_MINORTYPE_RESERVED since the value may change. zero suffices for unset. |
| 20:53.03 | CIA-53 | BRL-CAD: 03brlcad * r42724 10/brlcad/trunk/src/conv/dbupgrade.c: |
| 20:53.04 | CIA-53 | BRL-CAD: dbupgrade with a coredump flag makes the automatic endianness flip test fail |
| 20:53.04 | CIA-53 | BRL-CAD: since low-level libbn routines will bomb if they encounter a bad matrix. |
| 20:53.04 | CIA-53 | BRL-CAD: instead of bombing, let them return their error so we upgrade cleanly. |
| 20:53.43 | CIA-53 | BRL-CAD: 03brlcad * r42725 10/brlcad/trunk/src/tclscripts/mged/help.tcl: improve help, dbupgrade will begin processing if it's followed by 'upgrade' |
| 21:00.48 | CIA-53 | BRL-CAD: 03brlcad * r42726 10/brlcad/trunk/src/tclscripts/mged/help.tcl: list all three upgrade commands |
| 21:01.38 | CIA-53 | BRL-CAD: 03brlcad * r42727 10/brlcad/trunk/src/tclscripts/mged/dbupgrade.tcl: display the entire detailed help if the user requests help. 'help dbupgrade' will still just report a brief usage message. |
| 21:03.17 | CIA-53 | BRL-CAD: 03jordisayol * r42728 10/brlcad/trunk/ (31 files in 2 dirs): Upgraded the debian package build proccess. Added mged, archer, rtwizard, documentation and examples menus. Added brlcad mime type. Added brlcad mime file association. |
| 21:09.13 | CIA-53 | BRL-CAD: 03starseeker * r42729 10/brlcad/branches/cmake/src/ (CMakeLists.txt irprep/subroutines.c other/CMakeLists.txt): more Windows stuff... |
| 21:10.44 | CIA-53 | BRL-CAD: 03brlcad * r42730 10/brlcad/trunk/src/tclscripts/mged/dbupgrade.tcl: support all variations of 'yes' and inform the user that upgrade was cancelled and database is being reopened. this way, if it's a corrupt database, the warning makes sense. |
| 21:10.56 | brlcad | epileg: congratulations :) |
| 21:11.06 | epileg | thanks brlcad |
| 21:11.29 | epileg | but it do not compile on trunk |
| 21:11.46 | epileg | in 7.18.0 compile without problem |
| 21:12.30 | brlcad | you have to be way more specific for that to mean anything |
| 21:13.06 | epileg | ok, just a moment |
| 21:13.12 | brlcad | there are dozens or even hundreds of commits nearly every day, and they won't all work -- that's the nature of working on trunk |
| 21:14.08 | brlcad | first step is to always check and see if there's an update, read the recent commits from in here to see if the commit message says anything about breaking things |
| 21:15.02 | brlcad | then just read the failure to see what happened, because 90% of the time, it's a very simple syntax mistake that can be easily fixed by anyone |
| 21:15.38 | brlcad | then, if you can't make sense of the error, you can paste your build log somewhere and ask for help in here :) |
| 21:17.20 | brlcad | epileg: you also need to update misc/Makefile.am to list all of the file changes you made |
| 21:17.30 | brlcad | additions and deletions |
| 21:17.34 | brlcad | it's a simple list |
| 21:17.53 | epileg | ok, I'll do right now |
| 21:21.33 | CIA-53 | BRL-CAD: 03brlcad * r42731 10/brlcad/trunk/src/librt/db_open.c: |
| 21:21.33 | CIA-53 | BRL-CAD: enable automatic flipping of a binary-incompatible v4 geometry database if |
| 21:21.33 | CIA-53 | BRL-CAD: rt_db_flip_endian() returns true. this should only occur if flipping the |
| 21:21.33 | CIA-53 | BRL-CAD: database was observed to fix ALL matrix member failures during a quick db_open() |
| 21:21.33 | CIA-53 | BRL-CAD: scan of the file. the user should be able to force flipped values in mged with |
| 21:21.34 | CIA-53 | BRL-CAD: 'opendb -f'. |
| 21:26.33 | brlcad | epileg: you can check a build's suitability for distribution, particularly whether you missed any files, with "make distcheck" |
| 21:28.16 | epileg | ok |
| 21:28.55 | epileg | now i'm modifying misc/Makefile.am |
| 21:29.35 | brlcad | cool |
| 21:31.14 | brlcad | epileg: can you describe what all the changes were -- new menus, mime type file association, desktop icons, ... anything else? |
| 21:33.33 | epileg | brlcad: where i've to describe this? |
| 21:34.04 | brlcad | heh, here -- just asking for general info |
| 21:34.11 | epileg | ah |
| 21:34.32 | CIA-53 | BRL-CAD: 03starseeker * r42732 10/brlcad/branches/cmake/misc/CMakeLists.txt: whoopsie - to run in the build dir, need archer.ico in the build dir... |
| 21:36.10 | epileg | configure was changed from shared to build-in libraries |
| 21:37.20 | epileg | changed --prefix from /usr/ to /usr/brlcad/ |
| 21:39.06 | CIA-53 | BRL-CAD: 03starseeker * r42733 10/brlcad/branches/cmake/TODO.cmake: Update TODO.cmake |
| 21:39.06 | brlcad | anything else? |
| 21:43.56 | CIA-53 | BRL-CAD: 03brlcad * r42734 10/brlcad/trunk/AUTHORS: add epileg's aliases |
| 21:47.28 | *** join/#brlcad R0b0t1 (~Enigma@64-136-219-55.dyn.everestkc.net) |
| 21:47.28 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) |
| 21:50.24 | CIA-53 | BRL-CAD: 03brlcad * r42735 10/brlcad/trunk/NEWS: |
| 21:50.24 | CIA-53 | BRL-CAD: extensive work for the debian platform by jordi sayol (epileg). there are new |
| 21:50.24 | CIA-53 | BRL-CAD: application menus, icons, mime-type associations, and other usability essentials |
| 21:50.24 | CIA-53 | BRL-CAD: for a self-contained .deb installer. this should make it easier for a platform |
| 21:50.24 | CIA-53 | BRL-CAD: maintainer to publish .deb files on a more regular basis. |
| 21:58.18 | CIA-53 | BRL-CAD: 03brlcad * r42736 10/brlcad/trunk/NEWS: |
| 21:58.18 | CIA-53 | BRL-CAD: you can now run 'ls -la' on a v4 database and have it properly report primitive |
| 21:58.18 | CIA-53 | BRL-CAD: types. it was reporting NULL and even potential crash-inducing garbage due to |
| 21:58.18 | CIA-53 | BRL-CAD: directly indexing into the rt_functab based on an unset d_minor_type. |
| 21:59.09 | CIA-53 | BRL-CAD: 03erikgreenwald * r42737 10/brlcad/branches/bottie/src/librt/primitives/bot/ (bot.c btg.c btg.h btgf.c tie.c tie_kdtree.c): warning quellage |
| 22:20.08 | *** join/#brlcad epileg (~epileg@unaffiliated/epileg) |
| 22:22.16 | CIA-53 | BRL-CAD: 03jordisayol * r42738 10/brlcad/trunk/misc/Makefile.am: Updated debian/ files list |
| 22:22.53 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) |
| 22:25.29 | CIA-53 | BRL-CAD: 03brlcad * r42739 10/brlcad/trunk/NEWS: |
| 22:25.29 | CIA-53 | BRL-CAD: mged's opendb command now has a -f flip option for reading in a |
| 22:25.29 | CIA-53 | BRL-CAD: binary-incompatible v4 geometry database with a flipped endianness translation. |
| 22:25.29 | CIA-53 | BRL-CAD: the files are made read-only but may then be run through keep or dbupgrade to |
| 22:25.29 | CIA-53 | BRL-CAD: clean them up. this closes out a long-standing request from users and a class |
| 22:25.30 | CIA-53 | BRL-CAD: of modeling/rendering failures due to the platform-specific nature of v4 files, |
| 22:25.31 | CIA-53 | BRL-CAD: albeit only minimally tested with a few sample files. |
| 22:27.45 | *** join/#brlcad ibot (~ibot@rikers.org) |
| 22:27.45 | *** topic/#brlcad is BRL-CAD Open Source Solid Modeling || http://brlcad.org || http://sf.net/projects/brlcad || #brlcad logs: http://ibot.rikers.org/%23brlcad/ || BRL-CAD release 7.18.0 is posted (20101209) || Happy Open Source Anniversary 2010-12-21 !!! Six years... |
| 22:29.07 | CIA-53 | BRL-CAD: 03starseeker * r42740 10/brlcad/branches/cmake/src/other/tk/CMakeLists.txt: Doesn't look like winMain.c belongs in libtk, but that's not the only issue. |
| 22:29.23 | CIA-53 | BRL-CAD: 03brlcad * r42741 10/brlcad/trunk/NEWS: (log message trimmed) |
| 22:29.23 | CIA-53 | BRL-CAD: dbupgrade (along with ALL other BRL-CAD commands) now supports reading in |
| 22:29.23 | CIA-53 | BRL-CAD: binary-incompatible v4 files. the low-level librt db_open() inteface for |
| 22:29.23 | CIA-53 | BRL-CAD: reading in files will now auto-convert a v4 file to an alternate endian format |
| 22:29.24 | CIA-53 | BRL-CAD: if it finds that flipping all combination member matrices makes them valid. for |
| 22:29.24 | CIA-53 | BRL-CAD: safety, that means if an old v4 has even one actual bad matrix (so that it's |
| 22:29.25 | CIA-53 | BRL-CAD: invalid regardless of being flipped), then the dbupgrade cannot be performed |
| 22:46.23 | CIA-53 | BRL-CAD: 03brlcad * r42742 10/brlcad/trunk/misc/Makefile.am: M-x sort-lines |
| 22:46.53 | epileg | ops, sorry |
| 22:47.01 | brlcad | no problem |
| 22:51.04 | CIA-53 | BRL-CAD: 03brlcad * r42743 10/brlcad/trunk/TODO: dbupgrade support is done |
| 22:53.39 | CIA-53 | BRL-CAD: 03brlcad * r42744 10/brlcad/trunk/src/tclscripts/mged/help.tcl: document the new -f option to opendb |
| 23:02.47 | CIA-53 | BRL-CAD: 03brlcad * r42745 10/brlcad/trunk/doc/docbook/system/mann/en/opendb.xml: document the opendb -f option |
| 23:11.05 | CIA-53 | BRL-CAD: 03r_weiss * r42746 10/brlcad/trunk/src/libbn/bntester.c: Added to 'bntester' support for testing function 'bn_isect_line3_line3'. |
| 23:15.18 | CIA-53 | BRL-CAD: 03brlcad * r42747 10/brlcad/trunk/include/ged.h: |
| 23:15.18 | CIA-53 | BRL-CAD: refactor principle of DRY -- Don't Repeat Yourself. user usage statements don't |
| 23:15.18 | CIA-53 | BRL-CAD: belong in the public header regardless, but they're also already listed in the |
| 23:15.18 | CIA-53 | BRL-CAD: manual pages and src/tclscripts/mged/help.tcl and the bastard replications for |
| 23:15.18 | CIA-53 | BRL-CAD: archer and rtwizard (Db.tcl and Ged.tcl). approx 700 line reduction. |
| 23:15.18 | CIA-53 | BRL-CAD: ultimately, usage should be defined in the source C file directly in src/libged |
| 23:17.33 | brlcad | hm, distcheck does not like the misc/debian/brlcad-db.png symbolic link |
| 23:18.48 | epileg | is this a problem? |
| 23:19.05 | epileg | well, I can make a copy of the png |
| 23:22.40 | ``Erik | some os's don't support symlinks, can it cp when running the makedeb script shtuff? |
| 23:26.07 | brlcad | epileg: yeah, a copy would be better |
| 23:26.35 | CIA-53 | BRL-CAD: 03jordisayol * r42748 10/brlcad/trunk/misc/debian/ (brlcad-db.png brlcad-db.png): |
| 23:27.02 | brlcad | should always have *some* comment message, at least say why :) |
| 23:27.30 | epileg | ok. sorry |
| 23:28.33 | brlcad | could have been "copy instead of symlink" or "distcheck hates symlinks or even "sean complained so make a copy" .. all fine :) |
| 23:29.13 | brlcad | i.e., what would be useful to someone looking at that change years from now when the reason is no longer obvious |
| 23:30.06 | *** topic/#brlcad by brlcad -> BRL-CAD Open Source Solid Modeling || http://brlcad.org || http://sf.net/projects/brlcad || #brlcad logs: http://ibot.rikers.org/%23brlcad/ || BRL-CAD release 7.18.0 is posted (20101209) || http://itmanagement.earthweb.com/osrc/article.php/3922041/50-Open-Source-Applications-for-Sci-Tech-Education.htm |
| 23:54.54 | Ralith | I'll be needing to do inward polygon+arc offsetting in the gcode generator; a bit of research has pointed me towards M. Held. "On the Computational Geometry of Pocket Machining", LNCS500, Springer-Verlag, 1991, which "uses a Voronoi Diagram (VD) to produce rounded or constant-radius offset curves." I can get a copy of this at my univ library. Sound like I'm on the right track? |
| 23:55.13 | Ralith | ... which describes how to use "..., rather |