| 01:56.02 | *** join/#brlcad Nohla (~jesica@201.255.231.124) |
| 02:02.59 | CIA-73 | BRL-CAD: 03brlcad * r38259 10/brlcad/trunk/src/libfb/ (asize.c fb_generic.c fb_obj.c fb_util.c fbserv_obj.c if_tk.c): take care of a handful of strictness warnings. test params, mark unused params, fix sign conversions, remove unnecessary funcs (tk_close_existing), and lil more. |
| 02:45.32 | CIA-73 | BRL-CAD: 03brlcad * r38260 10/brlcad/trunk/include/fbio.h: looks like FBIO union 'l' param is a length parameter so make it a size_t instead of long. probably a good candidate for being a stdint fixed-size type. |
| 02:58.37 | CIA-73 | BRL-CAD: 03brlcad * r38261 10/brlcad/trunk/src/libfb/ (9 files): quellage of all strict verbose compilation warnings consisting of mostly parameter checking. additional static initializers on the FBIOs. lot of sanity tests and unused parameter decls. should be no logic changes. |
| 03:11.02 | brlcad | looking like the benchmark failure is possibly an inadvertent commit of code not intended to go into 7.16 |
| 03:11.48 | brlcad | not confirmed yet, though |
| 03:42.27 | CIA-73 | BRL-CAD: 03brlcad * r38262 10/brlcad/trunk/src/libfb/if_tk.c: include the netinet/in.h header for htonl and ntohl declarations (even though modern posix provides them via arpa/inet.h, legacy still works) |
| 03:43.15 | CIA-73 | BRL-CAD: 03brlcad * r38263 10/brlcad/trunk/src/libfb/Makefile.am: enable strict compilation on libfb now too too now that all warnings are squashed (tested on 32-bit macosx and 64-bit linux) |
| 04:12.55 | CIA-73 | BRL-CAD: 03brlcad * r38264 10/brlcad/trunk/src/libged/adc.c: quell warning about the really long string literal (exceeding the 509 lines that the standard minimally guarantees). make it a usage func. |
| 05:00.05 | CIA-73 | BRL-CAD: 03brlcad * r38265 10/brlcad/trunk/include/vmath.h: even relatively new code presenting problems with exact comparison assumptions. test AZEL_FROM_V3DIR() with floating fuzz. |
| 05:02.20 | CIA-73 | BRL-CAD: 03brlcad * r38266 10/brlcad/trunk/src/libged/ (bigE.c bot_dump.c clone.c color.c): |
| 05:02.20 | CIA-73 | BRL-CAD: slew of warning eliminations including conversion of exact floating point |
| 05:02.20 | CIA-73 | BRL-CAD: comparisons to nearness tests. this will require some testing (affects bigE, |
| 05:02.20 | CIA-73 | BRL-CAD: bot_dump, and clone in particular), but rudimentary testing looks good thus far. |
| 05:34.48 | CIA-73 | BRL-CAD: 03brlcad * r38267 10/brlcad/trunk/src/libged/ (draw.c dup.c): quell minor warnings, add extra param checking |
| 05:34.51 | brlcad | bahhhh |
| 05:35.07 | brlcad | false positive .. benchmarks are fine |
| 05:36.33 | CIA-73 | BRL-CAD: 03brlcad * r38268 10/brlcad/trunk/TODO: false positive. benchmark seems to be working just fine. stale build setup (edited rootsolver files in the way). two bugs to go. |
| 05:36.42 | brlcad | and on that fail! .. harumph |
| 05:36.45 | brlcad | onward |
| 05:38.35 | *** join/#brlcad cosurgi (~cosurgi@atak.bl.pg.gda.pl) |
| 05:40.44 | CIA-73 | BRL-CAD: 03brlcad * r38269 10/brlcad/trunk/src/ (libged/editit.c mged/tedit.c mged/utility1.c): make ged_editit() consistent with libged API. first arg is always the command name the user issued. pass it up through from the calling application. |
| 05:42.10 | CIA-73 | BRL-CAD: 03brlcad * r38270 10/brlcad/trunk/src/libged/edmater.c: heh, missing common.h |
| 05:54.02 | CIA-73 | BRL-CAD: 03brlcad * r38271 10/brlcad/trunk/src/libged/ (erase.c ged_private.h): remove _ged_eraseAllSubpathsFromSolidList() from the private API namespace as it is unused. make it static instead. |
| 06:00.34 | CIA-73 | BRL-CAD: 03brlcad * r38272 10/brlcad/trunk/TODO: the developer's pairing to the quick intro |
| 06:11.37 | CIA-73 | BRL-CAD: 03brlcad * r38273 10/brlcad/trunk/src/libged/ged.c: at least these are only data, but still an indicator of design failure that needs to be corrected. leave notes. |
| 06:11.49 | CIA-73 | BRL-CAD: 03brlcad * r38274 10/brlcad/trunk/src/libged/facetize.c: minor quelling via param checking |
| 06:13.24 | CIA-73 | BRL-CAD: 03brlcad * r38275 10/brlcad/trunk/src/libged/find.c: restructure to avoid forward decl, quiet warnings by checking params. cleanup. |
| 06:16.59 | CIA-73 | BRL-CAD: 03brlcad * r38276 10/brlcad/trunk/src/libged/get_comb.c: minor exact floating point comparison converted to fuzzy |
| 06:17.16 | CIA-73 | BRL-CAD: 03brlcad * r38277 10/brlcad/trunk/src/libged/get_obj_bounds.c: minor quieting. |
| 06:17.30 | CIA-73 | BRL-CAD: 03brlcad * r38278 10/brlcad/trunk/src/libged/gqa.c: cleanup, ws, indent. |
| 06:18.31 | brlcad | libged 30% complete |
| 09:25.15 | *** join/#brlcad mafm (~mafm@83.50.132.58) |
| 09:26.20 | *** join/#brlcad mafm (~mafm@83.50.132.58) |
| 10:19.15 | d-lo | heh, brlcad never ceases to amaze me with the 24/7 commits :) |
| 10:24.05 | roberthl | 9/n |
| 12:24.06 | *** join/#brlcad parigaudi (~quassel@pd95b7f5e.dip0.t-ipconnect.de) |
| 13:35.35 | *** join/#brlcad akafubu (~akafubu@unaffiliated/akafubu) |
| 14:30.32 | CIA-73 | BRL-CAD: 03erikgreenwald * r38279 10/brlcad/trunk/src/librt/primitives/nmg/nmg_tri_mc.c: Clear edge vectors before firing. Use midpoint if cross-trace fails to intersect in the specified range. Begin primary rays 1 step size further down. |
| 15:36.11 | *** join/#brlcad roberthl (~robert@2001:ba8:1f1:f03d::2) |
| 15:36.11 | *** join/#brlcad roberthl (~robert@silentflame/member/roberthl) |
| 17:29.26 | *** join/#brlcad roberthl (~robert@2001:ba8:1f1:f03d::2) |
| 17:29.26 | *** join/#brlcad roberthl (~robert@silentflame/member/roberthl) |
| 18:02.35 | CIA-73 | BRL-CAD: 03brlcad * r38280 10/brlcad/trunk/HACKING: mention where the tab settings are |
| 18:26.37 | CIA-73 | BRL-CAD: 03brlcad * r38281 10/brlcad/trunk/src/conv/bot_shell-vtk.c: quell a few msvc warnings about signatures not matching. interestingly sees k&r sigs as empty. mark statics as HIDDEN too. |
| 18:27.33 | CIA-73 | BRL-CAD: 03brlcad * r38282 10/brlcad/trunk/src/rt/ (main.c view_bot_faces.c): quell msvc warnings. expand function cast so it matches. reorder funcs to avoid forward decls, add missing params but marked as unused. |
| 18:28.22 | CIA-73 | BRL-CAD: 03brlcad * r38283 10/brlcad/trunk/src/libbu/ (bomb.c temp.c): {msvc is a bit picky about () != (void) with regards to making function sigs match, so expand them out. |
| 19:58.54 | starseeker | huh. Can anyone else do a make ehy.s ehy and raytrace the result? I'm seeing something funny locally |
| 19:59.45 | starseeker | or an epa |
| 20:01.03 | starseeker | only seeing it on the Mac |
| 20:13.22 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) |
| 20:20.25 | ``Erik | seems to trace ok on my mac? |
| 20:20.31 | ``Erik | been a few hours since I've updated, though |
| 20:29.39 | starseeker | probably some local oddity on my machine - 7.16.2 does it too |
| 20:31.59 | starseeker | er... nevermind |
| 20:46.00 | ``Erik | just did 100 rt's in default a/e, then 180 rotating -a, didn't see the issue O.o :) |
| 20:52.26 | CIA-73 | BRL-CAD: 03brlcad * r38284 10/brlcad/trunk/misc/win32-msvc8/brlcad/brlcad.sln: create project groupings for 'Other' and 'Libraries' to group projects/products from src/other and src/lib* respectively |
| 21:01.10 | starseeker | ``Erik: O.o |
| 21:19.59 | *** join/#brlcad Nohla (~jesica@201.255.231.124) |
| 21:37.00 | CIA-73 | BRL-CAD: 03r_weiss * r38285 10/brlcad/trunk/src/conv/obj-g_new.c: more rework for nmg |
| 21:49.11 | CIA-73 | BRL-CAD: 03starseeker * r38286 10/brlcad/trunk/src/fb/ (Makefile.am fbthreadtest.c): |
| 21:49.11 | CIA-73 | BRL-CAD: Toss in a temporary, minimalist 'display a pix image using a tk photo' program |
| 21:49.11 | CIA-73 | BRL-CAD: that will be used for isolated testing of Tcl threads and how they interact with |
| 21:49.11 | CIA-73 | BRL-CAD: Tk photo drawing - not (yet) simulating raytracing, just preparing to work on |
| 21:49.11 | CIA-73 | BRL-CAD: basics. This program will be removed once threads are integrated into |
| 21:49.11 | CIA-73 | BRL-CAD: libdm/libfb proper. |
| 23:57.56 | ``Erik | "try playing quake 4 on an intel video card and let us know how you get on" "looks as good as on an ATI/nvidia card so far... hold up, frame 2 is coming up... yup, still looks good" |