| 00:09.16 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) | |
| 01:09.25 | starseeker | ``Erik: heh - http://xkcd.com/729/ |
| 02:01.05 | starseeker | reads over the CGM technical report... hmm... |
| 02:30.50 | starseeker | humph - looks like the meshing algorithms live in CAMAL, which is not LGPL |
| 02:32.35 | starseeker | wonders if they could be persuaded |
| 02:52.16 | brlcad | ``Erik: care to run one more, replacing the BU_BITV_SHIFT macro with a) the one-liner constant and b) a numeric constant (like 6) instead of the func ... that should indicate whether it's just no longer inlined (maybe due to the extern) |
| 02:53.19 | brlcad | also matters if CHAR_BIT is set .. that makes it a constant and avoids the function |
| 03:44.17 | *** join/#brlcad dtidrow (~dtidrow@c-71-238-51-148.hsd1.mi.comcast.net) | |
| 03:55.32 | CIA-73 | BRL-CAD: 03brlcad * r39349 10/brlcad/trunk/src/libged/ (search.c typein.c wdb_obj.c wdb_qray.c): quellage and cleanup. avoid setting vars within expressions. init vars. |
| 03:56.08 | CIA-73 | BRL-CAD: 03brlcad * r39350 10/brlcad/trunk/src/libtclcad/ged_obj.c: cleanup, init variables, avoid sets within exprs. |
| 03:57.08 | CIA-73 | BRL-CAD: 03brlcad * r39351 10/brlcad/trunk/src/libdm/dm_obj.c: expand forward decl function prototypes, de-k&r dmo_fbs_callback() |
| 03:57.34 | CIA-73 | BRL-CAD: 03brlcad * r39352 10/brlcad/trunk/src/libdm/query.c: avoid unreachability |
| 03:59.04 | CIA-73 | BRL-CAD: 03brlcad * r39353 10/brlcad/trunk/src/libdm/dm-wgl.c: |
| 03:59.04 | CIA-73 | BRL-CAD: move struct dm dm_wgl down to the bottom in order to avoid forward decls but |
| 03:59.04 | CIA-73 | BRL-CAD: don't restructure functions to remove them all just yet. rmeove the few that |
| 03:59.04 | CIA-73 | BRL-CAD: seem to be kosher for removal. update slew of k&r signature to ansi-style. |
| 03:59.04 | CIA-73 | BRL-CAD: this marks the end of more than 3000 win32/win64 verbose warnings that have been |
| 03:59.05 | CIA-73 | BRL-CAD: reviewed and/or addressed. |
| 04:03.21 | CIA-73 | BRL-CAD: 03brlcad * r39354 10/brlcad/trunk/src/other/tk/generic/tk.h: also file-scope protect y1 to avoid shadow compilation warnings on mac with opt enabled. |
| 04:03.53 | CIA-73 | BRL-CAD: 03brlcad * r39355 10/brlcad/trunk/include/opennurbs_ext.h: initialize variables 'just in case' |
| 04:08.25 | brlcad | and with that, they're done done |
| 04:08.36 | brlcad | now to recompile and see what broke :) |
| 04:37.19 | CIA-73 | BRL-CAD: 03brlcad * r39356 10/brlcad/trunk/src/libpkg/pkg.c: win32 quellage. funky ssize_t fun. |
| 06:10.11 | *** join/#brlcad stevegt_ (~stevegt@c-69-181-134-76.hsd1.ca.comcast.net) | |
| 10:13.35 | *** join/#brlcad dtidrow (~dtidrow@c-71-238-51-148.hsd1.mi.comcast.net) | |
| 10:14.40 | *** join/#brlcad stevegt_ (~stevegt@c-69-181-134-76.hsd1.ca.comcast.net) | |
| 11:30.13 | ``Erik | BU_BITV_SHIFT and bu_bitv_shift() defined as 3 (the result from the func), commented out the bu_bitv_shift definition in libbu/bitv.c, 20 samples, avg 27128, stdev 422.8 |
| 11:47.11 | ``Erik | http://paste.lisp.org/display/99453 |
| 12:10.13 | brlcad | hm, that makes me think CHAR_BIT is set and you were getting a constant |
| 12:10.30 | brlcad | making the inline/no-inline irrelevant unused |
| 12:11.03 | d-lo | kind of an interesting read: http://www.gamesradar.com/f/what-all-developers-should-learn-from-eve-online/a-20100517113116512049 |
| 12:17.10 | ``Erik | removes the CHAR_BIT test stuff in bu.h and recompiles without the inline O.o |
| 12:31.04 | ``Erik | this is looking a wee bit slower |
| 12:32.32 | ``Erik | around 5% slower |
| 12:39.58 | brlcad | okay, that's sounding better |
| 12:40.35 | brlcad | it should be about 5-10% iirc, but consistently slower |
| 12:41.03 | ``Erik | still thinks it simply makes no sense to try inlining something from a library, one of the fundamentals of a library is being able to fix stuff in the lib without recompiling dependant executables O.o |
| 12:41.26 | ``Erik | and the inline breaks windows projects trying to link against bu *shrug* |
| 12:42.35 | brlcad | don't disagree |
| 12:45.40 | brlcad | if windows has CHAR_BIT, it might be appropriate to resimplify the whole mess back to compile-time constants |
| 12:46.45 | brlcad | the intent was to replace the determination of a shift size from compile-time to run-time, as the type can be changed on the fly during compilation by changing the bitv_t type |
| 12:47.15 | brlcad | it has to match bitv_t, so the function was written (with the intent of fully replacing the macro) |
| 12:47.33 | brlcad | since replacing the macro would require deprecation, it was instead just defined to that function |
| 13:11.57 | CIA-73 | BRL-CAD: 03bob1961 * r39357 10/brlcad/trunk/misc/win32-msvc8/tclsh/library/installTree.tcl: Added code to clean up unwanted files in the install dir. |
| 13:12.03 | CIA-73 | BRL-CAD: 03brlcad * r39358 10/brlcad/trunk/src/other/tk/generic/ (tk.h tkDecls.h): oops, y1 is used within structures so we can't rename it. just change the two decls that were causing shadow warnings in the first place. |
| 13:13.22 | CIA-73 | BRL-CAD: 03bob1961 * r39359 10/brlcad/trunk/misc/win32-msvc8/tkhtml/tkhtml.vcproj: Copy the tkhtml.dll to bin/Tkhtml3.0 in the install dir. |
| 13:15.37 | CIA-73 | BRL-CAD: 03bob1961 * r39360 10/brlcad/trunk/misc/win32-msvc8/ (3dm2g/ 3dm2g/3dm2g.vcproj brlcad/brlcad.sln): Added a project for 3dm-g. |
| 13:23.37 | ``Erik | msvc8 has "#define CHAR_BIT 8" in limis.h |
| 13:24.54 | CIA-73 | BRL-CAD: 03brlcad * r39361 10/brlcad/trunk/src/libbn/tplot.c: compare floats against float literals |
| 13:25.40 | ``Erik | might be what flipped it out, "extern inline unsigned int 8" O.o |
| 13:29.35 | brlcad | er, it doesn't define anything to CHAR_BIT |
| 13:29.43 | brlcad | it just uses that to pick a constant size |
| 13:30.10 | brlcad | so it was just bitching on the declaration if anything, as it wouldn't even get used if CHAR_BIT is defined |
| 13:43.19 | CIA-73 | BRL-CAD: 03bob1961 * r39362 10/brlcad/trunk/src/tclscripts/archer/ (Archer.tcl ArcherCore.tcl): Tie the display of the "Center Dot" with the "Viewing Parameters". |
| 13:44.37 | brlcad | z:\brlcad\src\other\tkhtml3\src\html.h(68) : fatal error C1083: Cannot open include file: 'htmltokens.h': No such file or directory |
| 13:44.45 | brlcad | (win32 build) |
| 13:46.06 | ``Erik | ah, yeh, gotcha |
| 13:46.23 | ``Erik | should get sleep before trying to wind through preprocessor code blind O.o heh |
| 13:47.36 | ``Erik | ok, with the CHAR_BIT logic gutted, no inline, avg 25281.3985 stdev 352.7. with the inline avg 25282.4435, stdev 425.59 |
| 13:51.41 | CIA-73 | BRL-CAD: 03brlcad * r39363 10/brlcad/trunk/src/libdm/dm-wgl.c: have to declare dm_wgl if we're going to use it. restructure so wgl_setBGColor is defined before use. quell warning about setting var inside conditional. |
| 13:56.01 | CIA-73 | BRL-CAD: 03brlcad * r39364 10/brlcad/trunk/NEWS: erik fixed a memory leak in isst caused by a re-malloc instead of a realloc during window resizing. |
| 13:58.40 | CIA-73 | BRL-CAD: 03brlcad * r39365 10/brlcad/trunk/NEWS: bob made the various archer view commands work as if a database were open, allowing the view to be manipulated before opening a database. (this probably needs testing) |
| 14:00.12 | CIA-73 | BRL-CAD: 03erikgreenwald * r39366 10/brlcad/trunk/src/librt/primitives/metaball/metaball.c: cope with the ray originating inside of a solid in the disabled approach |
| 14:05.03 | CIA-73 | BRL-CAD: 03brlcad * r39367 10/brlcad/trunk/NEWS: keith improved the nurbs curve evaluation (r38915), exposed via the brep command, which should help generate better iso curves for a wireframe approximation. |
| 14:06.35 | CIA-73 | BRL-CAD: 03brlcad * r39368 10/brlcad/trunk/NEWS: bob added a list view to archer, which wasn't line-item documented in the last release, so at least not the various minor behavior enhancements that have occurred since then. keys, icons, etc. |
| 14:10.06 | CIA-73 | BRL-CAD: 03brlcad * r39369 10/brlcad/trunk/src/librt/cut.c: it's not dead code, it's just only used in the NEW_WAY sections which mike never got to finish |
| 14:14.45 | CIA-73 | BRL-CAD: 03brlcad * r39370 10/brlcad/trunk/src/librt/ (Makefile.am primitives/nmg/nmg_junk.c): nmg_junk should be enabled for compilation as the routines it provides are part of a work-in-progress that should be kept working. compilation is enabled to make sure it doesn't get out of sync. |
| 14:17.43 | brlcad | notes that we're actually probably ready to release again RSN |
| 14:33.57 | CIA-73 | BRL-CAD: 03starseeker * r39371 10/brlcad/trunk/src/isst/isst_tcltk.c: (slightly) better rotation behavior |
| 14:48.19 | CIA-73 | BRL-CAD: 03erikgreenwald * r39372 10/isst/trunk/sdl/ (event.c main.c): move setting the dirty flag into the action functions |
| 14:53.42 | CIA-73 | BRL-CAD: 03erikgreenwald * r39373 10/isst/trunk/sdl/event.c: make changing the demo plugin value a repeating key |
| 14:56.34 | CIA-73 | BRL-CAD: 03starseeker * r39374 10/brlcad/trunk/src/isst/ (isst.tcl isst_tcltk.c): More rotation behavior improvements |
| 14:57.46 | CIA-73 | BRL-CAD: 03brlcad * r39375 10/brlcad/trunk/ (17 files in 2 dirs): make the if_read/if_write callbacks take a size_t parameter instead of an int in order to propagate better to lower-level read/write functions that expect a size_t. |
| 15:01.10 | CIA-73 | BRL-CAD: 03brlcad * r39376 10/brlcad/trunk/src/librt/tcl.c: var init quellage |
| 15:01.34 | CIA-73 | BRL-CAD: 03brlcad * r39377 10/brlcad/trunk/src/librt/primitives/submodel/submodel.c: set off_t to off_t |
| 15:01.53 | CIA-73 | BRL-CAD: 03starseeker * r39378 10/brlcad/trunk/src/isst/isst_tcltk.c: Clear some printf debugging lines. |
| 15:04.07 | CIA-73 | BRL-CAD: 03brlcad * r39379 10/brlcad/trunk/src/librt/primitives/part/part.c: use the corresponding vmath constant instead of a truncated version here. |
| 15:04.37 | ``Erik | O.o |
| 15:07.14 | CIA-73 | BRL-CAD: 03starseeker * r39380 10/brlcad/trunk/src/isst/isst_tcltk.c: Couple more printf removals. |
| 15:08.03 | CIA-73 | BRL-CAD: 03brlcad * r39381 10/brlcad/trunk/src/librt/primitives/nmg/nmg_inter.c: clear up logic, remove set in conditional, remove debug statement (we already blathered) |
| 15:15.34 | CIA-73 | BRL-CAD: 03brlcad * r39382 10/brlcad/trunk/src/librt/primitives/nmg/nmg_plot.c: initialize the edge_g_cnurb to zero |
| 15:17.47 | CIA-73 | BRL-CAD: 03brlcad * r39383 10/brlcad/trunk/src/librt/memalloc.c: off_t vs size_t quellage |
| 15:19.03 | CIA-73 | BRL-CAD: 03brlcad * r39384 10/brlcad/trunk/src/librt/primitives/generic.c: initialize avs to zero |
| 15:25.13 | CIA-73 | BRL-CAD: 03brlcad * r39385 10/brlcad/trunk/src/librt/primitives/hyp/hyp.c: protect from division by zero (looks like there are several potentials in here) |
| 15:28.12 | CIA-73 | BRL-CAD: 03bob1961 * r39386 10/brlcad/trunk/src/libtclcad/tclcadAutoPath.c: Added path for Tkhtml3.0 so a "package require Tkhtml 3.0" will work on Windows. |
| 15:28.56 | CIA-73 | BRL-CAD: 03brlcad * r39387 10/brlcad/trunk/src/librt/comb/db_comb.c: more size_t quellage |
| 15:33.46 | CIA-73 | BRL-CAD: 03brlcad * r39388 10/brlcad/trunk/src/librt/db5_io.c: you can't have our bomb |
| 15:34.16 | CIA-73 | BRL-CAD: 03brlcad * r39389 10/brlcad/trunk/src/librt/ (db5_alloc.c db5_scan.c db_alloc.c): size_t off_t mismatching matched |
| 15:36.16 | CIA-73 | BRL-CAD: 03brlcad * r39390 10/brlcad/trunk/src/librt/primitives/bspline/bspline.cpp: remove dead code. |
| 15:39.33 | CIA-73 | BRL-CAD: 03brlcad * r39391 10/brlcad/trunk/src/libged/wdb_obj.c: more size_t node counting |
| 15:57.14 | CIA-73 | BRL-CAD: 03brlcad * r39392 10/brlcad/trunk/src/libged/wdb_bigE.c: massive style, ws, indent update. added a slew of missing semicolons on macros to keep things looking like code. |
| 16:03.29 | CIA-73 | BRL-CAD: 03brlcad * r39393 10/brlcad/trunk/src/libged/ (search.c tables.c typein.c): never-ending size_t quellage |
| 16:08.25 | CIA-73 | BRL-CAD: 03brlcad * r39394 10/brlcad/trunk/src/libged/red.c: unused var, check_comb needs to return negative status values. |
| 16:24.38 | CIA-73 | BRL-CAD: 03brlcad * r39395 10/brlcad/trunk/src/libged/ (ged.c get_comb.c lt.c): assert and compare size_t |
| 16:25.22 | CIA-73 | BRL-CAD: 03brlcad * r39396 10/brlcad/trunk/src/libged/png.c: fix infinite loop bug, needs to be a signed type if we're going to iterate past zero. removed unused vars. |
| 16:25.39 | *** join/#brlcad Ralith (~ralith@216.162.199.202) | |
| 16:25.54 | d-lo | Looks like brlcad is on a roll. |
| 16:58.53 | *** join/#brlcad stevegt_ (~stevegt@c-69-181-134-76.hsd1.ca.comcast.net) | |
| 17:13.25 | CIA-73 | BRL-CAD: 03starseeker * r39397 10/brlcad/trunk/src/isst/isst_tcltk.c: Have the zero_view just make the model visible, like in the sdl version. Add in some more functions for other motions, will try hooking up to tk bindings. |
| 17:44.19 | CIA-73 | BRL-CAD: 03r_weiss * r39398 10/brlcad/trunk/src/conv/obj-g_new.c: fixed some test_closure bugs |
| 17:45.11 | CIA-73 | BRL-CAD: 03bob1961 * r39399 10/brlcad/trunk/ (include/bu.h src/libtclcad/tclcadAutoPath.c): Make the default BU_DIR_SEPARATOR be a '/' |
| 17:47.05 | CIA-73 | BRL-CAD: 03bob1961 * r39400 10/brlcad/trunk/src/tclscripts/mged/man.tcl: Mods to get man pages on Windows. |
| 18:35.00 | *** join/#brlcad stevegt_1 (~stevegt@cislunar.TerraLuna.Org) | |
| 19:50.47 | CIA-73 | BRL-CAD: 03brlcad * r39401 10/brlcad/trunk/BUGS: |
| 19:50.47 | CIA-73 | BRL-CAD: keith found an interesting background pixel difference comparing a BoT raytrace |
| 19:50.47 | CIA-73 | BRL-CAD: with a NURBS ray trace. The background pixels on the BoT image were off |
| 19:50.47 | CIA-73 | BRL-CAD: slightly by one for a large portion of scanlines (in the lower portion of the |
| 19:50.47 | CIA-73 | BRL-CAD: image). entire scanline was affected, but only the background pixels were off |
| 19:50.47 | CIA-73 | BRL-CAD: slightly, iirc. |
| 20:27.53 | CIA-73 | BRL-CAD: 03starseeker * r39402 10/brlcad/trunk/src/isst/ (isst.tcl isst_tcltk.c): Can now move forward, backward, left and right |
| 20:33.12 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) | |
| 20:51.03 | ``Erik | ahhh |
| 21:16.59 | CIA-73 | BRL-CAD: 03bob1961 * r39403 10/brlcad/trunk/include/bu.h: Undo the previous commit. |
| 21:18.36 | CIA-73 | BRL-CAD: 03starseeker * r39404 10/brlcad/trunk/src/isst/ (isst.tcl isst_tcltk.c): Support changing resolution |
| 22:00.59 | *** join/#brlcad stevegt_1 (~stevegt@cislunar.TerraLuna.Org) | |
| 22:21.40 | *** join/#brlcad stevegt_ (~stevegt@cislunar.TerraLuna.Org) | |
| 22:24.15 | CIA-73 | BRL-CAD: 03bob1961 * r39405 10/brlcad/trunk/src/libtclcad/tclcadAutoPath.c: Using braces around path arguments to prevent Tcl from evaluating backslaches. |
| 23:03.37 | *** join/#brlcad Owner (~chatzilla@va-67-233-102-229.sta.embarqhsd.net) | |
| 23:04.09 | Owner | i'm new with brlcad, and importing isn't working for me it's saying that stl-g (for importing stl files) isn't a valid command? |
| 23:04.26 | Owner | am typing it in the mged command line area |
| 23:05.00 | Owner | when i go to import in the file menu it only talks about .g databases |
| 23:05.14 | Owner | are there some plugins somewhere that i'm missing? thank you |
| 23:10.59 | Owner | i'll leave this channel open for a bit = please feel free to chime in whenever |
| 23:12.08 | ``Erik | stl-g is a program, not an mged command |
| 23:21.17 | *** join/#brlcad stevegt_1 (~stevegt@cislunar.TerraLuna.Org) | |
| 23:24.21 | Owner | thanks :) |
| 23:24.39 | Owner | i see it now, in the bin directory |
| 23:34.58 | CIA-73 | BRL-CAD: 03bob1961 * r39406 10/brlcad/trunk/src/mged/mged.c: If on Windows add bin/Tkhtml3.0 to the auto_path. |
| 23:39.09 | Owner | now it's giving me an "unrecognized line" error... any thoughts? it's the first line in the file, however i've tried removing it to no avail... something to do with how blender saves the files, although every other software i've tested these stl files with have opened them straight away.. |
| 23:51.56 | ``Erik | hrm, wonder if it's a unix vs dos newline issue? |