File Release Notes and Changelog
Release Name: 7.12.0
Notes: Coalescing after just over a year of concentrated development efforts, this release unveils a plethora of Windows-specific enhancements, bug fixes, and additional tools ported. Notably, the Windows improvements include fixing various issues with file management and porting many more ray tracers, geometry converters, framebuffer tools, and MGED classic mode support. This release also includes a new geometry converter for the United Kingdom's INTAVAL format from Her Majesty's Ministry of Defence. The new importer was jointly developed by TNO PML and IABG mbH and is included in the BRL-CAD sources as a contributed module. Of the numerous bug fixes and feature enhancements in this release, one of the biggest productivity enhancements is the addition of the mged "clone" and "tracker" commands. Clone is a very powerful yet simple command that provides an intuitive means for performing deep copies of objects. Tracker provides an interface for replicating objects along a spline path. These 'new' commands originate from modifications originally developed by the now defunct Geometric Solutions, Inc., and after receiving considerable updates and rewrites are now once again available to users. Finally, this release also marks a major change to the BRL-CAD source code repository with a move from CVS to Subversion recently completed. The conversion was successfully completed with the entire history of BRL-CAD development preserved, accounting for nearly 25 years of continuous active development. * automatic redraw updates after Geometry Browser edits - Bob Parker * improved Geometry Browser mouse behavior in mged - Bob Parker * make fast4-g use GIFT_material=1 and los=100 - Sean Morrison * fixed Mac OS X 'ogl' framebuffer crash - Sean Morrison, Bob Parker * removed ogl framebuffer multiple window 'm' option - Sean Morrison * tree -d option in MGED for limiting tree display depth - Cliff Yapp * added mouse color query support to OpenGL framebuffer - Bob Parker * fixed X11 framebuffer colormapping bug - Bob Parker * libpng updated to 1.2.25 - Erik Greenwald * fixed several mged I/O bugs for stdin/stdout/stderr - Bob Parker * nirt now uses bounding sphere for -b backout option - Erik Greenwald * document the signals rt responds to for progress - Sean Morrison * libpng updated to 1.2.24 - Erik Greenwald * TCL/TK updated to 8.5.1 - Erik Greenwald * MGED now runs in the foreground by default - Erik Greenwald * fixed g-dxf bug where incorrect object name is assigned - Bob Parker * fixed browser-based mged help on Mac OS X - Sean Morrison * added rest of BRL-CAD's geometry converters for Windows - Bob Parker * added more missing framebuffer tools to Windows port - Bob Parker * rtwizard enhancments for ghosting and display - Bob Parker * mged reports failure if ray-tracer returns non-zero - Sean Morrison * improved mged corrupt geometry file handling - Sean Morrison * improved dbconcat object suffix and prefix options - Sean Morrison * Tcl/Tk updated to 8.5 final from 8.5b1 - Erik Greenwald * fixed PIX image file creation bug on Windows - Bob Parker * ported rtwizard to function properly on Windows - Bob Parker * fixed mged text-editing commands on Windows platform - Bob Parker * removed duplicate points text-editing ARBs in mged - John Anderson * improved mged ted line-ending support using wordpad - Bob Parker * fixed g_diff bug where some differences were missed - John Anderson * fixed bug evaluating metaball normals - Erik Greenwald, Ed Davisson * improved metaball primitive rendering - Erik Greenwald * fixed bug in mged rcc-cap command when using 't' option - Bob Parker * mged classic mode now works on Windows platform - Bob Parker * fixed Windows-specific raytrace framebuffer delay - Bob Parker * new tracker tool for extracting NEWS tracker data - Sean Morrison * converted source repository from CVS to Subversion - Sean Morrison * new 'dir2ae' mged command for getting az/el from dir - Sean Morrison * new 'ae2dir' mged command for getting dir from az/el - Bob Parker * fixed empty name crash when using mged 'lt' command - Sean Morrison * Pro/E plugin allows spaces in part name mapping file - Sean Morrison * Pro/E plugin creates 'top.#' if 'top' already exists - Sean Morrison * fixed asc2g bug on Windows that caused a crash on exit - Bob Parker * fixed bugs in several Utah Raster Toolkit tools - Pedro Giffuni * fixed mged start-up initialization bug on Windows - Daniel Roßberg * graceful handling of mesh normal failures - Sean Morrison * fixed variety of g-iges and other exporter crashes - Sean Morrison * fixed crash-on-exit cleanup bug in g-off exporter - Sean Morrison * fixed crash when xpush was called on a primitive - Erik Greenwald * fixed Windows port 'opendb' command crash - Bob Parker * added configurable max_dist depth option to rtedge - Erik Greenwald * added rest of BRL-CAD's ray-tracers to the Windows port - Bob Parker * enabled embedded framebuffer support in Windows port - Bob Parker * fixed garbled output during parallel crash reports - Sean Morrison * fixed memory deallocation bug in the raytracers - Bob Parker * renamed 'hd' hexadecimal dump utility to 'hex' - Sean Morrison * renamed procedural 'wall' generator to 'masonry' - Sean Morrison * mged 'tol' command now accepts multiple tolerances - Sean Morrison * TCL and TK updated to 8.5b1. itcl/itk to dev 3.4. - Erik Greenwald * fixed non-normalized vector bugs with ARBN objects - John Anderson * pix-png warns if PNG is smaller than the PIX input - Sean Morrison * new tgf-g INTAVAL importer - TNO PML, IABG mbH * removed longjmp warning message from applications - Sean Morrison * upgraded bundled incrTcl to version 3.3 - Erik Greenwald * fixed bug with -g/-G cell size options to rt - John Anderson * 'tree' command without args now shows e'd objects - Sean Morrison * new pix-fb -p option to pause before exiting - Karel Kulhavy * fixed mged 'B' command behavior with -A and no args - John Anderson * updated bundled libpng to version 1.2.21 - Erik Greenwald * removed raytrace image continuation code - Erik Greenwald * new mged 'clone' command for performing deep copies of objects - Adam Ross, Sean Morrison, Erik Greenwald * new mged 'tracker' object routing command - Sean Morrison * fixed crash in mged ls -A attribute listing command - Sean Morrison * fixed dbconcat of empty combinations, crash in mged - John Anderson
Changes: 2007-10-24 20:30 erikgreenwald * src/other/incrTcl/CHANGES, src/other/incrTcl/ChangeLog, ... lots of other files ..., src/other/incrTcl/itk/win/rc/itk.rc: Upgrade to itcl/itk 3.3 2007-10-24 21:38 erikgreenwald * src/other/libpng/ANNOUNCE, src/other/libpng/CHANGES, ... lots of other files ..., src/other/libpng/scripts/pngw32.def: upgrade to 1.2.22 ("A new bug with decoding iCCP chunks was fixed.") 2007-10-26 15:42 brlcad * regress/Makefile.am: delete the solids.rt.log file 2007-10-26 19:41 brlcad * src/lgt/lgt.c: remove dead iris code 2007-10-26 21:51 brlcad * src/irprep/all_sf.c, src/irprep/firpass.c, src/irprep/ir-X.c, ... lots of other files ..., src/util/pl-X.c, src/util/rle-pix.c, src/vas4/vas4.c: yet another surge of level 4 flawfinder string length checks. somewhere around 30-40% done. 2007-10-26 22:25 brlcad * configure.ac: re-enable the checks for tcl8.4, see if we can get both 8.4 and 8.5 working so beta isn't required. 2007-10-26 22:28 brlcad * NEWS: merge in the release notes from STABLE 2007-10-26 22:29 brlcad * ChangeLog: merge in 7.10.4 ChangeLog from STABLE 2007-10-27 06:51 brlcad * src/liboptical/material.c: don't blather verbose if we can't find the dynamic shader 2007-10-27 21:51 brlcad * src/util/dbcp.c: quell compilation warnings, remove prs() 2007-10-27 21:58 brlcad * src/vdeck/vdeck.c: quell compilation warnings for constness changes 2007-10-28 04:21 brlcad * src/util/dbcp.c: fix fpritnf typo 2007-10-28 05:10 brlcad * src/burst/error.c: ws 2007-10-28 05:41 brlcad * src/conv/euclid-g.c, src/conv/g2asc.c, src/conv/iges/iges.c, src/conv/iges/revolve.c, src/conv/jack/g-jack.c, src/conv/nmg-sgp.c, src/conv/patch/rpatch.c, src/conv/poly-bot.c, src/mged/clone.c: more flawfinder and bu_exit conversions 2007-10-28 17:27 brlcad * src/fb/fbfade.c: remove the non-stdarg interfaces, we can assume c89 now. c-set-style stroustrup ws too. 2007-10-28 17:35 brlcad * src/fb/fbgamma.c: convert to libbu memory management 2007-10-29 03:08 brlcad * src/other/blt/Makefile.am: blt actually makes direct calls to X11, so need to include the goods here else face the wrath of unresolved symbols on some plats (encountered on RHEL) 2007-10-29 03:27 brlcad * configure.ac: blt *requires* X11 to function, turn it off if x11 isn't available 2007-10-29 03:27 erikgreenwald * src/mged/clone.c: I have no clue what a "BUFXIZE" is, but I know what BUFSIZ is... 2007-10-29 04:53 brlcad * src/libdm/Makefile.am: more undefined references on RHEL, need LIBGL if opengl is enabled 2007-10-29 04:55 brlcad * src/librtserver/Makefile.am: more undefined references on RHEL -- need the deps for rt, bn, and bu 2007-10-29 05:17 brlcad * configure.ac: more RHELness, multispectral needs symbols from optical 2007-10-29 05:21 brlcad * src/adrt/libtie/Makefile.am: uses libm 2007-10-29 05:25 brlcad * src/external/Cubit/g-sat.cxx, src/external/ProEngineer/proe-brl.c, src/external/Unigraphics/ug-g.c, src/fb/cmap-fb.c, src/fb/fb-cmap.c, src/fb/fbgammamod.c, src/fb/fbpoint.c, src/fb/fbscanplot.c: more flawfinding quellage and bu_exiting 2007-10-29 05:27 brlcad * src/mged/tedit.c: quell constness warnings, make the editor string const 2007-10-30 09:32 d_rossberg * configure.ac, src/other/Makefile.am, src/other/intaval, src/other/intaval/.cvsignore, src/other/intaval/Makefile.am, src/other/intaval/glob.h, src/other/intaval/read_dra.cpp, src/other/intaval/read_dra.h, src/other/intaval/readme.txt, src/other/intaval/regtab.cpp, src/other/intaval/regtab.h, src/other/intaval/tgf-g.cpp, src/other/intaval/tgf2g.dsp, src/other/intaval/write_brl.cpp, src/other/intaval/write_brl.h: initial checkin of TNO's/IABG's INTAVAL to BRL-CAD converter 2007-10-30 09:51 brlcad * configure.ac: intaval/Makefile 2007-10-30 10:05 brlcad * configure.ac: search the opennurbs dir for header files, it's required when compiling c++ sources that use rtgeom.h (which uses brep.h, which requires opennurbs.h) 2007-10-30 10:12 brlcad * src/other/intaval/Makefile.am: ah, blasted automake (or my assumptions of it). just include the opennurbs cppflags here since c++ files are still in the minority. 2007-10-30 10:16 brlcad * configure.ac: don't make up var names, just comment it out for now. as more C++ is integrated, it can be made default 2007-10-30 10:20 brlcad * NEWS: new tgf-g INTAVAL importer 2007-10-30 10:25 brlcad * NEWS: add PML to TNO's name. would preferably reference the developers directly but don't know yet if there were others coding besides Wim B. and Daniel R. working on it 2007-10-30 14:04 brlcad * src/other/intaval/readme.txt: annotate what DRA and TGF stand for 2007-10-30 15:27 d_rossberg * configure.ac: always set the path to the openNURBS headers (C++) 2007-10-30 15:36 brlcad * BUGS: apparently not yet annotated, so make a note of the stupid unable to find brl-cad message when rt encounters a shader it doesn't recognize. the dynamic shader loading code has some bad juju in it. 2007-10-31 04:48 brlcad * src/other/intaval/Makefile.am: ah crap. the converter uses libwdb which isn't/can't be compiled before src/other -- one of the few ways where recursive make actually bites us in the arse. need to move the converter to a directory that is traversed after src/other and after src/lib*, maybe misc/intaval or contrib/intaval or src/contrib/intaval etc. for now, just make the converter be compile-on-demand (i.e. cd src/other/intaval && make tgf-g) only. 2007-10-31 07:38 brlcad * src/librt/db_walk.c: remove old comment 2007-10-31 08:07 brlcad * misc/win32-msvc7/Makefile.am, misc/win32-msvc7/dxf2g, misc/win32-msvc7/dxf2g/.cvsignore, misc/win32-msvc7/dxf2g/dxf2g.vcproj, misc/win32-msvc7/g2dxf, misc/win32-msvc7/g2dxf/.cvsignore, misc/win32-msvc7/g2dxf/g2dxf.vcproj: by popular demand, add the g2dxf and dxf2g converters to the windows release build system 2007-10-31 08:34 brlcad * src/sig/Makefile.am, src/sig/coswin.c: quell malloc warning, use libbu allocator 2007-10-31 08:39 brlcad * src/mged/tedit.c: quell warnings 2007-11-01 15:01 brlcad * include/conf/Makefile.am: need all-am to be dependent upon built_sources else someone doing a make install without first running make will end up failing to generate the COUNT, DATE, HOST, USER, PATH files 2007-11-01 20:33 brlcad * src/util/pix-png.1: clarify that the user must specify the size if the image is not 512x512 2007-11-01 21:50 brlcad * src/util/pix-png.c: warn if the output PNG image dimensions are smaller than the input PIX image, related to support request from mallory 2007-11-01 21:56 brlcad * NEWS: pix-png now warns the user if the PNG being created has image dimenions smaller than the PIX file that was given as input. The tool compares the size of the input and output images, and reports the difference if there is one; it also suggest what flags the user may need to use if the input has recognizable image dimensions. This feature was implemented in response to a support request from Mallory where he was confused by the black PNG image produced (which happened to be the top of the input PIX image) since he had not specified the input image dimensions. 2007-11-01 22:00 brlcad * BUGS: mged isn't updating the graphics window with the new 8.5 tcl (sometimes) 2007-11-04 16:08 bob1961 * src/tclscripts/lib/Display.tcl, src/tclscripts/lib/Dm.tcl, src/tclscripts/lib/QuadDisplay.tcl, src/tclscripts/lib/RtControl.tcl: Activate the fbserv capability on Windows. 2007-11-04 16:15 bob1961 * src/tclscripts/archer/Archer.tcl, src/tclscripts/archer/LoadArcherLibs.tcl: Activate the fbserv capability on Windows. 2007-11-04 16:18 bob1961 * src/other/tk/win/tkWinMenu.c: Add option to have double bar tearoffs on windows. 2007-11-04 16:24 bob1961 * src/mged/mged_dm.h: Changed "#ifdef _WIN32" to "#if defined(_WIN32) && \!defined(__CYGWIN__)" 2007-11-04 16:26 bob1961 * src/mged/ged.c: Activate the fbserv capability on Windows. 2007-11-04 16:32 bob1961 * src/mged/fbserv.c: Mods to get the fbserv capability working on Windows. This required using Tcl_OpenTcpServer instead of pkg_permserver which beget other mods. 2007-11-04 16:39 bob1961 * src/mged/dm-wgl.c: Mods to accommodate the new hdc parameter to _wgl_open_existing(). 2007-11-04 16:51 bob1961 * src/librt/dg_obj.c, src/libtclcad/tclcadAutoPath.c: Mods to get things working on Windows. 2007-11-04 16:58 bob1961 * src/libfb/tcl.c: Mods to get the embedded fbserv working in the cadwidgets. 2007-11-04 17:02 bob1961 * src/libfb/fbserv_obj.c: Mods to get the fbserv capability working on Windows. This required using Tcl_OpenTcpServer instead of pkg_permserver which beget other mods. 2007-11-04 17:09 bob1961 * src/libdm/dm_obj.c: Mods to accommodate the signature change of fbs_open and fbs_close. 2007-11-04 17:12 bob1961 * src/fbserv/fbserv.c: Mods to get things working on Windows. 2007-11-04 17:25 bob1961 * include/fbserv_obj.h: Added fbsl_chan member to struct fbserv_listener. Added fbsc_chan and fbsc_handler members to struct fbserv_client. 2007-11-04 17:42 bob1961 * src/libfb/if_wgl.c: Mods to accommodate the new hdc parameter to _wgl_open_existing(). 2007-11-04 17:45 bob1961 * misc/win32-msvc7/libfb/libfb.vcproj: Using fbserv_obj.c instead of fbserb_obj_win32.c 2007-11-04 17:47 bob1961 * misc/win32-msvc7/fbclear/fbclear.vcproj, ... lots of other files ..., misc/win32-msvc7/stl2g/stl2g.vcproj: Minor updates. 2007-11-04 18:25 bob1961 * misc/win32-msvc7/itkstub/itkstub.vcproj: Update include dirs. 2007-11-05 04:38 brlcad * src/fb/fb-pix.c, src/fb/fb-png.c, src/fb/fbclear.c, src/fb/pix-fb.c, src/fb/png-fb.c, src/libfb/fbserv_obj.c, src/libpkg/pkg.c, src/mged/fbserv.c: use the feature-specific HAVE_WINSOCK_H header define instead of platform-specific defines 2007-11-05 14:28 brlcad * configure.ac: libpng needs zlib to link (on some platforms) 2007-11-05 17:56 brlcad * NEWS: erik upgraded the bundled incrtcl/tk from 3.3b1 to 3.3.0 2007-11-06 20:00 brlcad * NEWS: Mr. Anderson fixed various non-normalized vector bugs with the ARBN primitive that was causing all sorts of problems with rotations, shading, mirroring, and more. the primitive was assuming the vectors were normalized, thus causing the problems. Johns fixes take care of sf bugs 1800148 and 1800161 reported by clock3/karel that specifically reported the mirroring and shading problems. 2007-11-06 20:27 bob1961 * src/libfb/if_wgl.c: We don't need to check for Windows in this file ;-). 2007-11-06 20:37 bob1961 * src/libfb/if_remote.c: Fixed call to strncpy so as not to overwrite buf. 2007-11-06 20:41 bob1961 * src/libfb/fbserv_obj.c: Check fbsl_fd before closing. 2007-11-06 20:43 bob1961 * src/other/incrTcl/itcl/generic/itclInt.h: Looks like Sean's earlier fix got lost. Putting it back. 2007-11-06 21:35 erikgreenwald * src/other/tcl/ChangeLog, src/other/tcl/ChangeLog.1999, ... lots of other files ..., src/other/tcl/win/tclsh.rc: tcl 8.5a6->8.5b1 2007-11-06 21:38 erikgreenwald * src/other/tk/ChangeLog, src/other/tk/README, ... lots of other files ..., src/other/tk/xlib/xutil.c: tk 8.5a6->8.5b1 2007-11-06 21:41 erikgreenwald * src/other/incrTcl/itcl/generic/itcl.decls, src/other/incrTcl/itcl/generic/itcl.h, src/other/incrTcl/itcl/generic/itclDecls.h, src/other/incrTcl/itcl/generic/itclInt.h, src/other/incrTcl/itcl/generic/itclIntDecls.h, src/other/incrTcl/itcl/generic/itclStubInit.c, src/other/incrTcl/itcl/generic/itclStubLib.c, src/other/incrTcl/itcl/generic/itcl_bicmds.c, src/other/incrTcl/itcl/generic/itcl_class.c, src/other/incrTcl/itcl/generic/itcl_cmds.c, src/other/incrTcl/itcl/generic/itcl_ensemble.c, src/other/incrTcl/itcl/generic/itcl_linkage.c, src/other/incrTcl/itcl/generic/itcl_methods.c, src/other/incrTcl/itcl/generic/itcl_migrate.c, src/other/incrTcl/itcl/generic/itcl_objects.c, src/other/incrTcl/itcl/generic/itcl_parse.c, src/other/incrTcl/itcl/generic/itcl_util.c: update a good bit out of itcl development version, to cope with changes in tcl85b1 2007-11-06 21:42 erikgreenwald * src/other/incrTcl/itk/generic/itk.decls, src/other/incrTcl/itk/generic/itk.h, src/other/incrTcl/itk/generic/itkDecls.h, src/other/incrTcl/itk/generic/itkStubInit.c, src/other/incrTcl/itk/generic/itkStubLib.c, src/other/incrTcl/itk/generic/itk_archetype.c, src/other/incrTcl/itk/generic/itk_cmds.c, src/other/incrTcl/itk/generic/itk_option.c, src/other/incrTcl/itk/generic/itk_util.c, src/other/incrTcl/itk/library/Archetype.itk, src/other/incrTcl/itk/library/Toplevel.itk, src/other/incrTcl/itk/library/Widget.itk: update a good bit out of itk development version, to keep in sync with itcl 2007-11-06 21:43 erikgreenwald * configure.ac: note that we use (and search for) itcl/itk 3.4 2007-11-06 22:41 erikgreenwald * NEWS: mention version bumps on contrib stuff 2007-11-06 23:58 brlcad * NEWS, src/librt/wdb_obj.c, src/tclscripts/helplib.tcl: the mged 'tol' command now accepts multiple tolerance arguments so that users can set multiple tolerances with just one command invocation. this change makes the tol command interatively accept pairs of tolerance types with the corresponding value. 2007-11-07 15:43 brlcad * TODO: review the root solver 2007-11-07 19:31 bob1961 * src/librt/timer-nt.c: time() requires a time_t 2007-11-07 23:35 brlcad * configure.ac: update the adrt comment, it actually builds fairly cleanly not IF you have the dependencies installed, but still don't force-enable since they are unmanaged external 2007-11-08 00:12 brlcad * configure.ac: don't just check whether the option is set to the default -- check if it's been set at all so that you can use --enable-all with --disable options that may be default 'no' already. ran into this situation trying to disable opennurbs (which was already default off). 2007-11-08 03:40 brlcad * src/librt/timerunix.c: the unix timer has the same time_t problem that the nt timer had, being that it would be crash-worthy if time_t is actually a 64-bit type 2007-11-08 04:15 brlcad * src/conv/g-nmg.c, src/conv/viewpoint-g.c, src/libbu/whereis.c, src/libbu/which.c, src/librt/vdraw.c, src/mged/dir.c, src/mged/points/points_scan.l, src/nirt/command.c, src/nirt/if.c, src/nirt/nirt.c, src/proc-db/molecule.c: minor tweaks, preventive null terminating of strncmp'd strings 2007-11-08 04:55 brlcad * NEWS, src/proc-db, src/proc-db/.cvsignore, src/proc-db/Makefile.am, src/proc-db/masonry.c, src/proc-db/wall.c: the 'wall' procedural geometry generator tool was renamed to 'masonry' so as to not conflict with the commonly named 'wall' linux tool that writes a message to all users. source renamed from wall.c to masonry.c 2007-11-08 05:08 brlcad * NEWS, src/util, src/util/.cvsignore, src/util/Makefile.am, src/util/hd.1, src/util/hd.c, src/util/hex.1, src/util/hex.c: rename the 'hd' hexdump utility to 'hex' as well given the similar linux facilities and cause for conflict/confusion. source renamed from hd.c to hex.c (and manpage in kind) 2007-11-08 05:09 brlcad * TODO: renamed wall and hd commands, someone(tm) should review what else other than our libs that might be common conflict cases where we don't have a justifiable basis/need to keep the name 2007-11-08 05:35 brlcad * doc/html/manuals/Anim_Tutorial/chapt9.html, src/mged/vld_std.h: obliterate BRL.MIL references 2007-11-08 05:41 brlcad * sh/cray.sh, src/fb/cell-fb.1, src/lgt/lgt.1, src/libfb/libfb.3: more brl.mil references begone 2007-11-08 05:54 brlcad * doc/book/VolumeII.xml, doc/book/VolumeIV.xml, doc/html/manuals/Anim_Tutorial/index.html, doc/html/manuals/mged/animmate/anim_doc.html, doc/html/manuals/mged/mged_cmd_index.html: more address updating to avoid mil domain 2007-11-08 14:27 brlcad * src/proc-db/Makefile.am: don't choke here if openNURBS is disabled! 2007-11-08 14:32 erikgreenwald * src/mged/points/points_scan.l: Curly braces and brackets cannot be used interchangeably. 2007-11-08 15:32 erikgreenwald * src/mged/clone.c: add a break; to the -p option (been missing since the original import from GSI, wow) 2007-11-08 15:46 brlcad * sh/make_dmg.sh: dynamically calculate the size that the DMG needs to be (since the 7.10.4 release already surpassed the 250MB barrier) .. and give me at least a minute to find that background image :) 2007-11-08 18:04 bob1961 * src/libbu/fchmod.c: Initial check-in. 2007-11-08 18:53 bob1961 * src/librt/prep.c: rt_unprep() was using paths (i.e. a struct bu_ptbl) without being initialized. In fact, paths is never initialized. I'm guessing the intent was to use objs->paths. 2007-11-08 19:04 bob1961 * include/bu.h: Added declaration for bu_fchmod(). 2007-11-08 19:09 bob1961 * include/fbserv_obj.h: The fbsl_fd and fbsc_fd member types should be int. 2007-11-08 19:13 bob1961 * src/mged/fbserv.c: If windows call closesocket() instead of close(). 2007-11-08 19:15 bob1961 * src/mged/mged_dm.h: The c_fd of struct client should be int. 2007-11-08 19:16 bob1961 * src/mged/rtif.c: Using bu_fchmod() instead of fchmod. 2007-11-08 19:20 bob1961 * src/fbserv/fbserv.c: No longer calling close() in drop_client() because the call to pkg_close() already does this. 2007-11-08 19:25 bob1961 * src/tclscripts/archer/Archer.tcl, src/tclscripts/lib/RtControl.tcl: Apparently Itk's Toplevel widget now wants the -menu option directly. 2007-11-08 19:26 brlcad * NEWS: bob found/fixed a memory deallocation bug in the raytracers 2007-11-08 19:35 bob1961 * src/rt/do.c: Using bu_fchmod() instead of fchmod(). 2007-11-08 19:39 bob1961 * src/other/libregex/regerror.c: Reformat regerror() slightly. 2007-11-08 19:42 bob1961 * src/other/libpng/png.h: Mods to expose png_read_destroy() and png_write_destroy(). 2007-11-08 19:43 bob1961 * src/other/iwidgets/pkgIndex.tcl: Update to 4.0.2 2007-11-08 19:48 bob1961 * src/libpkg/pkg.c: If windows call closesocket() instead of close(). Mods to use PKG_SEND with the proper number of parameters. Mods to have pkg_open() return the correct type on error. 2007-11-08 19:50 bob1961 * src/fb/fb-pix.c, src/fb/fb-png.c: Using bu_fchmod() instead of fchmod(). 2007-11-08 19:56 bob1961 * src/other/tk/generic/tkFileFilter.c: Initialize a few variables. 2007-11-08 19:59 bob1961 * src/other/tcl/generic/tclExecute.c, src/other/tcl/generic/tclInt.h, src/other/tcl/generic/tclIntDecls.h: Miscellaneous mods for compiling on Windows. 2007-11-08 21:59 erikgreenwald * configure.ac: MFC to pick up solaris fix 2007-11-08 22:06 brlcad * misc/win32-msvc8, misc/win32-msvc8/asc2g, misc/win32-msvc8/asc2g/asc2g.vcproj: testing 2007-11-09 00:16 brlcad * BUGS: the Framebuffer -> Rectangle Area option doesn't seem to work, at least on Mac OS X. it turns disables the embedded framebuffer display (with ogl or X) instead of just showing the rect. otherwise it does render the rect correctly and displays if you turn it back to Framebuffer -> All 2007-11-09 03:46 brlcad * src/libbu/Makefile.am: add fchmod.c to the build and dist 2007-11-09 04:47 brlcad * src/mged/points/points_scan.l: declare extern YYSTYPE yylval to appease the files generated from flex+byacc on stock freebsd 6.3 2007-11-09 13:20 bob1961 * misc/Makefile.am: Added an entry for win32-msvc8 2007-11-09 13:23 bob1961 * misc/win32-msvc8/Makefile.am: Initial check-in. 2007-11-09 13:25 bob1961 * configure.ac: Added an entry for win32-msvc8 2007-11-09 21:09 bob1961 * misc/win32-msvc8/btclsh, misc/win32-msvc8/btclsh/btclsh.vcproj, misc/win32-msvc8/bwish, misc/win32-msvc8/bwish/bwish.vcproj, src/bwish/cadAppInit.c, src/bwish/consoleMain.c, src/bwish/unixMain.c, src/bwish/winMain.c: Initial check-in. 2007-11-09 21:32 bob1961 * src/archer/archer: Archer now depends on bwish. 2007-11-09 21:34 bob1961 * src/other/blt/src/blt.h: Mods for Windows port. 2007-11-09 21:36 bob1961 * src/tclscripts/archer/LoadArcherLibs.tcl: Archer now depends on bwish so this script gets simplified. 2007-11-09 21:41 bob1961 * src/other/blt/src/bltVecMath.c: Temporary hack for compiling on Windows. 2007-11-10 15:10 brlcad * src/libtclcad/tclcadAutoPath.c: make tclcad_tcl_library() print the path if we're in debug mode, otherwise have it ensure that the library path is actually set (via TclSetLibraryPath() if needed). clean up and remove some of the win32ness 2007-11-10 15:27 brlcad * src/bwish/tcl.c: just say it failed -- the actual error should already be printed by the init routine 2007-11-10 15:31 brlcad * src/bwish/main.c: yet another restructuring of the init routine. give Tcl_Init() a chance to work before calling tclcad_auto_path() so that we don't end up trying to force the init.tcl that is provided in the source distribution. if there's a system tcl being used, it will gawk at trying to load the wrong init script. this goes even further such that we keep track of which packages init properly and only call tclcad_auto_path() if one of them fail (or when we get to the brl-cad lib loading) using a two-pass loop. 2007-11-10 15:36 brlcad * src/mged/setup.c: drop the disabled code for setting PATH. we can/should now locate execable items individually via bu_brlcad_root instead of relying on PATH. 2007-11-10 15:42 brlcad * src/bwish/main.c: a few too many results reset 2007-11-10 15:48 brlcad * src/mged/setup.c: add a similar new two-pass initialization approach that bwish/btclsh are now using such that tclcad_auto_path is only called if one of the init routines actually fails. this helps ensure that we can use a system tcl (though doesn't take care of itcl loading/version problems). make the error printing a little more consistent as well. 2007-11-10 19:31 brlcad * include/tclcad.h: make tclcad_tcl_library take an interp 2007-11-10 19:42 brlcad * doc/deprecation.txt: make a note that the various *_version variables are all now *_version() functions now since 7.10.2 even though the globals were never intended for external use (they were in headers merely for compilation convenience as they were used across directories in multiple files). additionally added, but not commented is the 'new' brlcad_version() function which isn't as library-specific as the other variables. 2007-11-13 09:41 d_rossberg * include/bu.h, src/libbu/vls.c: protect the vararg function declaration in bu.h with the same defines as used in the definition in libbu/vls.c remove redundant declaration in libbu/vls.c 2007-11-13 09:43 d_rossberg * src/libbu/libbu.dsp: add fchmod.c (according to libbu/Makefile.am) 2007-11-13 09:51 d_rossberg * src/librt/db_open.c: It is hard to decide in db_close if bu_close_mapped_file should be called or not: db_open calls bu_open_mapped_file, db_clone_dbi does not. Now, db_open calls bu_close_mapped_file if it is a re-use of a previously mapped file. 2007-11-13 12:05 bob1961 * misc/win32-msvc8/dbupgrade, ... lots of other files ..., misc/win32-msvc8/wish, misc/win32-msvc8/wish/wish.vcproj: Initial check-in. 2007-11-13 12:42 bob1961 * src/bwish/cadAppInit.c: Only include windows.h for btclsh on Windows. 2007-11-13 15:57 brlcad * src/other/intaval, src/other/intaval/.cvsignore: ignore tgf-g product 2007-11-13 16:15 brlcad * src/tclscripts/archer/images/Themes/Crystal (Large): remove duplicate theme, src/tclscrtips/archer/images/Themes/Crystal_Large seems to be a complete copy plus later updates, and it's actually included in the dist (this one is not) 2007-11-13 16:21 bob1961 * src/libtclcad/tclcadAutoPath.c: No longer need SEARCH_ARCHER. 2007-11-13 17:32 bob1961 * include/config_win.h: Modify default values for BRLCAD_ROOT and BRLCAD_DATA. 2007-11-13 17:39 erikgreenwald * configure.ac: carry the solaris fixes over from STABLE 2007-11-13 21:15 bob1961 * src/librt/db_open.c: Initialize dbip->dbi_fd in db_open() to -1. 2007-11-14 02:48 brlcad * src/other/tcl/generic/tclInt.h: since our build flags aren't getting passed through to tcl for some reason (which isn't all that bad so far), just reference our common.h directly 2007-11-14 03:50 brlcad * src/other/incrTcl/itcl/Makefile.am: fix dist, apparently no longer a mac dir here? 2007-11-14 03:53 brlcad * src/other/libpng/Makefile.am: wrong ver file in dist 2007-11-14 03:55 brlcad * misc/win32-msvc8/Makefile.am: fix dist, there is no brlcad dir 2007-11-14 06:47 brlcad * src/vdeck/Makefile.loc: pointless and out-of-date old build file removed 2007-11-14 06:49 brlcad * src/util/Makefile.am: add missing remapid manual page to the install/dist 2007-11-14 06:51 brlcad * src/fbed/Makefile.loc, src/libbn/Makefile.loc, src/util/Makefile.loc: more obsolete and out-of-date old build files, now removed 2007-11-14 06:57 brlcad * src/libbu/bomb.c: semaphore protect the crash report being generated on bomb so that only one thread/process is writing out to the log file at a time. only allow one of them to write to the file, depending on whether the log file for that process already exists. we use BU_SEM_MAPPEDFILE since that really shouldn't be in use during bombing. 2007-11-14 07:04 brlcad * NEWS: fixed garbled output during parallel crash reports. if you were on a parallel system with mutiple threads/procs in action when bu_bomb() is invoked, multiple threads/procs ended up writing out to the crash report file. this fix synchronizes the output and removes the garbled interleaving. 2007-11-14 11:50 bob1961 * misc/win32-msvc8/brlcad, misc/win32-msvc8/brlcad/brlcad.sln: Initial check-in. 2007-11-14 12:21 brlcad * misc/win32-msvc8/Makefile.am: there is that brlcad.sln now.. 2007-11-14 17:12 bob1961 * src/rt/viewhide.c, src/rt/viewxray.c: AmbientIntensity is already declared in optical.h 2007-11-15 03:08 brlcad * src/mged/setup.c: AHA! I had added the final tclcad_auto_path() call... just not to the mged init routine. it was only added to bwish (those two really should syncronize their init routines into tclcad or something). this "should" fix the "gui" error. 2007-11-15 04:23 brlcad * src/bwish/main.c: match what mged does order-wise 2007-11-15 13:27 bob1961 * src/mged/qray.c: Mods to use librt's qray routines. This fixes the inability to adjust the format strings and shrinks MGED's code base a tiny bit. Note - after testing on unix the if'ed out code can be removed. 2007-11-15 13:29 bob1961 * include/raytrace.h: Expose a few more functions in librt. 2007-11-15 14:20 bob1961 * misc/win32-msvc8/Makefile.am: Added a few more projects. 2007-11-15 14:28 bob1961 * misc/win32-msvc8/rt_bot_faces, misc/win32-msvc8/rt_bot_faces/rt_bot_faces.vcproj, misc/win32-msvc8/rtarea, misc/win32-msvc8/rtarea/rtarea.vcproj, misc/win32-msvc8/rtcell, misc/win32-msvc8/rtcell/rtcell.vcproj, misc/win32-msvc8/rtfrac, misc/win32-msvc8/rtfrac/rtfrac.vcproj, misc/win32-msvc8/rthide, misc/win32-msvc8/rthide/rthide.vcproj, misc/win32-msvc8/rtpp, misc/win32-msvc8/rtpp/rtpp.vcproj, misc/win32-msvc8/rtrad, misc/win32-msvc8/rtrad/rtrad.vcproj, misc/win32-msvc8/rtrange, misc/win32-msvc8/rtrange/rtrange.vcproj, misc/win32-msvc8/rtray, misc/win32-msvc8/rtray/rtray.vcproj, misc/win32-msvc8/rtsil, misc/win32-msvc8/rtsil/rtsil.vcproj, misc/win32-msvc8/rtweight, misc/win32-msvc8/rtweight/rtweight.vcproj, misc/win32-msvc8/rtxray, misc/win32-msvc8/rtxray/rtxray.vcproj: Initial check-in. 2007-11-15 14:33 bob1961 * misc/win32-msvc8/btclsh/btclsh.vcproj, misc/win32-msvc8/bwish/bwish.vcproj: Increase optimization. 2007-11-15 14:36 bob1961 * misc/win32-msvc8/asc2g/asc2g.vcproj: Turn off debugging. 2007-11-15 14:38 bob1961 * misc/win32-msvc8/brlcad/brlcad.sln: Add 12 more projects. 2007-11-15 14:45 brlcad * NEWS: bob enabled/fixed the embedded framebuffer in the windows port and has added a slew of other missing utilities (too many to enumerate, but it's all of the ray-tracers). 2007-11-15 17:52 erikgreenwald * src/rt/viewedge.c: minor clean-up 2007-11-15 18:01 erikgreenwald * src/rt/viewedge.c: allow users to set arbitrary distance threshholds 2007-11-15 18:06 erikgreenwald * src/rt/rtedge.1: mention the new max_dist option 2007-11-15 18:07 erikgreenwald * NEWS: mention max_dist option for rtedge 2007-11-15 21:56 brlcad * NEWS: reword mostly just so it can be annotated that erik's fix was in response to a support request from karel, and in turn fixes his sf bug 1203346 (Piece of edge omitted) that he first reported back in 2005. 2007-11-15 22:10 brlcad * configure.ac: check for fchmod() 2007-11-15 22:10 brlcad * include/config_win.h: don't have fchmod or _fchmod apparently 2007-11-15 22:14 brlcad * src/other/iwidgets/pkgIndex.tcl: according to the sources, this is still 4.0.1 2007-11-15 22:16 brlcad * src/libbu/fchmod.c: rework bu_fchmod() to use HAVE_FCHMOD feature, falling back to chmod if it's not available. add missing footer. convert to unix line endings. 2007-11-15 22:49 erikgreenwald * src/bwish/cadAppInit.c, src/bwish/winMain.c: dos->unix line terminators 2007-11-16 03:16 brlcad * NEWS: bob fixed a bug exposed with the opendb command where it would crash if you said no to create a new file. it was a windows-specific bug related to trying to close fd 0 (which happened to be the default descriptor value). now set to -1 and is checked. 2007-11-16 04:24 brlcad * ., .cvsignore: on second thought, ignore the install logs 2007-11-16 15:55 erikgreenwald * src/librt/wdb_obj.c: fail gracefully when xpush is called with a primitive, instead of crashing 2007-11-18 01:11 brlcad * NEWS, src/conv/off/g-off.c: fix crash-on-exit bug during clean-up where the NOP curtree being returned wasn't having its magic number set (causing a bu type check to fail) 2007-11-18 01:59 brlcad * src/librt/db_walk.c: if there's no directory pointer, or no callbacks, there's nothing to do 2007-11-18 02:02 brlcad * src/conv/iges/g-iges.c: make sure the object may be looked up 2007-11-18 20:13 brlcad * src/librt/nmg_misc.c: speling 2007-11-19 00:43 brlcad * src/librt/dg_obj.c, src/librt/wdb_obj.c: remove dead code. there's closedb instead of overriding default close command, tol is a wdb_obj command. 2007-11-19 06:44 brlcad * src/librt/nmg_rt_isect.c: prevent a fatal crash/shutdown case, just return class unknown instead of bombing. if the inbound or outbound face use are null, don't try to use them. 2007-11-19 06:49 brlcad * src/librt/nmg_manif.c: there is some nasty bug in nmg_misc.c where nmg_classify_s_vs_s() ends up in an infinite recursion, causing the paint_color to increment out of bounds and eventually resulting in an invalid point_meaning array index. use an unsigned int instead of a char for the counter/arrayindex, but still add an explicit test as to whether the paint_color ends up out of bounds whereby we break out of the iteration. 2007-11-19 07:30 brlcad * src/librt/nmg_rt_isect.c: yet another place to report an unknown class instead of bombing out. 2007-11-19 07:31 brlcad * src/librt/nmg_class.c: if we exhaust the retry count, just give up instead of bombing out. otherwise this can cause havoc for even simple optional operations like trying to fix normals. 2007-11-19 07:36 brlcad * src/librt/nmg_misc.c: if we encounter an invalid shell with bad results, stop processing entirely. this avoids an avalanche of cascaded failures and potential bomb situations where we can usually proceed. this particular problem was encountered during g-iges that had a shell that could not be classified (resulted in infinite loop and array indices that went out of bounds, eventually crashing). 2007-11-19 07:52 brlcad * NEWS: fixed variety of g-iges and other exporter crashes and graceful handling of mesh normal failures. started with a particular model that was failing in the bot's tess() routine during the expensive nmg_fix_normals() processing. turned out that the model was going amuck while trying to determine shell orientation eventually overflowing a char in an inf loop until it crashed. the specific cause of the inf loop wasn't determined, but it does now detect the shell failure and aborts early instead of proceeding and running into a variety of bombing situations. 2007-11-19 15:28 erikgreenwald * NEWS: note fix to xpush 2007-11-19 20:18 brlcad * misc/win32-msvc8/Makefile.am: typo, brrrrcad 2007-11-19 20:41 brlcad * src/tclscripts/Makefile.am, src/tclscripts/archer/Makefile.am, src/tclscripts/lib/Makefile.am, src/tclscripts/mged/Makefile.am, src/tclscripts/nirt/Makefile.am, src/tclscripts/pl-dm/Makefile.am, src/tclscripts/rtwizard/Makefile.am, src/tclscripts/rtwizard/lib/Makefile.am, src/tclscripts/sdialogs/Makefile.am, src/tclscripts/sdialogs/scripts/Makefile.am, src/tclscripts/swidgets/Makefile.am, src/tclscripts/util/Makefile.am: so long as they're in CVS and being provided for the windows folks, include the pkgIndex.tcl and tclIndex files in the source dist. 2007-11-19 20:44 brlcad * src/tclscripts/archer/images/Themes/Crystal/Makefile.am, src/tclscripts/archer/images/Themes/Windows/Makefile.am: few archer resource images files missing from the source dist 2007-11-19 21:44 brlcad * src/rttherm/pixtest.c: update pixtest to the current api, mostly rt->bn changes plus warning quellage 2007-11-19 21:44 brlcad * src/rttherm/Makefile.am: re-enable pixtest compilation since it's now updated but don't install it. 2007-11-19 21:45 brlcad * src/rttherm, src/rttherm/.cvsignore: ignore pixtest 2007-11-19 22:06 brlcad * src/tclscripts/geometree/Makefile.am, src/tclscripts/swidgets/scripts/Makefile.am: missed a couple dirs, include pkgIndex.tcl and tclIndex in the dist 2007-11-19 22:06 brlcad * src/tclscripts/archer/images/Themes/Windows/Makefile.am: missed command.png, add to dist 2007-11-20 05:54 brlcad * src/rt/reshoot.c: anything using the bu offset macros needs to include stddef.h for the standard offset macros. other minor mods for bu_exit. 2007-11-20 05:58 brlcad * src/rt/Makefile.am: add reshoot to the build/install including its manual page documentation. 2007-11-20 05:58 brlcad * src/rt, src/rt/.cvsignore: ignore reshoot product 2007-11-20 06:00 brlcad * src/rt, src/rt/.cvsignore: the version files are no longer generated, remove from ignore list 2007-11-20 06:01 brlcad * src/rt, src/rt/.cvsignore: version file too 2007-11-20 06:03 brlcad * src/rt/Makefile.am: compile hurt, just don't install it. 2007-11-20 06:03 brlcad * src/rt, src/rt/.cvsignore: ignore hurt 2007-11-20 06:05 brlcad * db/Makefile.am: install the terra.dsp file since it's needed by terra.g; adds missing file to dist 2007-11-20 06:07 brlcad * doc/Makefile.am: install the brep.txt file for kicks and wiggles so it's not missing from source dist 2007-11-20 06:13 brlcad * misc/archlinux/Makefile.am: include PKGBUILD.static in the dist if it's in the checkout 2007-11-20 06:13 brlcad * misc/Makefile.am: include doxygen_structure in the dist 2007-11-20 06:18 brlcad * misc/win32-msvc7/Makefile.am, misc/win32-msvc8/Makefile.am: missing files from dist 2007-11-20 06:21 brlcad * regress/Makefile.am: add script files mising from dist 2007-11-20 06:24 brlcad * src/archer/Makefile.am: archer.bat missing from source dist, add it 2007-11-20 16:32 brlcad * src/other/URToolkit/tools/Makefile.am: add rleccube to the build 2007-11-20 16:32 brlcad * src/other/URToolkit/cnv/Makefile.am: readd the disabled converters so they make the dist 2007-11-20 16:33 brlcad * src/other/URToolkit/man/Makefile.am: add pgmtorle manpage 2007-11-20 18:25 brlcad * src/other/tkimg/Makefile.am, src/other/tkimg/base/Makefile.am, src/other/tkimg/png/Makefile.am: add files missing from source dist 2007-11-20 18:28 brlcad * src/other/tcl/libtcl_nil.a: oop, don't put the archive in cvs.. 2007-11-20 18:33 brlcad * src/other/openNURBS/Makefile.am, src/other/tcl/Makefile.am, src/other/tk/Makefile.am: more files missing from the source dist 2007-11-20 18:35 brlcad * src/other/tcl/unix, src/other/tcl/unix/.cvsignore, src/other/tk/unix, src/other/tk/unix/.cvsignore: more generated files to ignore 2007-11-20 18:50 brlcad * src/conv/example_geom.c, src/conv/walk_example.c: rename example_geom.c to walk_example.c to be consistent name-wise with the other *_example program samples. 2007-11-20 19:15 brlcad * src/conv/walk_example.c: restructure cleanup for bu_exit, move struct comments to where they're actually used, cleanup usage, remove progname. 2007-11-20 19:15 brlcad * src/conv/Makefile.am: add walk_example to the build, but don't install it. 2007-11-20 19:18 brlcad * db/Makefile.am: need terra.dsp in the dist 2007-11-20 19:19 brlcad * src/bwish/Makefile.am, src/other/incrTcl/itcl/generic/Makefile.am, src/other/incrTcl/itcl/library/Makefile.am, src/other/incrTcl/itk/library/Makefile.am, src/other/libpng/Makefile.am, src/other/libz/Makefile.am: add more files missing from the source dist 2007-11-20 19:24 brlcad * src/conv, src/conv/.cvsignore: ignore walk_example 2007-11-20 20:25 brlcad * src/librt/plane.h: obsolete private header -- the contents were added to bn.h a long time ago 2007-11-20 20:28 brlcad * src/librt/Makefile.am, src/mged/Makefile.am, src/other/blt/Makefile.am, src/other/blt/demos/Makefile.am, src/other/blt/html/Makefile.am, src/other/blt/library/Makefile.am, src/other/blt/man/Makefile.am, src/other/blt/src/Makefile.am, src/other/libpng/Makefile.am: and now what I think is the final batch of files missing from the source distribution, woot! 2007-11-20 20:43 brlcad * src/other/blt/Makefile.am: blt.mak not blt.make 2007-11-20 20:44 brlcad * sh/Makefile.am: include gforge.sh in dist 2007-11-20 20:44 brlcad * doc/html/manuals/Makefile.am: include eagleCAD.bmp in source dist 2007-11-20 20:47 brlcad * Makefile.am: enhance the dist hook so that it checks whether files listed in CVS are actually included in the source dist that we end up with. makes it a heck of a lot easier to ensure that we have all the files that we're supposed to come time to make a source release. 2007-11-20 20:58 brlcad * src/librt/Makefile.am: gah, one more. add brep-test.cpp to dist 2007-11-21 07:48 brlcad * TODO: itemize the critical items remaining for this release .. bu_bombing, rt parallel crash bug, incrtcl search woes, and testing parallel crash reporting. 2007-11-21 13:50 brlcad * src/other/tcl/Makefile.am, src/other/tk/Makefile.am: if we're building against system tcl/tk, then there's no unix dir to traverse into if we're doing an out-of-src build. this fixes distcheck on said platform situations too 2007-11-21 15:21 erikgreenwald * src/conv/asc-nmg.c, src/conv/iges/iges.c, src/conv/nmg-rib.c, src/conv/proe-g.c, src/conv/stl-g.c, src/conv/tankill-g.c, src/conv/viewpoint-g.c: change bu_bomb to bu_exit when the sky isn't really falling 2007-11-21 15:28 erikgreenwald * src/util/png_info.c: change bu_bomb to bu_exit when the sky isn't really falling. Include stdlib for EXIT_FAILURE. 2007-11-21 15:29 erikgreenwald * src/util/bw-png.c, src/util/dsp_add.c, src/util/png-bw.c, src/util/png-pix.c, src/util/remapid.c: change bu_bomb to bu_exit when the sky isn't really falling 2007-11-21 15:32 erikgreenwald * src/conv/iges/iges.c: char*label isn't munged, so make it const 2007-11-21 15:45 erikgreenwald * src/conv/g-euclid1.c: change bu_bomb to bu_exit when the sky isn't really falling 2007-11-21 15:51 erikgreenwald * src/mged/chgtree.c: change bu_bomb to bu_exit when the sky isn't really falling. Include stdlib for EXIT_FAILURE. 2007-11-21 15:56 erikgreenwald * src/mged/animedit.c, src/mged/clone.c, src/mged/cmd.c, src/mged/edsol.c, src/mged/solids_on_ray.c, src/rttherm/spectrum.c, src/rttherm/ssamp-bw.c, src/rttherm/ssampview.c, src/rttherm/viewtherm.c, src/tab/script-tab.c: change bu_bomb to bu_exit when the sky isn't really falling 2007-11-21 16:04 erikgreenwald * src/rttherm/spectrum.c: stdlib.h needed for EXIT_FAILURE 2007-11-21 16:10 erikgreenwald * src/fb/fb-png.c, src/fb/png-fb.c, src/gtools/beset/fitness.c, src/gtools/beset/population.c, src/gtools/g_qa.c, src/gtools/g_transfer.c: change bu_bomb to bu_exit when the sky isn't really falling 2007-11-21 16:18 erikgreenwald * src/rt/do.c, src/rt/hurt.c, src/rt/opt.c, src/rt/read-rtlog.c, src/rt/reshoot.c, src/rt/view.c, src/rt/view_bot_faces.c, src/rt/viewcell.c, src/rt/viewdir.c, src/rt/viewedge.c, src/rt/viewg3.c, src/rt/viewhide.c, src/rt/viewrange.c, src/rt/viewray.c, src/rt/worker.c: change bu_bomb to bu_exit when the sky isn't really falling 2007-11-21 16:19 erikgreenwald * src/rt/rt.1: note bu_exit() in addition to bu_bomb() 2007-11-21 18:09 erikgreenwald * src/rt/hurt.c: print using bu_exit instead of bu_log 2007-11-21 18:17 erikgreenwald * src/gtools/g_qa.c: print using bu_exit instead of bu_log 2007-11-21 18:19 erikgreenwald * src/gtools/g_qa.c: print using bu_exit instead of bu_log 2007-11-21 18:55 erikgreenwald * src/util/azel.c, src/util/bary.c, src/util/bw-a.c, ... lots of other files ..., src/util/wavelet.c, src/util/yuv-pix.c: exit->bu_exit where applicable 2007-11-21 19:09 erikgreenwald * src/conv/asc-nmg.c, src/conv/asc2g.c, src/conv/bot-bldxf.c, ... lots of other files ..., src/fb/rle-fb.c, src/fb/spm-fb.c: exit->bu_exit where applicable 2007-11-21 19:28 erikgreenwald * src/anim/anim_hardtrack.c, src/anim/anim_orient.c, ... lots of other files ..., src/rttherm/viewtherm.c: exit->bu_exit where applicable 2007-11-21 19:51 erikgreenwald * src/external/ProEngineer/proe-brl.c, src/external/Unigraphics/ug-g.c, src/external/Unigraphics/ug_misc.c, src/mged/bool_rewrite.c, src/mged/cad_boundp.c, src/mged/cad_parea.c, src/mged/comb_bool.h, src/mged/ged.c, src/mged/points/points_parse.y, src/mged/points/points_scan.l, src/mged/rtif.c, src/mged/solids_on_ray.c, src/mged/tedit.c, src/nirt/command.c, src/nirt/if.c, src/nirt/interact.c, src/nirt/nirt.c, src/nirt/parse_fmt.c, src/nirt/read_mat.c, src/nirt/showshot.c, src/remrt/remrt.c, src/remrt/rtsrv.c, src/tab/tabinterp.c, src/tab/tabsub.c, src/vas4/vas4.c, src/vas4/vas_io.c, src/vdeck/vdeck.c, src/vdeck/vproc.c: exit->bu_exit where applicable 2007-11-21 20:38 erikgreenwald * src/conv/asc2g.c, src/conv/dbupgrade.c, src/conv/enf-g.c, src/conv/g4-g5.c, src/conv/g5-g4.c, src/conv/iges/convtree.c, src/conv/iges/iges.c, src/conv/iges/main.c, src/conv/iges/readglobal.c, src/conv/iges/readstart.c, src/conv/iges/readtree.c, src/conv/iges/usage.c, src/conv/proe-g.c, src/conv/stl-g.c, src/conv/viewpoint-g.c, src/fb/cell-fb.c, src/gtools/beset/population.c, src/gtools/g_lint.c, src/gtools/g_qa.c, src/gtools/g_transfer.c, src/gtools/solshoot.c, src/lgt/do_options.c, src/mged/bool_rewrite.c, src/mged/solids_on_ray.c, src/nirt/interact.c, src/nirt/nirt.c, src/proc-db/masonry.c, src/remrt/remrt.c, src/remrt/rtsrv.c, src/rt/do.c, src/rt/hurt.c, src/rt/rtshot.c, src/util/bary.c, src/util/double-asc.c, src/util/imgdims.c, src/util/mst.c, src/util/pixclump.c, src/util/pixcount.c, src/util/pl-dm.c, src/util/remapid.c: collapse bu_log();bu_exit(#,""); as much as possible 2007-11-21 20:49 erikgreenwald * src/anim/anim_hardtrack.c, src/anim/anim_orient.c, ... lots of other files ..., src/vas4/vas_io.c, src/vdeck/vdeck.c, src/vdeck/vproc.c: pass bu_exit a NULL instead of an empty string 2007-11-21 21:21 erikgreenwald * src/rt/rtshot.c: quell type warning on 64b build 2007-11-21 21:46 erikgreenwald * src/mged/fbserv.c: pad with 0's to keep the compiler from complaining about casting 32b int->ptr on 64b systems (and just hope we don't cross 4g) 2007-11-22 06:40 brlcad * src/rt/rt.1: bu_exit() won't generate a core dump because of the bu_setjmp exception handler. never gets to the abort. mention gdb. 2007-11-22 07:04 brlcad * src/rt/rtshot.c: bu_exit formats the exit statement distinctly, don't confuse it as a complete drop-in replacement for bu_log+exit 2007-11-26 12:17 bob1961 * misc/win32-msvc8/asc2g/asc2g.vcproj, ... lots of other files ..., misc/win32-msvc8/wish/wish.vcproj: Updates related to new Tcl/Tk. 2007-11-26 12:19 bob1961 * src/other/incrTcl/itcl/library/pkgIndex.tcl, src/other/incrTcl/itk/library/pkgIndex.tcl: Update version. 2007-11-27 07:44 brlcad * src/other/incrTcl/itcl/pkgIndex.tcl.in, src/other/incrTcl/itk/pkgIndex.tcl.in: unclobber the pkg templates. the IT?.LIB_FILE names aren't presently expanded properly until we use incrtcl's configure. 2007-11-27 07:50 brlcad * configure.ac: use the Tcl TEA build infrastructure, namely the tcl.m4 macro set, so we can locate/load the tcl/tkConfig.sh scripts that can in turn be loaded so we get the correct TCL_VERSION 2007-11-27 08:00 brlcad * configure.ac: set ITCL_LIB_FILE to something half-sensible (even if we probably can't package require archives) 2007-11-27 20:44 brlcad * TODO: review and integrate David Loman's bot-processing and object diagram/labelling scripts. need a better libdm interface for doing textual overlays in mged. 2007-11-27 21:12 brlcad * BUGS: Open dialog in MGED seems rather busted for scrolling and selection 2007-11-27 21:47 brlcad * include/machine.h: comment on the extensive DEPRECATED nature of the defines, types, and values provided by this file indicating whether it's a symbols that's going away, turning into a compile- or run-time check, or simply moving somewhere else. ultimately, this entire header should get absorbed into other headers (particularly those of libbu), though the method differs for each symbol. 2007-11-28 04:00 brlcad * doc/README.IRIX: beef up the IRIX readme with addition details encountered with the problem I ran into with (yet another stupid) libtool linkage problems/bugs. more specifically, libtool was leaving out three librt object files that were in the librt_nil.la convenience library. turns out they were getting prelinked into a '.al' pre-link archive with just the .l nm symbol instead of .lo causing subsequent linkage failures due to the symbols that get left out. 2007-11-28 04:18 brlcad * TODO: make byteoffset and run-time byteorder detections necessary in order for Mac OS X universal binaries to actually work 2007-11-28 13:54 erikgreenwald * TODO: fix (type)(size_t)val hacks 2007-11-28 13:55 erikgreenwald * src/rt/view_bot_faces.c: (type)(size_t)val hack to quell compiler warning... 2007-11-28 14:11 erikgreenwald * src/libbn/mat.c, src/libbn/plane.c, src/libbn/poly.c, src/liboptical/refract.c, src/liboptical/sh_flat.c, src/liboptical/wray.c, src/librt/bool.c, src/librt/g_arb.c, src/librt/g_arbn.c, src/librt/g_ars.c, src/librt/g_bot.c, src/librt/g_bot_include.c, src/librt/g_cline.c, src/librt/g_dsp.c, src/librt/g_ebm.c, src/librt/g_ehy.c, src/librt/g_ell.c, src/librt/g_epa.c, src/librt/g_eto.c, src/librt/g_extrude.c, src/librt/g_half.c, src/librt/g_hf.c, src/librt/g_nmg.c, src/librt/g_part.c, src/librt/g_pg.c, src/librt/g_pipe.c, src/librt/g_rec.c, src/librt/g_rhc.c, src/librt/g_rpc.c, src/librt/g_sketch.c, src/librt/g_sph.c, src/librt/g_submodel.c, src/librt/g_superell.c, src/librt/g_tgc.c, src/librt/g_torus.c, src/librt/g_vol.c, src/librt/g_xxx.c, src/librt/many.c, src/proc-db/kurt.c, src/rt/hurt.c, src/rt/view.c, src/rt/viewedge.c, src/rt/viewpp.c, src/rt/viewrad.c, src/rt/viewscat.c, src/rt/worker.c, src/tclscripts/archer/tclIndex, src/tclscripts/geometree/tclIndex, src/tclscripts/lib/tclIndex, src/tclscripts/mged/tclIndex, src/tclscripts/rtwizard/lib/tclIndex, src/tclscripts/sdialogs/scripts/tclIndex, src/tclscripts/swidgets/scripts/tclIndex: LOCAL->static, per machine.h deprecation list 2007-11-28 20:56 erikgreenwald * src/adrt/libtie/tie.c, src/adrt/libtie/tie.h, src/adrt/libtie/tie_define.h, src/adrt/libtie/tie_kdtree.c, src/adrt/libtie/tie_struct.h: merge of the "other" libtie... 2007-11-28 22:41 bob1961 * misc/win32-msvc8/tclsh/tclsh.vcproj: Add post-build commands for creating an install tree. 2007-11-28 22:43 bob1961 * misc/win32-msvc8/asc2g/asc2g.vcproj, ... lots of other files ..., misc/win32-msvc8/wish/wish.vcproj: Mods to use the new install tree. 2007-11-28 22:45 bob1961 * misc/win32-msvc8/tclsh/library, misc/win32-msvc8/tclsh/library/treeInit.sh: Initial check-in (not finished, checking in for safety) 2007-11-29 06:25 brlcad * misc/win32-msvc8/asc2g/asc2g.vcproj, ... lots of other files ..., misc/win32-msvc8/tkstub/tkstub.vcproj: make the AdditionalIncludeDirectories paths match for Debug and Release 2007-11-29 13:13 bob1961 * misc/win32-msvc8/brlcad/brlcad.sln: Changing the build order to force tclsh to build sooner. Tclsh is used early in the build to create the install tree. 2007-11-29 17:58 brlcad * BUGS: annotate the opengl display manager problems where the display doesn't automatically update any longer if the context is invalidated; mention that the libfb ogl interface seems horribly broken at the moment as it just crashes (consistently, at least on Mac OS X); and mention that the open dialog problems seem to be Mac OS X specific too as they worked under Linux (7.10.1 Mac also seemed to work fine, so something since then). 2007-11-29 18:04 bob1961 * misc/win32-msvc8/tclsh/library/installTree.tcl: Initial check-in. This script will be called by Visual Studio to build the install tree. 2007-11-29 18:06 bob1961 * misc/win32-msvc8/tclsh/tclsh.vcproj: Calling installTree.tcl instead of treeInit.sh. 2007-11-29 18:09 brlcad * misc/win32-msvc8/Makefile.am: include the new scripts in the source dist 2007-11-29 18:11 bob1961 * misc/win32-msvc8/tclsh/library/treeInit.sh: Now using installTree.tcl. 2007-11-29 18:12 brlcad * TODO: annotate 2007-11-29 18:14 brlcad * src/other/tcl/unix/configure.in, src/other/tk/unix/configure.in: 2.52 is our autoconf minimumu, not 2.59 2007-11-29 18:30 brlcad * librt: once again, try to delete these zombie files that have again mysteriously shown up after an update 2007-11-29 19:01 bob1961 * misc/win32-msvc8/tclsh/tclsh.vcproj: Remove command to copy clock.tcl. 2007-11-29 19:08 bob1961 * include/config_win.h: No longer need to define BRLCAD_DATA and BRLCAD_ROOT. 2007-11-29 19:09 bob1961 * src/archer/archer.bat: Archer now uses bwish. 2007-11-29 19:11 bob1961 * src/mged/mged.bat: Define WEB_BROWSER to point to IEXPLORE.EXE. 2007-11-29 19:14 bob1961 * src/tclscripts/archer/Archer.tcl: brlcadDataPath can now be set the same on Windows as it is on Unix. 2007-11-29 19:15 bob1961 * src/tclscripts/mged/mged.tcl, src/tclscripts/mged/openw.tcl: Normalize the path when setting mged_default(html_dir). 2007-11-29 20:09 bob1961 * misc/win32-msvc8/tclsh/library/installTree.tcl: Copying many more files to the install tree. 2007-11-29 20:53 bob1961 * src/tclscripts/lib/tclIndex: Put things back. 2007-11-29 21:37 bob1961 * src/archer/archer: html is no longer under doc on Windows. 2007-11-29 21:44 bob1961 * src/tclscripts/archer/tclIndex: Looks like things accidently got wacked. 2007-11-29 21:55 erikgreenwald * src/adrt/libcommon/unpack.c, src/adrt/librender/plane.c, src/adrt/librender/spall.c: update the tie_init() calls with the new parm 2007-11-29 21:58 brlcad * src/other/incrTcl/itk/pkgIndex.tcl.in: there is no ITK_VERSION, use ITCL_VERSION 2007-11-29 22:22 brlcad * INSTALL: sync options with configure: add dtrace, remove automatic 2007-11-30 22:42 brlcad * src/other/intaval/regtab.cpp: if there's a region entry missing from the material file, just fake an entry for it instead of aborting. this lets processing continue for the rest of the regions that follow at least, and placeholds the unknown region as such. 2007-11-30 23:31 brlcad * BUGS: mged tab-completion doesn't work if the object names have spaces 2007-12-01 15:18 bob1961 * misc/win32-msvc8/libtcl/libtcl.vcproj: Added a call to build the missing include/conf files. 2007-12-01 18:27 bob1961 * misc/win32-msvc8/tclsh/library/installTree.tcl: No longer need to create the missing include/conf files. Added code to create tclIndex files. Added the shareDir variable to simplify things a bit. 2007-12-01 18:30 bob1961 * misc/win32-msvc8/brlcad/brlcad.sln, misc/win32-msvc8/libItcl/libItcl.vcproj, misc/win32-msvc8/tclsh/tclsh.vcproj, misc/win32-msvc8/tclstub/tclstub.vcproj: Mods for getting the installTree script to create tclIndex files. 2007-12-01 18:36 bob1961 * src/tclscripts/ami.tcl: Created the make_tclIndex proc. 2007-12-01 20:21 brlcad * src/lgt/error.c, src/lgt/prnt.c, src/libfb/fb_log.c: mass simplification since we can assume stdarg is available, it's c89. remove the alternatives. 2007-12-01 21:47 brlcad * src/burst/ui.c, src/fb/fbstretch.c, src/fb/gif-fb.c, src/fbserv/fbserv.c, src/gtools/g_qa.c, src/halftone/main.c, src/libbu/crashreport.c, src/libbu/fopen_uniq.c, src/libbu/parallel.c, src/libbu/parse.c, src/libfft/fftc.c, src/libfft/ifftc.c, src/liborle/rle.c, src/libpkg/pkg.c, src/librt/cmd.c, src/librt/db_path.c, src/librt/db_tree.c, src/librt/dg_obj.c, src/librt/g_bot.c, src/librt/g_dsp.c, src/librt/g_ebm.c, src/librt/g_extrude.c, src/librt/g_vol.c, src/librt/nirt.c, src/librt/nmg_bool.c, src/librt/nmg_ck.c, src/librt/nmg_plot.c, src/librt/nmg_rt_isect.c, src/librt/parse.c, src/librt/pr.c, src/librt/tcl.c, src/librt/track.c, src/librt/vdraw.c, src/librt/vlist.c, src/librt/wdb_obj.c: more flawfinder level 4 quellage, mostly str*() buffer boundary limits. 2007-12-01 21:49 brlcad * src/libfft/fftc.c, src/libfft/ifftc.c: don't use bu_exit() so we can still avoid the libbu dependency 2007-12-01 22:22 brlcad * src/librtserver/rtserver.c, src/librtserver/rtserverTest.c: more finder quelling 2007-12-02 14:56 bob1961 * src/archer/archer.bat, src/mged/mged.bat: Minor mod related to install tree changes. 2007-12-02 18:40 bob1961 * misc/win32-msvc8/tclsh/library/installTree.tcl: Create include directory (only BRL-CAD header files for now). 2007-12-02 18:44 bob1961 * misc/win32-msvc8/brlcad/brlcad.sln: Force asc2g to be built last. 2007-12-02 18:46 bob1961 * misc/win32-msvc8/asc2g/asc2g.vcproj: Added post-build commands to create .g files. 2007-12-02 18:48 bob1961 * misc/nsis/brlcad.nsi: Mods to work with new install tree. 2007-12-02 19:33 bob1961 * src/tclscripts/archer/Archer.tcl: Archer's documentation has moved. 2007-12-02 21:51 brlcad * src/mged/cad_boundp.c, src/remrt/rtsrv.c: no longer need the pre-ansi hacks with c89 functionality assumption, remove the non stdard-based variable-length argument functions 2007-12-02 21:56 brlcad * src/mged/track.c: reformat parameters to crname() and crregion() so that they now take a max length parameter. make all the str() funcs use their n-variant to check the string bounds 2007-12-02 22:05 brlcad * src/mged/cad_boundp.c: want the other HAVE_STDARG_H section 2007-12-02 22:34 brlcad * src/mged/animedit.c, src/mged/chgmodel.c, src/mged/clone.c, ... lots of other files ..., src/util/sgi-pix.c, src/util/ttcp.c: woo hoo, finally got through most of the flawfinder level 4 issues related to string buffer limits (leaving issues that can probably/mostly be ignored). these changes mostly invovlved using the n-variants of the various string processing functions so that the buffers being written to are properly clamped. 2007-12-02 22:43 brlcad * src/rt/rtscale.c: strncat has to account for what's already in there 2007-12-02 22:45 brlcad * TODO: most of the flawfinder issues that we should probably care about are all pretty much accounted for now -- having reviewed all the level 4 and 5 issues individually 2007-12-02 22:50 brlcad * configure.ac, include/config_win.h: no longer need the vprintf test, c89 is a given and all usages of HAVE_VPRINTF are gone 2007-12-02 23:16 brlcad * src/conv/iges/g-iges.c, src/librt/comb.c, src/mged/clone.c, src/proc-db/brep_cube.cpp, src/proc-db/brep_simple.cpp: make sure we check db_lookup() for a null return value throughout and do something sensible 2007-12-03 09:33 d_rossberg * src/libbu/brlcad_path.c: do not return a pointer to a temporary variable, made variable static 2007-12-03 09:36 d_rossberg * src/librt/dg_obj.c: typing error in preceding commit 2007-12-03 11:49 brlcad * src/conv/iges/g-iges.c: argv not ptr 2007-12-03 15:07 brlcad * NEWS: Daniel fixed a mged start-up initialization bug on Windows where a local var was being returned for the brlcad_data path. yuck. 2007-12-03 15:34 brlcad * src/libfb/if_X.c, src/libfb/if_X24.c, src/libfb/if_ogl.c, src/libfb/if_tk.c, src/libfb/if_wgl.c, src/libwdb/strsol.c: quell empty compilation unit warning 2007-12-03 17:48 erikgreenwald * src/mged/dodraw.c: silly coder, snprintf needs a length, too 2007-12-03 17:51 brlcad * src/mged/dodraw.c: double oops, snprintf doesn't need %s size limits 2007-12-03 18:47 brlcad * configure.ac: disable the new freetype font system in Tk 8.5 for now until they're final -- this should 'fix'/avoid a bug reported by dwayne regarding the mged font manager not working right in 7.10.4 2007-12-03 21:55 brlcad * src/conv/proe-g.c: be more consistent on the buffer lengths and limits (needs testing on the name generation) using MAX_LINE_LENGTH for the names too 2007-12-03 22:07 brlcad * src/external/ProEngineer/proe-brl.c: move the format string to the (single) use 2007-12-03 22:24 erikgreenwald * src/adrt/libtie/Makefile.am, src/adrt/libtie/tie.c, src/adrt/libtie/tie.h, src/adrt/libtie/tie_define.h, src/adrt/libtie/tie_kdtree.c, src/adrt/libtie/tie_kdtree.h: build execution paths for both single and double precision floats in the same lib 2007-12-03 22:27 brlcad * src/fb/fbstretch.c: don't need the HAVE_STDARG_H checks any more, keep it simple 2007-12-03 22:27 brlcad * src/fb/fbfade.c: const labelification 2007-12-03 22:28 brlcad * src/fb/cell-fb.c: make sure we don't go too far on the buffer 2007-12-03 22:29 brlcad * src/fb/fbstretch.c: const lables 2007-12-03 22:30 erikgreenwald * src/adrt/libtie/tie.c, src/adrt/libtie/tie.h: make triangle degeneracy testing a runtime selectable feature 2007-12-03 22:57 brlcad * src/burst/extern.h, src/burst/prnt.c, src/fb/fbfade.c, src/fb/fbstretch.c, src/fb/gif-fb.c, src/fbed/extern.h, src/fbed/prnt.c, src/lgt/error.c, src/lgt/extern.h, src/lgt/prnt.c, src/libbu/color.c, src/libbu/log.c, src/libbu/mro.c, src/libbu/vls.c, src/libfb/fb_log.c, src/other/jove/jove.h, src/util/remapid.c: c89 simplification. assume stdarg is a given. 2007-12-04 00:43 brlcad * src/fb/cell-fb.c: strNcat 2007-12-04 00:56 brlcad * src/fb/fbstretch.c: too many lines 2007-12-04 01:31 brlcad * include/fb.h: don't need the non __STDC__ decls any longer, include bu.h in order to declare the stdarg compiler hint 2007-12-04 01:36 brlcad * configure.ac, include/bu.h, include/config_win.h, src/fbserv/fbserv.c, src/libbu/log.c, src/libbu/mro.c, src/libbu/vls.c: remove the last traces of stdarg/vararg testing. c89 conformance provides the stdarg interface. 2007-12-04 13:08 brlcad * HACKING: update to the details on the C++age since there is some now 2007-12-04 16:10 brlcad * src/libfb/Makefile.am, src/libfb/fbserv_obj_win32.c: looks like the windows build is using the real fbserv_obj.c file now, no longer need the fbserv_obj_win32.c empty stub 2007-12-04 16:14 brlcad * src/libfb/libfb.dsp: let vc6 try to use the fbserv_obj.c file too instead of the empty stub 2007-12-04 19:54 brlcad * Makefile.am, sh/make_dmg.sh, sh/make_pkg.sh, sh/make_tar.sh: update the helper build scripts to the new version variables now that MAJOR_, MINOR_, and PATCH_ are no longer being individually subst'd. simplify them to only require a single version number in triplet form instead of separated out individually (only mac make_pkg script needed it broken up anyways). 2007-12-04 21:16 erikgreenwald * src/adrt/libtie/tie.c, src/adrt/libtie/tie_kdtree.c: removal of the "bit hack" 2007-12-04 21:45 erikgreenwald * src/libtclcad/tclcadAutoPath.c: remove garbage in preprocessor line 2007-12-04 21:46 bob1961 * src/other/tk/library/panedwindow.tcl: For the moment catch the " identify " commands. Every now and then they like to make noise. 2007-12-04 21:48 bob1961 * src/tclscripts/archer/Archer.tcl: Add a catch around utility plugin creation. 2007-12-04 21:52 bob1961 * misc/nsis/brlcad.nsi: Do a recursive copy of the bin dir. 2007-12-04 21:57 erikgreenwald * src/adrt/libtienet/tienet.h, src/adrt/libtienet/tienet_master.c: minimize use of tie.h 2007-12-04 21:58 bob1961 * misc/win32-msvc8/tclsh/library/installTree.tcl: Copy visual studio redist files. 2007-12-04 21:59 erikgreenwald * src/adrt/libtie/tie_struct.h: shuffle and pad the tie_s struct for slightly safer use outside of the dual-path environment... 2007-12-04 22:00 bob1961 * misc/win32-msvc8/asc2g/asc2g.vcproj, ... lots of other files ..., misc/win32-msvc8/wish/wish.vcproj: Mods to NOT treat wide chars as a built-in type. 2007-12-04 22:43 brlcad * sh/make_dmg.sh: bah, convert the du size to megabytes for hdiutil, otherwise just caps out at a gb 2007-12-05 05:04 brlcad * src/tclscripts/geometree/GeometryBrowser.tcl: go ahead and have it use draw instead of e so it's clear what's going on to new users 2007-12-05 05:27 brlcad * configure.ac: let tcl and tk know where they're coming from, and don't cache the results of tclConfig.sh or tkConfig.sh so that Tcl's subconfigure will detect properly itself when we're compiling tcl/tk 2007-12-05 15:33 erikgreenwald * src/adrt/libtienet/tienet.h, src/adrt/libtienet/tienet_define.h, src/adrt/libtienet/tienet_master.c, src/adrt/libtienet/tienet_master.h, src/adrt/libtienet/tienet_slave.c, src/adrt/libtienet/tienet_slave.h, src/adrt/libtienet/tienet_util.c, src/adrt/libtienet/tienet_util.h: upgrade to the updated version 2007-12-05 15:34 erikgreenwald * src/adrt/Makefile.am: always build libtienet 2007-12-05 17:03 erikgreenwald * configure.ac: AC_DEFINEs for python and SDL 2007-12-05 17:17 erikgreenwald * src/adrt/Makefile.am: add libutil to the "always compile" list. Back out the += for those using automake 1.6. 2007-12-05 17:18 erikgreenwald * src/adrt/libutil/display.c, src/adrt/libutil/display.h: workaround for lack of SDL 2007-12-05 17:19 erikgreenwald * src/adrt/libutil/Makefile.am, src/adrt/libutil/util_image.c, src/adrt/libutil/util_image.h, src/adrt/libutil/util_math.c, src/adrt/libutil/util_math.h, src/adrt/libutil/util_rand.c, src/adrt/libutil/util_rand.h: merge in the updated libutil stuff (introduces capability duplication for now, but no symbol conflicts) 2007-12-05 17:24 erikgreenwald * src/adrt/libutil/util_image.c, src/adrt/libutil/util_image.h, src/adrt/libutil/util_math.c, src/adrt/libutil/util_math.h, src/adrt/libutil/util_rand.c, src/adrt/libutil/util_rand.h: decorate the new sources 2007-12-05 17:38 erikgreenwald * src/adrt/libutil/Makefile.am: rename libutil to libtieutil, to avoid any possible confusion 2007-12-05 18:49 erikgreenwald * src/adrt/Makefile.am: always build libtexture. Include the new adrt headers 2007-12-05 18:50 erikgreenwald * src/adrt/adrt.h, src/adrt/adrt_struct.h: Include the new adrt headers 2007-12-05 18:53 erikgreenwald * src/adrt/libtexture/Makefile.am, src/adrt/libtexture/texture.h, src/adrt/libtexture/texture_blend.c, src/adrt/libtexture/texture_blend.h, src/adrt/libtexture/texture_bump.c, src/adrt/libtexture/texture_bump.h, src/adrt/libtexture/texture_camo.c, src/adrt/libtexture/texture_camo.h, src/adrt/libtexture/texture_checker.c, src/adrt/libtexture/texture_checker.h, src/adrt/libtexture/texture_clouds.c, src/adrt/libtexture/texture_clouds.h, src/adrt/libtexture/texture_gradient.c, src/adrt/libtexture/texture_gradient.h, src/adrt/libtexture/texture_image.c, src/adrt/libtexture/texture_image.h, src/adrt/libtexture/texture_internal.h, src/adrt/libtexture/texture_mix.c, src/adrt/libtexture/texture_mix.h, src/adrt/libtexture/texture_perlin.c, src/adrt/libtexture/texture_perlin.h, src/adrt/libtexture/texture_stack.c, src/adrt/libtexture/texture_stack.h: merge in the updated libtexture stuff 2007-12-05 18:54 erikgreenwald * src/adrt/bench/Makefile.am, src/adrt/isst/master/Makefile.am, src/adrt/isst/observer/Makefile.am, src/adrt/isst/slave/Makefile.am, src/adrt/rise/master/Makefile.am, src/adrt/rise/observer/Makefile.am, src/adrt/rise/slave/Makefile.am: reflect the libutil->libtieutil change 2007-12-05 19:12 erikgreenwald * src/adrt/libutil/Makefile.am, src/adrt/libutil/util_image.c, src/adrt/libutil/util_image.h, src/adrt/libutil/util_math.c, src/adrt/libutil/util_math.h, src/adrt/libutil/util_rand.c, src/adrt/libutil/util_rand.h: remove duplication 2007-12-05 19:14 erikgreenwald * src/adrt/libtexture/Makefile.am, src/adrt/libtexture/texture_bump.c, src/adrt/libtexture/texture_camo.c, src/adrt/libtexture/texture_checker.c, src/adrt/libtexture/texture_clouds.c, src/adrt/libtexture/texture_gradient.c, src/adrt/libtexture/texture_image.c, src/adrt/libtexture/texture_mix.c, src/adrt/libtexture/texture_perlin.c: util_math.h -> umath.h 2007-12-05 19:35 erikgreenwald * src/adrt/libutil/Makefile.am: make a libtool archive, with SDL deps noted 2007-12-05 19:49 erikgreenwald * src/adrt/libutil/umath.h: common.h doesn't seem to be needed 2007-12-05 19:53 erikgreenwald * src/adrt/bench/Makefile.am, src/adrt/isst/master/Makefile.am, src/adrt/isst/observer/Makefile.am, src/adrt/isst/slave/Makefile.am, src/adrt/rise/master/Makefile.am, src/adrt/rise/observer/Makefile.am, src/adrt/rise/slave/Makefile.am: libtieutil is now a libtool archive 2007-12-05 20:44 erikgreenwald * src/adrt/Makefile.am: always build librender 2007-12-05 21:24 erikgreenwald * src/adrt/librender/Makefile.am, src/adrt/librender/camera.c, src/adrt/librender/camera.h, src/adrt/librender/component.c, src/adrt/librender/cut.c, src/adrt/librender/cut.h, src/adrt/librender/depth.c, src/adrt/librender/flat.c, src/adrt/librender/flos.c, src/adrt/librender/flos.h, src/adrt/librender/grid.c, src/adrt/librender/hit.c, src/adrt/librender/normal.c, src/adrt/librender/path.c, src/adrt/librender/phong.c, src/adrt/librender/render.h, src/adrt/librender/render_internal.h, src/adrt/librender/render_util.c, src/adrt/librender/spall.c, src/adrt/librender/surfel.c, src/adrt/librender/surfel.h: merge in updated librender 2007-12-05 21:52 erikgreenwald * src/adrt/librender/render_internal.h: redefine the plane method 2007-12-05 22:01 erikgreenwald * src/adrt/bench/Makefile.am, src/adrt/isst/master/Makefile.am, src/adrt/isst/observer/Makefile.am, src/adrt/isst/slave/Makefile.am, src/adrt/libcommon/Makefile.am, src/adrt/libcommon/unpack.h, src/adrt/librender/camera.c, src/adrt/librender/camera.h, src/adrt/librender/hit.c, src/adrt/librender/plane.c, src/adrt/librender/render_internal.h, src/adrt/libutil/Makefile.am, src/adrt/rise/master/Makefile.am, src/adrt/rise/observer/Makefile.am, src/adrt/rise/slave/Makefile.am: include path shtuff 2007-12-06 03:21 brlcad * src/tclscripts/geometree/geometree.tcl: don't assume /bin/tclsh, ask /usr/bin/env like a good boy 2007-12-06 05:23 brlcad * src/adrt/libtienet/tienet_master.c: check the length 2007-12-06 14:09 brlcad * src/burst/error.c, src/burst/extern.h, src/burst/prnt.c, src/bwish/winMain.c: const format specifier 2007-12-06 14:44 brlcad * src/fb/Makefile.am, src/fb/fbfade.c, src/fb/fbstretch.c, src/fb/gif-fb.c, src/fb/ioutil.c: refactor the handful of failure and printing functions that are common to gif-fb, fbfade, and fbstretch into their own ioutil.c file to reduce the repetition. clean-up the interface while we're at it with some constness. 2007-12-06 14:53 brlcad * src/gtools/g_qa.c: put the usage string into the usage func, don't need a progname var when there's a bu_getprogname() 2007-12-06 15:06 brlcad * include/fb.h, src/libfb/fb_log.c: make fb_log()'s format specifier constant 2007-12-06 15:08 brlcad * doc/deprecation.txt: denote the fact that fb_log format became const 2007-12-06 15:14 brlcad * src/lgt/error.c: const fb_log 2007-12-06 15:15 brlcad * src/lgt/extern.h, src/lgt/prnt.c: constness 2007-12-06 15:20 brlcad * src/libbn/axis.c: add a check to make sure we don't exceed available buffer space for the dynamic number-of-digit printing 2007-12-06 15:26 brlcad * src/libbu/brlcad_path.c: use bu_which() to find pwd instead of relying on popen 2007-12-06 15:49 brlcad * src/libbu/parallel.c: bu_get_load_average() has a horrible implementation but fortunately we don't even use it. mark it deprecated. 2007-12-06 16:11 brlcad * doc/deprecation.txt, src/libbu/brlcad_path.c, src/libbu/bu_tcl.c: finally obsolete bu_brlcad_path() and bu_tcl_brlcad_path(), which were marked deprecated back in 7.4; callers should now be using bu_brlcad_data() and bu_brlcad_root() depending on the nature of the lookup. 2007-12-06 17:37 erikgreenwald * src/adrt/librender/Makefile.am: allow dynamic lib to be build 2007-12-06 17:37 erikgreenwald * src/adrt/librender/camera.h, src/adrt/librender/render_internal.h: fixes for third party consumers 2007-12-06 17:39 erikgreenwald * src/libbu/brlcad_path.c: add missing closing paren. 2007-12-06 17:54 erikgreenwald * src/fbed/prnt.c, src/fbserv/fbserv.c: change fb_log definition to match the prototype in fb.h 2007-12-06 18:23 erikgreenwald * src/adrt/librender/Makefile.am, src/adrt/librender/camera.c, src/adrt/libutil/Makefile.am, src/adrt/libutil/camera.c: use bu_avail_cpu()s instead of the FreeBSD-only get_procs() version 2007-12-06 18:28 erikgreenwald * src/adrt/ChangeLog, src/adrt/acinclude.m4, src/adrt/configure.ac: unnecessary files 2007-12-06 18:28 erikgreenwald * src/adrt/Makefile.am: remove unnecessary files 2007-12-06 19:02 erikgreenwald * src/adrt/librender/flos.c, src/adrt/librender/path.c: hack to quell warnings, surely breaks in double precision mode 2007-12-06 20:08 erikgreenwald * src/adrt/librender/Makefile.am, src/adrt/librender/plane.c, src/adrt/librender/plane.h, src/adrt/librender/render.h: plane is now cut 2007-12-06 20:09 erikgreenwald * src/adrt/librender/cut.c: bring in cleanup that was done on plane.c 2007-12-06 20:11 erikgreenwald * src/adrt/librender/render_internal.h: RENDER_METHOD_PLANE is a synonym for RENDER_METHOD_CUT now 2007-12-06 20:15 erikgreenwald * src/adrt/adrt.h: mirror the mesh hit/select in libcommon (pre-migration) 2007-12-06 20:15 erikgreenwald * src/adrt/isst/isst.h: build the isst/adrt synonyms 2007-12-06 20:17 erikgreenwald * src/adrt/isst/master/master.c: use the new tienet api... sorta... 2007-12-06 20:17 erikgreenwald * src/adrt/isst/slave/Makefile.am, src/adrt/isst/slave/load.c, src/adrt/isst/slave/load.h: bring in the mysql load stuff (kinda) before removing it... O.o 2007-12-06 20:18 erikgreenwald * src/adrt/libcommon/env.c, src/adrt/libcommon/unpack.c: plane is now cut 2007-12-06 20:25 brlcad * BUGS: once again, rtwizard seems to be busted. getting bad screen distance ":41.25" processing the -width option on the itk_component(bar) on the RTWizard::FeedbackDialog 2007-12-06 20:36 erikgreenwald * src/adrt/isst/slave/slave.c: bring in changes from the new adrt stuff 2007-12-07 04:50 brlcad * include/bu.h: remove bu_brlcad_path and bu_tcl_brlcad_path decls, they're now obsolete 2007-12-07 05:10 brlcad * src/other/blt/src/blt.h: quell compilation warning about data definition having no type or storage class in libtclcad when including the blt.h header 2007-12-07 05:17 brlcad * src/libbu/semaphore.c: don't mention compat4.h 2007-12-07 05:19 brlcad * src/librt/bomb.c: compat4.h is no longer needed or wanted, pretend it doesn't exist .. so don't undefine things it might define 2007-12-07 05:22 brlcad * src/libbu/libbu.3: remove the RECOMMENDATION section that refers to compat4.h and updating sources several major versions old 2007-12-07 05:31 brlcad * NEWS: update the section on deprecation with a pointer to the doc/deprecation.txt transcript. remove the old blather about compat4.h and mged commands. 2007-12-07 05:41 brlcad * include/raytrace.h: provide the args for a_overlap() so it can perform type checks 2007-12-07 05:45 brlcad * src/rttherm/viewtherm.c: set the app's log overlap handler to rt_silent_logoverlap() instead of setting the overlap callback to rt_overlap_quietly() 2007-12-07 06:32 brlcad * doc/deprecation.txt, include/raytrace.h, src/librt/bool.c: rt_overlap_quietly() is now finally obsolete. no longer need the _rt_defoverlap since the verbosity flag is no longer used anyways as well. apps instead can set the a_logoverlap callback to a logging routine (like rt_silent_logoverlap()) as the overlap handler is supposed to be silent. 2007-12-07 06:36 brlcad * doc/deprecation.txt, include/Makefile.am, include/compat4.h: compat4.h is now finally fully obsolete. it's been deprecated since 5.0 and with a compilation pragma warning now for several major release. it's time. just let it go. buh-bye. 2007-12-07 06:42 brlcad * doc/deprecation.txt, src/libbn/mat.c: and then there were none. make the last of the pre-7.0 deprecations now finally obsolete including bn_mat_zero(), bn_mat_idn(), and bn_mat_copy(). they all have equivalent macros provided by the api. 2007-12-07 06:51 brlcad * src/librt/bool.c: reorder the functions so that no declarations are necessary. 2007-12-07 06:53 brlcad * src/librt/bool.c: doxygenify 2007-12-07 06:58 brlcad * src/libfft/fftc.c, src/libfft/ifftc.c: quell warning 2007-12-07 07:03 brlcad * src/libpkg/pkg.c: use stat() instead of access() (and it's a c89 function) 2007-12-07 14:21 brlcad * src/other/URToolkit/cnv/rlatorle.c, src/other/URToolkit/tools/into.c, src/other/URToolkit/tools/rleaddcom.c, src/other/URToolkit/tools/rlecat.c, src/other/libutahrle/rle_hdr.c, src/other/libutahrle/scanargs.c: apply a modified form of Pedro F. Giffuni - giffunip's sf patch [ 1845889 ] "Minor fixes to URT" (seems half his changes were already applied) 2007-12-07 14:21 brlcad * src/other/URToolkit/tools, src/other/URToolkit/tools/.cvsignore: ignore rleccube 2007-12-07 14:36 brlcad * NEWS: applied a modified form of Pedro F. Giffuni - giffunip's sf patch [ 1845889 ] "Minor fixes to URT" (seems half his changes were already applied). this fixed bugs in several Utah Raster Toolkit including libutahrle and a handful of the tools 2007-12-07 14:40 brlcad * src/rt/rt.1: apply a modified form of Karel Kulhavy - clock3's sf patch [ 1837469 ] "man rt" documentation tweak that clarifies that the -d option returns zero as the distance when it misses. 2007-12-07 16:19 erikgreenwald * src/librt/librt.3: remove mention of the obsolete (and removed) rt_overlap_quietly function 2007-12-07 17:47 erikgreenwald * src/mged/clone.c: record Dwayne's wishlist 2007-12-07 17:48 erikgreenwald * TODO: mention the todo list in clone.c 2007-12-07 18:12 brlcad * src/librt/g_bot.c: limit the length of the face_mode copy into the bizzare use of bot_data (where we intentionally go past the sizeof(struct bot_rec) with the bot data for v4 files) 2007-12-08 22:17 brlcad * configure.ac: check for grp.h, check for geteuid() and getegid() -- used by libbu/stat.c for new file permission functions 2007-12-08 22:32 brlcad * include/config_win.h: clean up and expand the permission/access/chmod defines for the S_I* defines so we can do the same on windows we do on non-windows. add missing X_OK and remove the two bogus ones for eof and HAVE_PWD_H 2007-12-08 22:42 brlcad * include/bu.h, src/libbu/stat.c: add new file stat routines to libbu, bu_file_readable() and bu_file_writable() and bu_file_executable(), for determining permissions on a given file name for the current user. check the appropriate user, group, or other permission, and should work on windows too. these routines make it easy to completely avoid access() as well as a means to improve error checking throughout. 2007-12-08 22:52 brlcad * src/librt/wdb_obj.c: don't use access, use the new bu_file_readable() function instead 2007-12-08 22:53 brlcad * src/libbu/stat.c: remove the access() approach in bu_file_exists() and take stat() for granted since it's c89 and doesn't have the security problems access has 2007-12-08 22:56 brlcad * src/libbu/stat.c: just in case we don't have the access() interface defines for R_OK, W_OK, and X_OK that we're privately using, define them if needed 2007-12-08 22:57 brlcad * configure.ac: don't check for or use access(), leave a note that it's intentional. begin a comment about the functions we're specifically assuming are available (the well-behaved ones from c89). 2007-12-08 23:02 brlcad * include/config_win.h: don't need HAVE_STAT or HAVE_ACCESS any more 2007-12-08 23:02 brlcad * src/libbu/stat.c: don't need to check for HAVE_STAT, assume it's available per c89 2007-12-08 23:03 brlcad * configure.ac: don't check for stat(). allow/begin slow and careful dechecking of well-behaved c89 functions. 2007-12-08 23:06 brlcad * src/adrt/isst/master/isst_python.c: shouldn't need to undef HAVE_STAT, no longer checked (not that it should matter) 2007-12-08 23:16 brlcad * src/mged/ged.c: use the new bu_file_readable() routine instead of access() along with bu_file_exists() when unable to open a new file. 2007-12-08 23:21 brlcad * src/remrt/remrt.c: use the new bu_file_writable() routine instead of access() along with bu_file_exists() when creating the new file. 2007-12-08 23:23 brlcad * src/util/bw-rle.c, src/util/pix-orle.c, src/util/pix-rle.c: use bu_file_exists() instead of access() 2007-12-09 00:27 brlcad * src/libbu/fopen_uniq.c: add comment that we can't call bu_log here because of the semaphore 2007-12-09 00:36 brlcad * src/halftone/main.c: quell warning 2007-12-09 00:37 brlcad * src/mged/animedit.c: check buffer length whilest printing into the buffers 2007-12-09 00:37 brlcad * src/libbu/semaphore.c: use mkstemp instead of mktemp since we can (old sgi section) 2007-12-09 00:49 brlcad * src/util/remapid.c: looks like the libbu/file.c functions were just dumped into here. rename the functions to not look like they're still bu_ functions. ugly hacked file. no donut for pjt. 2007-12-09 00:51 brlcad * include/bu.h: wow, there are even two copies of the old file.c function decls, and none of them exist any longer. -1 donut for pjt 2007-12-09 04:01 brlcad * src/libbu/stat.c: handle null/empty path 2007-12-09 04:03 brlcad * src/libbu/stat.c: geez, finish typing before hitting the send 2007-12-09 17:32 brlcad * src/libbu/stat.c: heh, oops -- stat returns 0 on success, -1 on failure. 2007-12-09 21:40 brlcad * src/libbu/temp.c: initial addition and implementation of bu_temp_file() -- a relatively secure means to create a temporary file for use by applications. this implementation should work on most platforms including Windows that support either mkstemp (POSIX) or tmpfile_s (ISO/IEC TR 24731-1) and should conform to the CERT secure coding of temporary file recommendations. bu_temp_file() presently doesn't take any arguments, though that may change if needed by apps. it does, however, auto-delete the temporary files on close through an atexit handler and a linked list of temp files that have been opened. presently creates temp files in the first identifiable/usable temp dir searching TMPDIR, TEMP, TMP, /tmp, and the current directory, creating a file named BRL-CAD_temp_XXXXXXX if using mkstemp. 2007-12-09 21:40 brlcad * src/libbu/Makefile.am: add temp.c to the build 2007-12-09 21:41 brlcad * src/libbu/libbu.dsp: add temp.c to the vc6 build 2007-12-09 21:43 brlcad * include/bu.h: use BU_ASSERT instead of assert -- though it's questionable as to whether BU_LIST_CLOSE needs to even exist.. it's apparently not used anywhere. 2007-12-09 21:45 brlcad * include/config_win.h: windows have tmpfile_s() .. declare the underscored TMP_MAX_S 2007-12-09 21:46 brlcad * configure.ac: need to check fo mkstemp() and tmpfile_s() for the new bu_temp_file() interface. 2007-12-09 21:48 brlcad * misc/win32-msvc7/libbu/libbu.vcproj, misc/win32-msvc8/libbu/libbu.vcproj: add temp.c to the vc7 and vc8 build 2007-12-10 14:21 brlcad * src/adrt/libtie/tie_define.h, src/adrt/libtienet/tienet_util.h: include stdint.h for the various c99 standard integer types. this should fix sf bug [ 1847278 ] "Fails to compile at tienet" from James Millar - j_dmillar. 2007-12-10 21:56 bob1961 * src/tclscripts/archer/Archer.tcl: Removed a few unused methods. 2007-12-10 22:00 bob1961 * src/tclscripts/archer/ArcherCore.tcl: Initial check-in. This is intended to be the base class for application classes (i.e. Archer). 2007-12-11 07:51 brlcad * include/config_win.h: indent consistency 2007-12-13 16:03 bob1961 * src/tclscripts/archer/ArcherCore.tcl: Added a few more methods. Also some minor cleanup. 2007-12-13 16:19 brlcad * src/tclscripts/archer/Makefile.am: add ArcherCore.tcl to the dist 2007-12-13 20:01 brlcad * include/bn.h: remove authorship/origin info from header, innaccurately documents what is already tracked by version control 2007-12-13 20:53 bob1961 * src/tclscripts/archer/Archer.tcl: Modifications to use ArcherCore. Note - there are still a few issues that need attention. 2007-12-13 20:58 bob1961 * src/tclscripts/archer/CombEditFrame.tcl, src/tclscripts/archer/GeometryEditFrame.tcl, src/tclscripts/archer/LoadArcherLibs.tcl, src/tclscripts/archer/pkgIndex.tcl: Mods related to using ArcherCore. 2007-12-13 20:59 bob1961 * src/tclscripts/archer/ArcherCore.tcl: Removed a debug statement. 2007-12-13 21:04 bob1961 * src/bwish/cadAppInit.c: Minor cleanup. 2007-12-13 21:05 bob1961 * include/config_win.h: define uid_t and gid_t 2007-12-13 21:06 bob1961 * include/tclcad.h: declare Tclcad_Init 2007-12-13 21:08 bob1961 * misc/win32-msvc8/libtclcad/libtclcad.vcproj: Added tclcad.c 2007-12-13 21:10 bob1961 * src/archer/archer: Mods related to using ArcherCore 2007-12-13 21:11 bob1961 * src/libtclcad/tclcad.c: Initial check-in. This provides the interface for dynamically loading via Tcl. 2007-12-13 21:42 brlcad * src/libbu/temp.c: modify bu_temp_file() so that we can capture the name of the temporary file that was created along with the opened file descriptor. this requires ditching tmpfile_s() entirely on windows since it returns anonymous files, instead providing an implementation of mkstemp() for platforms like Windows that don't provide it. 2007-12-14 10:49 brlcad * src/libbu/temp.c: clean up the list of temp directories that we try to identify so they can all be treated/tested the same in a loop. the interface should be pretty solid now. 2007-12-14 11:03 brlcad * src/libbu/temp.c: document the bu_temp_file() parameters, don't truncate the filename if it won't fit. 2007-12-14 11:04 brlcad * include/bu.h: declare the two parameters to bu_temp_file() for storing the filename 2007-12-14 11:39 brlcad * src/libbu/temp.c: in order for the filename to be useful, we need to not automatically unlink it right after it's opened -- add it to the closure linked list and unlink the file later during application exit. 2007-12-14 12:25 brlcad * src/mged/chgmodel.c: use bu_temp_file() instead of mktemp/mkstemp for generalized temp file management. fixes assumption that there's a /tmp and that we can write to C:\ on windows 2007-12-14 12:31 brlcad * src/mged/mater.c: use bu_temp_file() instead of mktemp/mkstemp for better generalized temp file management. fixes assumption that there's a /tmp and that we can write to C:\ on windows and keeps track of the temp file better 2007-12-14 13:05 brlcad * src/mged/red.c: use bu_temp_file() instead of mktemp/mkstemp for better generalized temp file management. fixes directory assumptions and cleans up temp file creation some 2007-12-14 13:11 brlcad * src/mged/tedit.c, src/mged/utility1.c: more usage of bu_temp_file() instead of mktemp/mkstemp for better generalized temp file management. 2007-12-14 13:21 brlcad * src/util/buffer.c: use bu_temp_file() instead of using mkstemp directly and assuming there's a /usr/tmp 2007-12-14 13:21 brlcad * src/libbu/semaphore.c: use bu_temp_file instead of mkstemp directly, ws on sgi section 2007-12-14 13:31 brlcad * include/config_win.h: no longer need TMP_MAX_S or mktemp with the latest bu_temp_file() implementation 2007-12-14 13:32 brlcad * include/raytrace.h: a_overlap isn't identical to a_multioverlap -- one too many struct partitions 2007-12-14 13:33 brlcad * src/lgt/do_options.c: use bu_temp_file() for the batch file instead of assumping /tmp 2007-12-14 13:45 brlcad * include/raytrace.h: take it back, librt/bool.c does feed in the partition head for a_overlap, so it's the other older callers that are missing that added param 2007-12-14 13:45 brlcad * src/librt/bool.c: typo 2007-12-14 14:43 bob1961 * src/tclscripts/lib/Db.tcl: Fixed the help for ocenter. 2007-12-14 14:44 bob1961 * src/libbu/temp.c: if windows, include <sys/stat.h> 2007-12-14 17:09 brlcad * src/libbu/temp.c: system headers come before our lib headers and should never need to use _WIN32 for header inclusion protection -- use/create the corresponding HAVE_*_H define 2007-12-14 17:46 brlcad * src/libfb/fb_generic.c: can't be consistent 2007-12-14 17:47 brlcad * src/adrt/Makefile.am: one per line when more than one 2007-12-14 20:28 brlcad * configure.ac: try using the -bind_at_load option on Mac OS X so that undefined symbols are all resolved when the library/app is loaded instead of waiting until first use 2007-12-14 20:28 brlcad * src/rt/viewcheck.c, src/rt/viewweight.c: quell warnings, add the missing partition head pointer to the overlap callback 2007-12-14 20:30 brlcad * src/rt/viewweight.c: define the functions in order so that they don't need to be declared 2007-12-14 20:36 brlcad * src/adrt/librender/cut.c: needs another pair of eyeballs to verify, quell warning on second arg to tie_push as a TIE_3 ** 2007-12-14 20:42 brlcad * src/adrt/libutil/image.c: quell warnings, zlib likes its longs unsigned 2007-12-14 20:53 brlcad * src/gtools/g_lint.c, src/gtools/g_qa.c, src/gtools/solshoot.c: quell decl mismatch warnings 2007-12-14 20:58 brlcad * src/irprep/all_sf.c, src/irprep/firpass.c, src/irprep/secpass.c, src/irprep/shapefact.c, src/irprep/showtherm.c: quell warnings 2007-12-14 21:02 brlcad * src/mged/solids_on_ray.c: quell overlap callback decl warning 2007-12-14 21:07 brlcad * src/rttherm/viewtherm.c: quell warning by using NULL -- the logoverlap handler returns void, RT_AFN_NULL returns an int. 2007-12-14 21:13 brlcad * src/util/ttcp.c: give the compilation warnings one more shot at quelling the socklen_t vs int vs size_t inconsistency mix that has persisted over the years. posix defines socklen_t, len was int* during early bsd days. may need to create our own typedef if non-posixy so it can still compile stand-alone 2007-12-14 21:27 bob1961 * src/tclscripts/lib/Command.tcl: Added the reinitialize method. 2007-12-14 21:29 bob1961 * src/tclscripts/archer/LoadArcherLibs.tcl: Removed the dir parameter from both LoadArcherCoreLibs and LoadArcherLibs. 2007-12-14 21:31 bob1961 * src/tclscripts/archer/ArcherCore.tcl: Added a few class variables that were previously in the Archer class. 2007-12-14 21:33 bob1961 * src/tclscripts/archer/Archer.tcl: Added code to change the prompt. Also move a few class variables to ArcherCore. 2007-12-14 21:35 bob1961 * src/archer/archer: Mods related to class variables moving from Archer to ArcherCore. 2007-12-15 05:08 brlcad * include/config_win.h: don't need HAVE_MATHERR 2007-12-15 05:09 brlcad * src/mged/anal.c: remove the old acos() hack to quell implementation errors by overriding matherr() 2007-12-15 05:17 brlcad * src/librt/Makefile.am, src/librt/sqrt.gould.s, src/librt/sqrt.vax.s: remove the unused assembler sqrt() implementations for the gould and vax. relinquish to the bowels of revision history until there is (ever) a need to properly reintegrate into the build (where they probably belong in libbn) 2007-12-15 05:22 brlcad * configure.ac: libtool on mac chokes on the -bind_at_load, get rid of it 2007-12-15 05:37 brlcad * include/config_win.h: no longer need HAVE_TMPFILE_S, no longer using tmpfile_s() 2007-12-15 05:39 brlcad * configure.ac: no longer using tmpfile_s(), don't check for it 2007-12-15 05:42 brlcad * configure.ac: we don't use ftime() 2007-12-15 06:22 brlcad * configure.ac: include a comprehensive list of c89 functions that I'm aware of that should be a near-complete list of functions we should be able to rely on. this does NOT include some of the POSIX.1 functions, but it should be a near-complete list otherwise. functions compiled from the following headers: assert.h ctype.h locale.h math.h setjmp.h signal.h stdarg.h stddef.h stdio.h stdlib.h string.h time.h 2007-12-15 07:53 brlcad * include/config_win.h, src/libbu/malloc.c: consider calloc() a given with c89 2007-12-15 07:55 brlcad * configure.ac: comment on the libsysv calls we check for even though they're c89. don't check calloc() 2007-12-15 08:10 louipc * misc/archlinux/PKGBUILD.static: On second thought this won't really work too well. Removing. 2007-12-15 08:12 louipc * misc/archlinux/Makefile.am: Remove PKGBUILD.static from dist. 2007-12-15 08:44 brlcad * src/burst/prnt.c, src/fbserv/fbserv.c, src/lgt/execshell.c, src/libpkg/pkg.c, src/other/URToolkit/tools/into.c: strerror is c89, and does work on windows at least with vc8 so drop the obsolete and error-prone sys_errlist array alternates and configure check 2007-12-15 09:09 brlcad * src/mged/tedit.c: we have getenv() 2007-12-15 09:17 brlcad * include/config_win.h, src/libbu/linebuf.c, src/util/pixcut.c, src/util/pixpaste.c: take setvbuf for granted per c89, bye bye HAVE_SETVBUF 2007-12-15 09:17 brlcad * configure.ac: don't need to check for setvbuf and strerror funcs any more 2007-12-15 16:52 brlcad * configure.ac, include/config_win.h: don't need to check for getenv(), go c89 2007-12-15 17:20 brlcad * src/libbu/parallel.c: raise() works on windows and is c89, so bye bye 2007-12-15 18:01 brlcad * configure.ac: raise() works on windows and is c89, so bye bye 2007-12-15 18:03 brlcad * configure.ac: strdup and strsep don't seem to be c89 2007-12-15 20:13 brlcad * configure.ac, src/libpkg/pkg.c: get rid of our single use of strerror_r, minimize checks 2007-12-15 20:29 brlcad * configure.ac: remove a whole bunch of function checks that can be taken for granted since c89 provides for them. all of these functions are apparently not even used (any more) via HAVE_ decls regardless. removing the checks reduces configure time signficantly. function checks removed include atexit, fabs, floor, memchr, memmove, modf, pow, setlocale, sqrt, strcpy, strcspn, strpbrk, strrchr, strspn, strstr, strtol, strtoul, strtoull 2007-12-15 20:44 brlcad * configure.ac: comment consistency 2007-12-15 21:09 brlcad * configure.ac: list the POSIX.1 headers too for good measure 2007-12-15 21:17 bharder * doc/book/VolumeIV.xml: Working with new processing to facilitate img handling, re-worked <link> and <xref>, minor formatting. 2007-12-15 21:29 bharder * doc/book/VolumeIV.xml: Previous commit also re-worked the placement of authorship from <title> to <bookinfo> and the dedication from <prolog> to <dedication> -- not sure about the <author> change... will need to keep eye on this. 2007-12-16 15:14 brlcad * configure.ac: no longer check for bzero, begin conversion to memset 2007-12-16 15:31 brlcad * include/machine.h: bzero is POSIX.1, memset is c90 -- pick the latter for now and remove bzero from machine.h 2007-12-16 15:33 brlcad * include/bu.h, include/raytrace.h: ws 2007-12-16 15:34 brlcad * include/nmg.h: begin bzero -> memset conversion 2007-12-16 15:42 brlcad * configure.ac: remove a slew of function checks that have no corresponding HAVE_* define use in the code reducing configure time by another significant amount. functions included are: dup2, endgrent, endpwent, finite, fttruncate, getcwd, gethostbyaddr, gethostbyname, getpagesize, getpass, gettimeofday, inet_ntoa, isascii, isfinite, localtime_r, mkdir, mkfifo, munmap, ml_langinfo, panic, prtcl, putenv, re_comp, realpath, rmdir, select, setlinebuf, socket, strcasecmp, strncasecmp, tzset, uname, utime 2007-12-16 15:59 brlcad * misc/macosx/openUp.c, src/burst/fb.c, src/canon/canonlib.c, ... lots of other files ..., src/util/ttcp.c: convert all bzero calls to memset 2007-12-16 16:02 brlcad * include/machine.h: fully remove the bzero/bcopy section 2007-12-16 16:05 brlcad * src/libbu/bitv.c, src/libbu/malloc.c, src/librt/nmg_plot.c, src/sig/dwin.c, src/tab/tabinterp.c, src/util/pixtile.c: few multiline bzero to memset stragglers 2007-12-16 16:09 brlcad * HACKING: comment on using memset and memcpy instead of bzero and bcopy 2007-12-16 16:48 brlcad * src/libbu/convert.c: not clear whether the two regions can overlap, so convert bcopy to memmove 2007-12-16 16:49 brlcad * src/libbu/htond.c, src/libbu/htonf.c: memcpy has been part of strings.h since c90, get rid of the HAVE_MEMORY_H condition and bcopy alternate 2007-12-16 18:53 brlcad * src/adrt/isst/slave/load.c, src/adrt/isst/slave/slave.c, ... lots of other files ..., src/util/ttcp.c: convert usages of bcopy to memcpy throughout. although none of them should be buffers that overlap per machine.h history, all calls were individually checked regardless. new code should not use bzero/bcopy any more. 2007-12-16 20:13 brlcad * configure.ac: not using HAVE_GETOPT anywhere, no need to check for getopt 2007-12-16 20:13 brlcad * configure.ac: erm, adrt does use HAVE_GETOPT_LONG, though, so check for getopt_long 2007-12-16 20:54 brlcad * sh/trimconfig.sh: initial script that reports what defines are being declared in a configure template via function checks, header checks, and manual defines. it correlates what is being set/checked with what is used in the sources and reports how many uses there are. this is usefull in culling checks/defines that are not being used. 2007-12-16 20:55 brlcad * sh/Makefile.am: provide the trimconfig.sh script since it should be fairly generic to any configure.ac file 2007-12-16 21:03 brlcad * sh/trimconfig.sh: make the SKIP tests actually skip and make the AC_DEFINES checks actually work 2007-12-16 21:09 brlcad * sh/trimconfig.sh: cleanup printing 2007-12-16 21:27 brlcad * src/libbu/malloc.c: memset missing comma 2007-12-16 22:19 brlcad * src/proc-db/masonry.c: oops, subscripted value is neither array nor pointer 2007-12-16 22:23 brlcad * configure.ac: remove checks for a slew of headers that seem to have no corresponding HAVE_*_H in use anywhere in our sources. these include arpa/inet.h, ieeefp.h, itcl.h, itk.h, langinfo.h, malloc.h, netdb.h, socket.h, strings.h, sys/resource.h, sys/statfs.h, sys/timeb.h, utime.h, values.h, waitflags.h, and wchar.h 2007-12-17 01:42 brlcad * include/machine.h: BOOL_T isn't even used any more 2007-12-17 02:13 brlcad * include/raytrace.h: ws 2007-12-17 02:14 brlcad * include/config_win.h: no longer should be using nor needing the bzero/bcopy macros -- should be using memset and memcpy throughout now 2007-12-17 02:16 brlcad * include/bn.h, include/machine.h, src/libbn/complex.c, ... lots of other files ..., src/rttherm/viewtherm.c: removal of the FAST declaration throughout. now using register or letting the compiler sort things out. 2007-12-17 02:22 brlcad * src/libdm/dm-wgl.c: revert int back to BOOL, windows thing 2007-12-17 05:53 brlcad * src/tclscripts/geometree/tclIndex, src/tclscripts/mged/tclIndex, src/tclscripts/rtwizard/lib/tclIndex, src/tclscripts/sdialogs/scripts/tclIndex, src/tclscripts/swidgets/scripts/tclIndex: revert the rest of the bad tclIndex files that were clobbered during the 2007/11/28 commit with log "LOCAL->static, per machine.h deprecation list" 2007-12-17 06:33 brlcad * doc/html/manuals/Install.html: remove references to LOCAL and FAST 2007-12-17 07:53 brlcad * src/libbn/mat.c, src/libbn/plane.c, src/libbn/poly.c, ... lots of other files ..., src/rt/viewrad.c, src/rt/viewscat.c, src/rt/worker.c: my bad deprecation instruction, revert/remove the LOCAL -> static conversion. LOCAL is only static for non-SMP systems, but usually auto. 2007-12-17 07:54 brlcad * include/machine.h: poof, make LOCAL go away entirely. the deprecation instruction was wrong/bad -- LOCAL shouldn't go to static, it just goes away (so we get auto). 2007-12-17 08:17 brlcad * src/librt/g_hf.c: remove silly rcsid printing 2007-12-17 08:29 brlcad * src/lgt/reflect.c: add last partition param 2007-12-17 13:24 d_rossberg * src/libbu/temp.c: time() is declared in time.h 2007-12-17 13:33 d_rossberg * src/conv/comgeom/f2a.c: bu_exit() is declared in bu.h 2007-12-17 13:55 d_rossberg * src/librt/db_tree.c: do not forget to copy a string's terminating 0 2007-12-17 14:48 bob1961 * src/conv/asc2g.c: Fixed a bug related to trying to close a null file pointer. 2007-12-17 15:01 brlcad * NEWS: parker fixed asc2g bug on Windows that caused a crash on exit 2007-12-17 15:09 brlcad * src/librt/db_tree.c: sanity, manually null-terminate after the strncpy in case the buffer is 'full' -- should consider a bu_strncpy or bu_strlcpy that does this automatically 2007-12-17 18:23 bob1961 * src/libdm/dm-ogl.c, src/libdm/dm-wgl.c: Set flag to clear back buffer in drawBegin routine. 2007-12-17 19:17 brlcad * src/other/blt/src/Makefile.am: wrong case on tkMenubutton.c, missing from dist 2007-12-17 19:28 bob1961 * src/tclscripts/archer/Archer.tcl: Move call to itk_initialize up a bit. This fixes the problem of the main canvas' background not getting set. Added a call to writePreferences in the destructor. 2007-12-17 20:00 erikgreenwald * src/librt/db_tree.c: s/shaderlen/shader_len/ 2007-12-17 20:04 erikgreenwald * src/adrt/adrt.h: force enum values instead of "hoping". define the name size limit. 2007-12-17 20:05 erikgreenwald * src/adrt/libcommon/adrt_common.h: moving some stuff to ../adrt.h 2007-12-17 20:10 erikgreenwald * src/adrt/isst/slave/main.c: change naming prefix 2007-12-17 20:41 brlcad * NEWS, src/external/ProEngineer/proe-brl.c: update the pro/engineer exporter so that it not only checks if an object named 'top' already exists, but it will attempt to use various 'top.#' alternative names in order to avoid a clash and still create a top-level object for the user. this should fulfill the second half of sf request [ 1159469 ] "Pro/E converter improvements". this mod is alas untested because I don't have protk at the moment. 2007-12-17 20:58 erikgreenwald * src/adrt/isst/slave/Makefile.am, src/adrt/isst/slave/load.c, src/adrt/isst/slave/load.h, src/adrt/isst/slave/slave.c, src/adrt/isst/slave/slave.h: reorg to reduce prototypes 2007-12-17 21:05 bob1961 * src/tclscripts/archer/ArcherCore.tcl: Move a few methods related to editing from ArcherCore to Archer. 2007-12-17 21:07 bob1961 * src/tclscripts/archer/Archer.tcl: Move a few methods related to editing from ArcherCore to Archer. Added handlers for editing via the mouse. 2007-12-17 21:21 brlcad * NEWS, src/external/ProEngineer/proe-brl.c: update the pro/e plugin so that it also now implements the first half of sf request [ 1159469 ] "Pro/E converter improvements" whereby it now more appropriately parses the part number to part name mapping file and allows part names that have spaces in the name. instead of stopping at the first whitespace, it now reads until the end of the line and extracts the part name, allowing for an arbitrary amount of surrounding whitespace that it trims off. 2007-12-17 21:47 brlcad * TODO: improve the framebuffer support for HDR imagery, an alpha channel for transparency, and configurable timeouts on remove framebuffers. 2007-12-17 23:05 brlcad * src/nirt/usrfmt.h: ws consistency cleanup 2007-12-17 23:05 brlcad * src/nirt/nirt.c: major ws and style consistency cleanup 2007-12-17 23:06 brlcad * src/nirt/dist_def.c, src/nirt/interact.c: use c99 fmax instead of max macro, might need configure support 2007-12-17 23:07 brlcad * src/nirt/command.c, src/nirt/if.c: use c99 fabs() instead of abs macro, might need configure support 2007-12-17 23:08 brlcad * src/nirt/nirt.h: clean ws and style consistency, document origin, remove min/max/abs and DEBUG_FORMAT (raytrace.h provides it), add header wrapper protection 2007-12-17 23:14 brlcad * src/nirt/interact.c: ws and style consistency cleanup 2007-12-17 23:30 brlcad * src/nirt/bsphere.c, src/nirt/command.c, src/nirt/dist_def.c: more ws and style consistency cleanup 2007-12-18 00:11 brlcad * src/nirt/command.c: uffer 2007-12-18 00:13 brlcad * src/nirt/nirt.c: say it's natalie's 2007-12-18 00:14 brlcad * src/nirt/showshot.c: clean up ws and style, clean up bu_exit and error reporting calls, clean up header 2007-12-18 00:14 brlcad * src/nirt/conversion.c, src/nirt/if.c, src/nirt/parse_fmt.c, src/nirt/read_mat.c, src/nirt/str_manip.c: ws and style consistency cleanup 2007-12-18 00:20 brlcad * TODO: investigate rt_shootray() backing out the ray when inside an object and nirt's -b backout option. perhaps it used to not back out the ray automatically? should make sure the change wasn't unintentional. 2007-12-18 06:46 brlcad * src/adrt/isst/master/isst_python.c: eek, fix a few of the bad strncat/strncpy sizes that were put on arg 2 instead of 3. manually null-terminate for safety. 2007-12-18 07:07 brlcad * src/anim/anim_sort.c, src/canon/canonize.c, src/conv/euclid-g.c, src/conv/g-adrt.c, src/conv/iges/showtree.c, src/conv/proe-g.c, src/fb/cell-fb.c, src/libbn/mat.c, src/libbu/parse.c, src/librt/dg_obj.c, src/librt/nirt.c: null-terminate strncat'd buffers for sanity sake since we don't usually check if we filled/truncated 2007-12-18 15:47 brlcad * misc/win32-msvc8/Makefile.am: treeInit.sh no longer exists 2007-12-18 15:51 brlcad * misc/enigma/enigma.c: quell warning, need #define _XOPEN_SOURCE to get crypt from unistd.h on SGI linux 2007-12-18 16:07 brlcad * BUGS: annotate/migrate here the bug from sf bug tracker item [ 925408 ] "mk_lcomb() in libwdb fails on large combinations" .. the bug was too low-priority and developer-centric to have lingering in the sf tracker indefinitely, but worth keeping a note of here. 2007-12-18 17:02 brlcad * src/adrt/Makefile.am: have to include the always_compile files in the DIST_SUBDIRS else they feel left out 2007-12-18 17:11 brlcad * src/libtclcad/tclcad.c: remove carriage returns, add footer, add missing headers and a RESOURCE_INCLUDED define for using a system-installed itcl.h 2007-12-18 17:35 brlcad * src/libtclcad/tclcad.c: quell Tcl_PkgProvide warning 2007-12-18 17:37 brlcad * src/libtclcad/Makefile.am: add tclcad.c to libtclcad. this makes libtclcad loadable by Tcl! 2007-12-18 18:22 erikgreenwald * src/adrt/isst/master/Makefile.am, src/adrt/isst/master/compnet.c, src/adrt/isst/master/dispatcher.c, src/adrt/isst/master/dispatcher.h, src/adrt/isst/master/isst_python.c, src/adrt/isst/master/isst_python.h, src/adrt/isst/master/main.c, src/adrt/isst/master/master.c, src/adrt/isst/master/master.h: merge in the newer stuff 2007-12-18 18:32 erikgreenwald * src/adrt/isst/isst.h: remove unnecessary include 2007-12-18 18:33 erikgreenwald * src/adrt/isst/slave/slave.c: ISST_OP_SHOT is now ADRT_WORK_SHOTLINE 2007-12-18 19:09 erikgreenwald * src/anim/anim_sort.c: add missing arg... 2007-12-18 19:39 bob1961 * src/mged/chgview.c, src/mged/cmd.c, src/mged/cmd.h: Added the following two commands: ae2dir and viewdir. 2007-12-18 19:41 bob1961 * include/vmath.h: Added V3AE2DIR. 2007-12-18 19:43 bob1961 * include/raytrace.h: Declare vo_viewDir_cmd() and vo_ae2dir_cmd(). 2007-12-18 19:44 bob1961 * src/librt/view_obj.c: Added vo_ae2dir_cmd and vo_ae2dir_tcl. 2007-12-18 19:49 erikgreenwald * src/adrt/isst/master/Makefile.am: remove unnecessasry stuff 2007-12-18 20:00 erikgreenwald * src/adrt/isst/slave/Makefile.am: remove unnecessary stuff 2007-12-18 21:06 bob1961 * src/tclscripts/lib/QuadDisplay.tcl, src/tclscripts/lib/View.tcl: Added ae2dir. 2007-12-18 21:09 bob1961 * src/tclscripts/helplib.tcl: Added help string for ae2dir. 2007-12-18 21:33 erikgreenwald * configure.ac: libtclcad needs blt linked 2007-12-18 21:36 erikgreenwald * src/adrt/isst/observer/Makefile.am: don't build this. Ever. Yet... 2007-12-18 21:39 erikgreenwald * src/adrt/isst/master/compnet.c, src/adrt/isst/master/compnet.h, src/adrt/isst/master/dispatcher.c, src/adrt/isst/master/dispatcher.h, src/adrt/isst/master/main.c, src/adrt/isst/master/master.c, src/adrt/isst/master/master.h: generify function names a bit (hopefully leading to unification of the various compute chunks) 2007-12-18 21:42 erikgreenwald * src/adrt/Makefile.am: welcome to the "always build" club, libcommon and isst. 2007-12-18 22:05 erikgreenwald * src/adrt/libtie/tie_define.h: change to use vmath macros 2007-12-18 22:06 erikgreenwald * src/adrt/libutil/umath.h: remove redundant macros 2007-12-19 15:39 bob1961 * src/libtclcad/tclcad.c: This hack is required for compilation on windows, atleast until I figure out what's really going on. 2007-12-19 16:24 brlcad * src/other/tcl/Makefile.am: fix the distcheck, the upgrade incremented the version numbers of some of the .tm files, so change INSTALL_ARCHIVES to be *any* version 2007-12-19 16:28 brlcad * src/anim/anim_sort.c: reserve was already decreased, need to add space remaining/expected 2007-12-19 20:13 erikgreenwald * src/adrt/isst/slave/Makefile.am: force mysql off by default to allow the build to pass 2007-12-19 20:23 erikgreenwald * src/adrt/isst/slave/Makefile.am: turn mysql off harder... 2007-12-19 21:41 brlcad * NEWS: fixed empty name crash when using mged 'lt' command. this should fix a bug encountered in some crash logs that Craig Barker had and some mged crashes that he reported with 7.10.4. the problem was an empty name being passed to lt which ended up getting passed to bu_log as a null ... argument, which subequently caused the fwrite to fail and bu_bomb to kick off. 2007-12-19 21:42 brlcad * src/librt/db_lookup.c: fixed empty name crash when using mged 'lt' command. this should fix a bug encountered in some crash logs that Craig Barker had and some mged crashes that he reported with 7.10.4. the problem was an empty name being passed to lt which ended up getting passed to db_lookup and later bu_log as a null ... argument, which subequently caused the fwrite to fail and bu_bomb to kick off. 2007-12-19 21:44 brlcad * src/libbu/log.c: ws 2007-12-19 22:13 brlcad * src/libbu/vls.c: return gracefully instead of bombing if there is nothing to print into, or nothing to print in bu_vls_vprintf 2007-12-19 22:27 brlcad * src/libbu/vls.c: gah, I'm making a habit of this lately or apparently too much shell scripting, haven't made these typos in years 2007-12-19 22:29 brlcad * src/libbu/log.c: try stdout just in case something is wrong with stderr when logging before fatally aborting the app. there's probably something wrong either way, but until we can set/redirect bu_log -- try harder to not fail. 2007-12-19 22:32 brlcad * src/librt/nmg_ck.c, src/librt/track.c, src/mged/dir.c, src/mged/rtif.c, src/mged/track.c, src/proc-db/nmgmodel.c, src/proc-db/pipe.c, src/proc-db/torii.c, src/rt/rtscale.c, src/tab/scriptsort.c: manuallly null-terminate buffers after a strncat for sanity 2007-12-19 22:41 brlcad * src/libfb/if_remote.c: ws and style cleanup on rem_open, init vars 2007-12-19 22:46 brlcad * TODO: note that ::key=val;key2=val2; might work well for setting a timeout on remote framebuffers (e.g. -F::host=brlcad.org;port=3;timeout=40;) 2007-12-19 22:59 brlcad * src/libbu/linebuf.c: avoid passing a null FILE to setvbuf 2007-12-19 23:02 brlcad * src/libbu/log.c: do the same protections to bu_putchar that bu_log had to try stdout if stderr fails or is unavailable. instead of blindly trying stdout/stderr, make sure they're not NULL FILE pointers first and fail accordingly. 2007-12-19 23:41 brlcad * src/other/tcl/unix/Makefile.in, src/other/tk/unix/Makefile.in: apply the gentoo ebuild tcl/tk build system patch that conditionally installs the manpages (someone should try to push this upstream) 2007-12-20 08:47 brlcad * include/vmath.h: provide DEG2RAD and RAD2DEG constants for pi/180 and 180/pi; rename the new V3AE2DIR to V3DIR_FROM_AZEL to be more consistent with other vmath macros, similarly making first arg the dir instead of last; implement the complementary AZEL_FROM_V3DIR for setting an az/el from a direction vector as well (untested) 2007-12-20 08:48 brlcad * src/librt/view_obj.c: update from V3AE2DIR to newly renamed V3DIR_FROM_AZEL macro name 2007-12-20 08:54 brlcad * src/libbn/const.c: use the vmath.h/math.h macros where available (math.h are generally higher precision), convert all constants to fastf_t values. 2007-12-20 20:17 bob1961 * misc/win32-msvc8/bw2fb, misc/win32-msvc8/bw2fb/bw2fb.vcproj, ... lots of other files ..., misc/win32-msvc8/spm2fb, misc/win32-msvc8/spm2fb/spm2fb.vcproj: Initial check-in. 2007-12-20 20:21 bob1961 * misc/win32-msvc8/brlcad/brlcad.sln: Added 38 more projects. Mostly fb utils and their missing libs. 2007-12-20 20:26 bob1961 * include/config_win.h: Define HAVE_RAND and fmax. 2007-12-20 20:28 bob1961 * include/orle.h: Added ORLE_EXPORT for exposing variables and functions on windows. 2007-12-20 20:30 bob1961 * include/vfont-if.h: Mods to expose vfont_get() and vfont_free() on windows. 2007-12-20 20:34 bob1961 * src/other/libutahrle/include/rle.h: Mods to expose variables and functions. 2007-12-20 20:36 bob1961 * src/fb/fbanim.c, src/fb/fbcmrot.c, src/fb/fbfade.c, src/fb/fbstretch.c, src/fb/gif-fb.c, src/fb/ioutil.c, src/fb/pixflip-fb.c, src/fb/pl-fb.c, src/other/libutahrle/include/rle_config.h, src/other/libutahrle/rle_getrow.c, src/other/libutahrle/rle_open_f.c, src/other/libutahrle/scanargs.c: Mods to get things compiling on windows. 2007-12-20 22:56 bob1961 * misc/win32-msvc8/brlcad/brlcad.sln, ... lots of other files ..., misc/win32-msvc8/wish/wish.vcproj: Clean up the project files a little. 2007-12-22 15:01 brlcad * src/adrt/bench/main.c, src/adrt/isst/master/main.c, src/adrt/isst/observer/main.c, src/adrt/isst/slave/main.c: probably going to revert to a libbu solution, but manually null-terminate for now anyways 2007-12-22 18:26 brlcad * misc/win32-msvc8/Makefile.am: add newly added project files to the dist for: bw2fb cat2fb cell2fb cmap2fb fb2bw fb2cmap fb2fb fb2orle fb2rle fbanim fbcbars fbcmap fbcmrot fbcolor fbfade fbframe fbfree fbgamma fbgammamod fbgrid fbhelp fblabel fbline fbpoint fbscanplot fbstretch fbzoom gif2fb liborle libutahrle orle2fb pixautosize pixflip2fb pl2fb polar2fb pp2fb rle2fb spm2fb 2007-12-22 19:31 brlcad * misc/flawfinder: the message was wrong for scanf() warnings about the %s modifier when all we know in the default case is that it's not a constant. 2007-12-22 19:36 brlcad * regress/flawfinder.sh: ignore false positives, finished reviewing all of level four issues too but keep it at five for regression testing 2007-12-22 19:37 brlcad * src/adrt/isst/slave/load.c: quell flawfinder warnings, check buffer lengths when printing into them 2007-12-22 19:38 brlcad * src/mged/cad_boundp.c: get rid of Mess, just did silly print-header-to-stderr+print-body-to-stdout that was mostly used for debugging. just fprintf the output for now (consider bu_log later). 2007-12-22 19:39 brlcad * src/nirt/parse_fmt.c: -> ws 2007-12-22 19:39 brlcad * src/proc-db/fence.c, src/proc-db/fence.h, src/proc-db/pipe.c: const params 2007-12-22 19:40 brlcad * src/util/loop.c: check length while printing into buffers 2007-12-22 19:40 brlcad * src/vas4/init_ik.c: don't need popen decl. it's in stdio.h 2007-12-22 19:45 brlcad * configure.ac, include/config_win.h, src/fb/fbfade.c: rand() is c89/c90 so don't need/want to check for it. check for random() instead. 2007-12-22 19:51 brlcad * src/fb/pl-fb.c: right idea to simplify the loop, but take it a little further. 2007-12-22 20:03 brlcad * NEWS: bob added a new 'ae2dir' command to mged per request from dave loman where the command will report the direction vector for a given azimuth and elevation. there is an invert option to flip the vector around. 2007-12-22 20:18 brlcad * src/librt/view_obj.c: add the dir2ae pairing to the new ae2dir command. 2007-12-22 20:19 brlcad * include/raytrace.h: declare new vo_dir2ae_cmd view obj command for dir2ae 2007-12-22 20:20 brlcad * src/mged/cmd.c: header cleanup 2007-12-22 20:23 brlcad * src/mged/chgview.c, src/mged/cmd.c, src/mged/cmd.h: hook in the new dir2ae command 2007-12-22 20:25 brlcad * NEWS: just like ae2dir, added new dir2ae for calculating an az/el from a given direction vector. 2007-12-22 20:52 brlcad * TODO: query_ray seems to be blathering warnings/errors 2007-12-31 17:29 brlcad * TODO: tested parallel crash report generation. it's still busted. for some reason, the BU_MAPPED_FILE semaphore is getting acquired multiple times simultaneously. 2007-12-31 17:30 brlcad * include/config_win.h: move USE_PROTOTYPES up with the __STDC__ define since they're closely related 2007-12-31 17:34 brlcad * include/db.h, include/machine.h, include/nmg.h: c89 is assumed, remove the \!__STDC__ protections/sections 2007-12-31 17:34 brlcad * include/raytrace.h: remove the \!RECORD_DEFINED sections.. struct record (v4) should always be definable if we can assume v4. better to reduce the header file conditionals regardless 2007-12-31 17:35 brlcad * src/burst/Hm.c, src/burst/Hm.h, src/burst/HmGetc.c: remove the __STDC__ protections, c89 is a given 2007-12-31 17:36 brlcad * src/conv/g-nff.c: use the new bu_temp_file() instead of tmpfile() 2007-12-31 17:38 brlcad * src/conv/nastran-g.c: file pointer, not file descriptor. there's a difference. 2007-12-31 17:39 brlcad * src/conv/iges/g-iges.c: use new bu_temp_file() instead of tmpfile() for temporary files 2007-12-31 17:39 brlcad * src/fbed/std.h: remove the if not __STDC__ section 2007-12-31 17:40 brlcad * src/libbn/fortran.c: really odd wrapper.. the uppercase variant seems to never get used/declared contrary to the comment. either way, remove the non __STDC__ section 2007-12-31 17:41 brlcad * src/libbn/wavelet.c: remove the __STDC__ protections, it's a given 2007-12-31 17:44 brlcad * src/libbu/bomb.c: fix at least one race-condition problem with parallel crash report generation. acquire a semaphore when printing into the tracefile filename and acquire the BU_SEM_SYSCALL semaphore when printing so we don't interleave with other printing that may be going on. do the same when printing the intentional core dump message. 2007-12-31 17:45 brlcad * src/libbu/brlcad_path.c: consistently lowercase 2007-12-31 17:46 brlcad * src/libbu/crashreport.c: make a note that bu_crashreport() is NOT thread-safe. it could have an internal semaphore wrapper, but for now, just let callers protect (i.e. leave it up to bu_bomb()) 2007-12-31 17:47 brlcad * src/libbu/list.c: no longer need assert.h 2007-12-31 17:53 brlcad * src/libbu/malloc.c, src/libbu/vls.c: move bu_strdup_message out of malloc.c and into vls.c, add bu_vls_strcmp() and bu_vls_strncmp() as new functions instead of macros so the sanity checks can be added. 2007-12-31 17:54 brlcad * include/bu.h: declare bu_vls_strcmp() and bu_vls_strncmp() as functions instead of macros, remove the plethora of __STDC__ protections on the assertion macros and friends. 2007-12-31 17:55 brlcad * src/librt/nmg_fcut.c, src/librt/table.c, src/librt/tree.c, src/mged/dodraw.c: remove the non __STDC__ sections, c89 is a given now 2007-12-31 17:56 brlcad * src/vdeck/vdeck.c: comment and header cleanup, convert k&r prototypes to ansi, doxygenify 2007-12-31 18:28 brlcad * BUGS: the new linger-by-default option doesn't seem to be working at least with 7.10.4 on ia32 linux, reported by Paul58 2007-12-31 19:14 brlcad * src/libbu/str.c: add initial implementation of bu_strlcat() and bu_strlcpy() that provide safe wrappers around the string functions of similar name. if they're not available, then the strn family are used and appropriate null-termination behavior is matched from the strl family. warnings are printed if we exceed the specified max size. similar to bu_strdup(), the interface is wrapped in macros so that we can report file:line numbers when there is a problem. additionally, move bu_strdupm into here from malloc.c 2007-12-31 19:15 brlcad * src/libbu/malloc.c: moved bu_strdupm() to new src/libbu/str.c file 2007-12-31 19:16 brlcad * include/bu.h: include stddef.h for size_t and declare the new bu_strlcatm() and bu_strlcpym() functions and their associated bu_strlcat/bu_strlcpy macro wrappers. 2007-12-31 19:16 brlcad * src/libbu/Makefile.am: add str.c to the compilation 2007-12-31 19:18 brlcad * misc/win32-msvc7/libbu/libbu.vcproj, misc/win32-msvc8/libbu/libbu.vcproj, src/libbu/libbu.dsp: add the new src/libbu/str.c file to the windows build 2007-12-31 20:25 brlcad * doc/plot3.5, include/brep.h, include/db.h, include/fbmsg.h, ... lots of other files ..., src/external/Unigraphics/ug-g.1: header cleanup, more address removal 2008-01-11 18:54 brlcad * autogen.sh, bench/lgt.sh, bench/run.sh, bench/try.sh, ... lots of other files ..., src/rttherm/test.sh, src/tclscripts/nirt/prototype.sh: set the executable property on all shell scripts that cvs2svn failed to set even through CVS repository and auto-props both said they should have it set 2008-01-11 18:58 brlcad * include/conf/make.vbs, pix/cube.rt, src/adrt/misc/rename.pl, src/adrt/scripts/adrt.py, src/adrt/scripts/drawBSP.py, src/adrt/scripts/makemovie.pl, src/adrt/scripts/ptcloud.py, src/archer/archer.bat, src/conv/intaval-g.py, src/mged/mged.bat, src/other/iwidgets/doc/tk2html.perl, src/other/libpng/contrib/pngminus/png2pnm.bat, src/other/libpng/contrib/pngminus/pngminus.bat, src/other/libpng/contrib/pngminus/pnm2png.bat, src/other/libz/contrib/masmx64/bld_ml64.bat, src/other/libz/contrib/masmx86/bld_ml32.bat, src/other/libz/contrib/masmx86/mkasm.bat, src/other/tcl/win/buildall.vc.bat, src/other/tk/win/buildall.vc.bat, src/other/tk/win/mkd.bat, src/other/tk/win/rmd.bat: more executable scripts that need their property set 2008-01-11 19:00 brlcad * NEWS: converted the source repository from CVS to Subversion 2008-01-11 19:34 brlcad * COPYING, Makefile.am, autogen.sh, bench/benchmark.1, ... lots of other files ..., src/vdeck/vextern.h, src/vdeck/vproc.c: And I, for one, welcome our new rat overlords. All hail 2008, the year of the rat! (copyright update) 2008-01-13 18:54 brlcad * misc/macosx/openUp.c: quell ohloh APSL license detection since we can't yet specify (to ohloh) that it should ignore/exempt some files/dirs 2008-01-13 23:08 brlcad * src/mged/mged_dm.h: no need to define DEG2RAD and RAD2DEG 2008-01-13 23:17 brlcad * configure.ac: enable maintainer mode for any type of checkout, not just cvs (i.e. subversion too) 2008-01-13 23:24 brlcad * doc/book/VolumeIV.xml: use bu_exit for the examples 2008-01-13 23:26 brlcad * src/adrt/isst/master, src/adrt/isst/slave: ignore the binary products 2008-01-13 23:39 brlcad * src/conv/iges/n_iges.cpp, src/conv/iges/nmain.cpp, src/gtools/beset/beset.c, src/halftone/main.c, src/irprep/Makefile.am, src/irprep/irdisp.c, src/libbu/association.c, src/libbu/backtrace.c, src/libbu/fopen_uniq.c, src/libbu/htester.c, src/libbu/image.c, src/libbu/rb_internals.h, src/libbu/semaphore.c, src/libfb/if_ug.c, src/librt/g_ebm.c, src/librt/prep.c, src/sig/Makefile.am, src/sig/a-d.c, src/sig/bw-d.c, src/sig/d-bw.c, src/sig/d-f.c, src/sig/d-i.c, src/sig/d-u.c, src/sig/d2-c.c, src/sig/dconv.c, src/sig/ddisp.c, src/sig/dfft.c, src/sig/dmod.c, src/sig/dsel.c, src/sig/dstat.c, src/sig/f-d.c, src/sig/f-i.c, src/sig/i-d.c, src/sig/i-f.c, src/sig/imod.c, src/sig/istat.c, src/sig/smod.c, src/sig/u-d.c, src/sig/u-f.c, src/sig/umod.c, src/sig/ustat.c: refactor to use bu_exit instead of bu_log+exit where applicable 2008-01-14 07:42 brlcad * src/util/Makefile.am, src/util/pix-ci.c: remove the obsolete (irix-specific) pix-ci tool for preparing pix files for irix image processing tools 2008-01-14 07:47 brlcad * src/util/Makefile.am, src/util/pl-starbase.1, src/util/pl-starbase.c: also remove pl-starbase as it requires maintenance, custom headers, and hasn't been enabled (compilation-wise) in many years 2008-01-14 13:58 bob1961 * include/config_win.h: No longer need HAVE_RANDOM. 2008-01-14 14:36 brlcad * src/mged/hideline.c: remove the UAH and redstone arsenal source address/phone number, cleanup header 2008-01-14 14:36 brlcad * src/util/pl-hpgl.1: remove source 2008-01-14 14:51 brlcad * src/util/ttcp.c: return instead of exit in main 2008-01-14 14:56 brlcad * include/rtgeom.h: rtgeom uses/requires vmath and bu for typedef decls, so include them 2008-01-14 15:06 brlcad * src/adrt/bench/main.c, src/adrt/isst/slave/main.c, ... lots of other files ..., src/util/rle-pix.c, src/util/sgi-pix.c: extensive bu_exit refactoring with related cleanup on calls to exit and stderr, header cleanup along with way 2008-01-14 15:09 brlcad * .cvsignore, bench/.cvsignore, db/.cvsignore, doc/.cvsignore, ... lots of other files ..., src/vas4/.cvsignore, src/vdeck/.cvsignore: remove the .cvsignore files since they are no longer used. the ignore patterns are now svn:ignore properties on the directory and can be edited with propset/propedit. 2008-01-14 15:14 brlcad * src/util/Makefile.am, src/util/pix-both-ik.c, src/util/pix-filt-ik.c, src/util/pix-ik.c, src/util/pix-ikr.c, src/util/pix-lowp-ik.c, src/util/pixt-ik.c: remove the various Ikonas writer utilities that have been disabled from compilation (now that they're all updated, ugh) since they require an unspecified header and need maintenance updates. until someone needs them again, they can stay stashed in the bowels of revision history. includes pix-filt-ik, pix-ik, pix-ikr, pix-lowp-ik, pixt-ik. 2008-01-14 15:20 brlcad * src/util/Makefile.am: pix-ci.c was removed too 2008-01-14 15:28 brlcad * src/util/Makefile.am, src/util/fix_polysolids.c: update fix_polysolids to the new API specs, massive reformat and cleanup, make it compile again. still needs more work so that it's not v4 specific and not just polysolids but any nmg/bot. 2008-01-14 15:29 brlcad * src/util: ignore fix_polysolids 2008-01-14 15:34 brlcad * sh/tracker.sh: fix a bug where the page layout changed and the tracker type/name was no longer getting extracted correctly, also now extract the submitters description as part of a new COMMENT field in the output. take care to not save more than 1020 chars for now. 2008-01-14 15:38 brlcad * sh/Makefile.am, sh/news2tracker.sh: add a new script for generating the same style table as tracker.sh, but extracted from the NEWS file revision history. was written to work with svn commands, pulling the same data as it pertains to sf trackers using the commit history. cross references the annotate revisions per-line with the commit log messages for those lines, and uses the NEWS file's simple format as an assumption for extracting the title and assignees. 2008-01-14 15:48 brlcad * NEWS: new tracker tool for NEWS items 2008-01-14 16:58 brlcad * src/burst/Hm.c, src/burst/HmGlob.c, src/burst/burst.c, ... lots of other files ..., src/vdeck/vproc.c: delintify the ifndef linted RCSid $ variable blocks, i.e., remove them. they're not really useful any more, particularly with svn (or any modern scm for that matter), so we can make the compiler's job just a little bit easier. 2008-01-14 17:08 brlcad * include/conf.h, include/plot3.h, include/sysv.h, ... lots of other files ..., src/util/xyz-pl.c, src/vdeck/cgarbs.c: use consistent spacing around the common header inclusion using one and only blank line before and after. 2008-01-14 17:10 brlcad * src/adrt/bench/main.c, src/adrt/isst/master/main.c, src/adrt/isst/observer/main.c, src/adrt/libtie/tie.c, src/adrt/rise/observer/main.c: more common.h consistency 2008-01-14 17:46 erikgreenwald * src/libbu/htester.c: RCSid is actually used here 2008-01-14 17:50 erikgreenwald * src/libpkg/pkg.c, src/librtserver/rtserver.c: RCSid is actually used here 2008-01-14 18:04 bob1961 * misc/win32-msvc8/asc2g/asc2g.vcproj, ... lots of other files ..., misc/win32-msvc8/wish/wish.vcproj: Minor cleanup. 2008-01-14 18:59 brlcad * src/libpkg/pkg.c: for debug printing the version, use the (newish) pkg_version() instead 2008-01-14 19:00 brlcad * src/librtserver/rtserver.c: similarly, have the service be provided the full version string (if the intent is to parse a date/time/version, there are other/better routines as rt_version() is the human-readable paragraph form, so might need changing) 2008-01-14 19:02 brlcad * src/libbu/htester.c: meh, it's a noinst, just kill the (fairly useless) -v option. 2008-01-14 19:40 erikgreenwald * src/librt/g_metaball.c: Initial support for handling different evaluation methods. 2008-01-14 19:51 erikgreenwald * include/rtgeom.h: Prototype for point evaluation function. 2008-01-14 19:55 bob1961 * src/librt/dg_obj.c: Mods affecting use of rt and rtcheck on Windows. 2008-01-14 19:59 bob1961 * src/nirt/parse_fmt.c: uos not getting terminated properly. 2008-01-14 20:03 bob1961 * src/libdm/dm-wgl.c: Tk_CreateWindowFromPath not properly initializing the TkWindow structure for toplevel windows. 2008-01-14 20:08 erikgreenwald * include/rtgeom.h, src/librt/g_metaball.c: Add goo factor to metaball primitive (used in metaball and blob evaluation) 2008-01-14 20:14 bob1961 * src/libfb/if_wgl.c: Removed call to Sleep() from wgl_do_event(). This was causing a delay in wgl_close(). Uncomment/reactivate a few lines of code in wgl_open(). 2008-01-14 20:17 bob1961 * src/mged/setup.c: Initialize dgop->dgo_rtCmdNotify. This function is called by librt when rt and nirt finish. 2008-01-14 20:18 bob1961 * src/mged/ged.c: Mods to get classic mode working on Windows. 2008-01-14 20:21 bob1961 * include/raytrace.h: Added dgo_rtCmndNotify member. 2008-01-14 20:57 bob1961 * src/libpkg/pkg.c: Minor mods for Windows. 2008-01-14 20:59 bob1961 * src/rt/main.c: Mods for Windows (initialize sockets if using a framebuffer). 2008-01-14 21:03 erikgreenwald * src/adrt/isst/Makefile.am, src/adrt/isst/isst.h, src/adrt/isst/isst_struct.h, src/adrt/isst/slave/main.c, src/adrt/isst/slave/slave.c: remove vestigial headers 2008-01-14 21:03 bob1961 * src/fb/fbcmap.c, src/fb/fbfade.c, src/fb/fbgamma.c, src/fb/fbgammamod.c, src/fb/fblabel.c, src/fb/fbline.c, src/fb/fbscanplot.c: Mods to initialize sockets. This is really for Windows (i.e. calls to pkg_init() do nothing unless on Windows). 2008-01-14 21:10 erikgreenwald * src/adrt/rise/master/Makefile.am, src/adrt/rise/slave/Makefile.am: forcibly disable RISE for now 2008-01-14 21:22 erikgreenwald * src/adrt/slave, src/adrt/slave/main.c, src/adrt/slave/slave.c: hoist the isst master/slave nodes up to adrt's toplevel 2008-01-14 21:23 erikgreenwald * src/adrt/master, src/adrt/master/Makefile.am, src/adrt/master/compnet.c, src/adrt/master/compnet.h, src/adrt/master/dispatcher.c, src/adrt/master/dispatcher.h, src/adrt/master/main.c, src/adrt/master/master.c, src/adrt/master/master.h: hoist the isst master/slave nodes up to adrt's toplevel 2008-01-14 21:23 erikgreenwald * configure.ac: hoist the isst master/slave nodes up to adrt's toplevel 2008-01-14 21:24 erikgreenwald * src/adrt/isst/Makefile.am, src/adrt/isst/main.c, src/adrt/isst/master, src/adrt/isst/observer.c, src/adrt/isst/observer.h, src/adrt/isst/observer/Makefile.am, src/adrt/isst/observer/main.c, src/adrt/isst/observer/observer.c, src/adrt/isst/observer/observer.h, src/adrt/isst/observer/splash.h, src/adrt/isst/slave, src/adrt/isst/splash.h: hoist the isst master/slave nodes up to adrt's toplevel 2008-01-14 21:24 erikgreenwald * src/adrt/Makefile.am, src/adrt/doc/ChangeLog: hoist the isst master/slave nodes up to adrt's toplevel 2008-01-14 21:42 bob1961 * src/mged/ged.c: Fix pre-processor mishap. 2008-01-14 22:24 brlcad * HACKING: should use bu_strlcat/bu_strlcpy instad of strn?cat/strn?cpy 2008-01-14 23:11 brlcad * include/anim.h, include/bn.h, include/bu.h, include/db.h, ... lots of other files ..., src/vdeck/vdeck.c, src/vdeck/vproc.c: comma space ws consistency, include a space after commas in expressions (e.g., ', '). applied globally minus a few exceptions where the alignment was signficant to the formatting or on tables where it was used to indicate grouping. 2008-01-15 00:22 brlcad * bench/pixcmp.c, include/raytrace.h, include/vmath.h, ... lots of other files ..., src/util/pcd-pix.c, src/util/pixsubst.c, src/util/pl-dm.c: more ws comma consistency cleanup. remove the spaces before commas throughout. 2008-01-15 06:04 brlcad * NEWS: Bob fixed the extended rendering delays that users would see when ray-tracing on Windows. you could see via the log that the ray-trace would finish very quickly (nearly instantly sometimes) but then would stall for a minute or two before displaying the image to the framebuffer. the problem was actually a manual Sleep() call that was added a long time ago (heh) to the wgl framebuffer that made it interactively update the image while it rendered and made a completed image not starve out other processes with massive CPU use. now there's still problems interactively updating and it may starve others, but the delay is gone (and it was worse). 2008-01-15 06:22 brlcad * NEWS: (reworded line) Bob fixed the extended rendering delays that users would see when ray-tracing on Windows. you could see via the log that the ray-trace would finish very quickly (nearly instantly sometimes) but then would stall for a minute or two before displaying the image to the framebuffer. the problem was actually a manual Sleep() call that was added a long time ago (heh) to the wgl framebuffer that made it interactively update the image while it rendered and made a completed image not starve out other processes with massive CPU use. now there's still problems interactively updating and it may starve others, but the delay is gone (and it was worse). 2008-01-15 14:01 bob1961 * src/mged/ged.c: It looks like the Windows specific call to Tcl_CreateChannelHandler() was accidentally removed. 2008-01-15 16:25 bob1961 * src/mged/attach.c: The indices for the MGED_DISPLAY_VAR were accidentally changed. This prevented MGED's status bar on the command window from being updated. 2008-01-15 19:39 bob1961 * src/tclscripts/mged/accel.tcl: Fixed bug in rcc-cap that shows up when using the 't' option. 2008-01-16 06:29 brlcad * HACKING: new example footer using c-file-style so that the style is encouraged more strictly. setting the c-file-style effectly causes a c-set-style to get invoked for that buffer, with the predefined 'stroustrup' style most closely matching our convention (it's basically 'k&r' with 4-char indents), this fixes the braindead default 'gnu' style emacs uses that indents the c-offsets-alist substatement-open style parameter a level (i.e., open brace curlies are indented a level ahead of the proceeding statement). 2008-01-16 06:35 brlcad * sh/footer.sh: add a c-file-style to the C files and remove their corresponding c-basic-offset lines. can't override the c-basic-offset if using a style, not that we want/need to with stroustrup since it matches our offset 2008-01-16 06:38 brlcad * TODO: include the list of remaining mged tasks for this release and the next, including the questions and comments mentioned at the last arl/brl-cad configuration control board meeting. immediate tasks include fixing mged i/o on puts/gets; text flooding; disabling mged auto-backgrounding; and fixing tclConfig.sh problems. next are verifying/testing pix-fb -p, clone mirroring, dbconcat with a prefix, and pushing primitives. longer term is windows smp support, new tree traversal code, and exposing nmg_fix_normals through a userland command. 2008-01-16 06:40 brlcad * HACKING: change the ordering so the c-file-style is last 2008-01-16 07:06 brlcad * bench/pixcmp.c, include/RtServerImpl.h, include/anim.h, ... lots of other files ..., src/vdeck/vextern.h, src/vdeck/vproc.c: set a c-file-style globally in the local variables footer, setting it to the built-in 'stroustrup' style. that style is basically the 'k&r' style with 4-char indents (otherwise emacs' 'k&r' style has 5-char indents), which matches our convention. you can't specify c-basic-offset when you used c-file-style apparently (at least as of emacs 21.2.1) but conveniently the 'stroustrup' style matches so we don't have to care. 2008-01-16 14:29 brlcad * src/liborle/rle.c: cleanup and restructure, apply style, remove decls 2008-01-16 14:36 brlcad * configure.ac: only look for the tclConfig.sh and tkConfig.sh files if we're not building tcl or tk and need to load the config, otherwise having TCL_BIN_DIR set isn't too useful (the --with/--without flags are still available). this should fix the configure exit being encountered on some systems when a system tcl/tk isn't already installed (i.e., the WARNING: Can't find Tcl configuration definitions 'error') 2008-01-16 18:26 brlcad * src/adrt/isst/observer: you have to delete the directories too when they're empty with svn since they're tracked entities (you can svn rm an entire subdir as one-operation) 2008-01-16 18:26 brlcad * configure.ac: remove the src/adrt/isst/observer/Makefile entry since the dir is empty/gone 2008-01-16 21:03 brlcad * src/proc-db/Makefile.am, src/proc-db/sgi.sh: add an example of creating geometry using basic shell scripting. the example shows several ways for passing modeling commands to mged, creates the .g geometry file, sets material properties, and renders the model to an image. don't install it, just an example. 2008-01-16 21:05 bob1961 * src/mged/ged.c: Hack to get "puts" working on Windows. Still need to look at this on the other platforms. 2008-01-16 21:07 bob1961 * src/mged/dm-X.c, src/mged/dm-ogl.c, src/mged/dm-tk.c, src/mged/dm-wgl.c: if_name not getting properly terminated (i.e. not eos) 2008-01-16 21:18 brlcad * src/proc-db/sgi.sh: keep a png 2008-01-16 21:39 erikgreenwald * src/librt/g_metaball.c: Improve text display routines to provide more informative render method information as well as "good" when needed. Add functions to translate between a numeric method ID and a human readable string. Minor format cleanup. 2008-01-16 21:46 erikgreenwald * src/librt/g_metaball.c: Wrapper for multiple normal methods added. Isopotential normals fixed with the help of Ed. 2008-01-16 21:57 brlcad * src/proc-db/sgi.sh: match the cube rendering more closely lining up the center better, adding a little bit of perspective, and using a tube radius that better approximates. keep the png as sgi.png.. 2008-01-17 00:57 brlcad * src/proc-db/sgi.sh: add the same comments that were used for the wiki page so that it better explains all steps being taken 2008-01-17 02:34 johnranderson * src/gtools/g_diff.c: fixed two bugs. One was incorrect use of strcmp. Second was not counting some differences found. Fixes bug #1865987 2008-01-17 02:50 louipc * misc/archlinux/PKGBUILD.in: Install script actually shouldn't be in sources. No need to specify source tarball since the build scripts are included in the dist. 2008-01-17 12:48 bob1961 * src/mged/ged.c: Fixed a misplaced endif. 2008-01-17 16:00 brlcad * src/librt/g_brep.cpp: annotate that jason wrote this 2008-01-17 16:26 brlcad * src/librt/opennurbs_ext.cpp: add header and footer 2008-01-17 19:01 brlcad * src/libwdb/brep_stub.c: add trailing newline 2008-01-17 19:02 brlcad * configure.ac: turn brep stuff back on 2008-01-17 19:34 brlcad * src/proc-db/brep_simple.cpp: clarify the maybe comment .. and get rid of the conditional brep_simple.g creation -- always create the .g file 2008-01-17 19:55 brlcad * src/proc-db/brep_simple.cpp: make sure we could make the twisted cube before calling mk_brep 2008-01-17 23:17 bob1961 * src/libbu/temp.c: Closing a file descriptor after an fclose breaks on windows. 2008-01-17 23:21 bob1961 * src/mged/tedit.c: Make the default editor for windows wordpad. When writing out the primitives on windows use crlf. Lastly, on windows change the return value to 1 in editit(). 2008-01-18 00:19 johnranderson * src/mged/tedit.c: ted on an arb now displays the expected vertices 2008-01-18 00:31 johnranderson * src/librt/g_arb.c: added some minimal comments to rt_arb_get_cgtype 2008-01-18 00:49 johnranderson * src/mged/tedit.c: oops, sorry Bob, I stomped on one of your changes 2008-01-18 04:30 brlcad * HACKING: CVS is no more, update the docs for SVN 2008-01-18 06:09 brlcad * NEWS: Bob implemented support for MGED classic mode on Windows, adding a hack that goes through tcl i/o routines (i.e. doesn't use the same read/write calls as elsewhere) 2008-01-18 06:18 brlcad * NEWS: Bob fixed a bug in mged's rcc-cap command where using the 't' option didn't work due to curlie braces that wrapped a tcl var going to vadd2. removed the curlies and it now works. 2008-01-18 06:29 brlcad * NEWS: erik improved the metaball rendering with the addition of a blobbyness/gooeyness factor that can be set. also added improved text display routines. 2008-01-18 06:29 brlcad * NEWS: erik and ed fixed a bug in the calculation of normals for metaballs 2008-01-18 06:35 brlcad * NEWS: John fixed a bug in g_diff where differences between objects were getting missed due to the command not keeping track of the count of differences that had accummulated up to that point. there was also an incorrect strcmp call that had the wrong truth value getting tested. this fixes a bug reported by David Loman ([ 1865987 ] g_diff fails to see BoT Orientation change) 2008-01-18 06:45 brlcad * NEWS: bob improved the mged ted command on windows, making the files being written out actually use windows line terminators so they don't wrap around as one line. he also made wordpad the default editor instead of notepad (user can still override) -- this responds to requests for functional text edit support on windows from several bnd modelers 2008-01-18 06:51 brlcad * NEWS: John Anderson improved the way the ted command in mged deals with ARBs that are less than ARB8s (e.g. an ARB6) .. instead of duplicating the points and requiring users to edit the duplications, only present them with the # points for their ARB#. this fixes a bug report submitted by karel/clock3 where there was unexpected behavior that the edits where intenional and that the user had to edit the duplicates ([ 1842596 ] ted on arbn6 doesn't change vertices) 2008-01-18 19:45 brlcad * src/libwdb/brep.cpp: need a copy of the users brep here because wdb_export will free the brep when it's done with it. mk_brep doesn't need to claim the brep, so we can let callers keep using that brep with mods for future mk_brep() calls (with a different name) 2008-01-18 19:47 brlcad * src/proc-db/brep_cube.cpp, src/proc-db/brep_simple.cpp: the two lines aren't needed, and the edge orientations do seem to make sense so remove the XXX ???'s. since mk_brep changed, we need to delete the brep ourselves (which is good because we had to new it in the first place) 2008-01-18 19:49 brlcad * src/librt/g_brep.cpp: provide an initial implementation of the describe command that uses the ON Dump() call for describing arbitrary breps consistently. ... so nice to be able to use mid-scope decls.. *sigh*. also remove the m_object_table count debug statement. 2008-01-18 19:51 brlcad * src/proc-db/Makefile.am, src/proc-db/breplicator.cpp: add a new breplicator proc-db test app for creating a variety of brep geometry. not yet complete but it will make a brep with some points in it, saving out to a .g file. next step, create a variety of arbs. 2008-01-18 20:00 brlcad * src/librt/g_nurb.c: call it a NURBS solid 2008-01-18 20:01 brlcad * src/librt/g_brep.cpp: give the initial describe line a more informative label of BREP object. 2008-01-18 23:25 brlcad * src/librt/g_brep.cpp: remove the first line (ON_Brep:) of the description 2008-01-18 23:27 brlcad * src/proc-db/breplicator.cpp: create vertices, curves, and edges for the test cube 2008-01-18 23:28 brlcad * src/proc-db/breplicator.cpp: use SMALL_FASTF instead of VDIVIDE_TOL 2008-01-18 23:43 brlcad * INSTALL, NEWS, README, doc/description.txt, doc/html/manuals/Obtain.html, misc/win32-msvc8/tclsh/library/installTree.tcl, sh/enumerate.sh, sh/indent.sh, sh/template.sh, sh/ws.sh: more cleanup, referencing, and checking for subversion dirs where it used to check for cvs 2008-01-18 23:43 brlcad * src/other/openNURBS/opennurbs_brep.h: fix typo, s/trin/trim/ 2008-01-18 23:44 brlcad * Makefile.am: dist-hook is probably busted for finding files not added to the dist, need to update for svn so leave a FIXME comment for now. 2008-01-19 20:57 brlcad * include/bu.h: change the IND_ routines to END_ for endian, even thou twas a funny joke 2008-01-19 21:05 brlcad * src/libbu/convert.c: change indian to endian 2008-01-19 23:53 brlcad * HACKING: comment that bu_exit is intended to replace most printing statements that are merely followed by an exit() 2008-01-19 23:53 brlcad * src/conv/g-acad.c, src/conv/g-dxf.c, src/conv/g-euclid.c, src/conv/g-euclid1.c, src/conv/g-nff.c, src/conv/g-nmg.c, src/conv/g-obj.c, src/conv/g-stl.c, src/conv/g-tankill.c, src/conv/g-vrml.c, src/conv/g-x3d.c, src/conv/g-xxx_facets.c, src/conv/iges/g-iges.c, src/conv/jack/g-jack.c, src/conv/off/g-off.c, src/libbn/noise.c, src/libbn/plane.c, src/libbn/tabdata.c, src/libbu/malloc.c, src/libbu/parse.c, src/libbu/rb_delete.c, src/libbu/rb_extreme.c, src/libbu/rb_walk.c, src/librt/pr.c, src/librt/shoot.c, src/mged/animedit.c: bu_bomb cleanup 2008-01-20 00:12 brlcad * TODO: bu_bomb's in the front ends seem to be all properly converted to bu_exit 2008-01-20 00:16 brlcad * TODO: tclConfig.sh problems should be gone now that it doesn't try to locate those files until it thinks we're not building tcl 2008-01-20 17:09 brlcad * bench/pixcmp.c, include/anim.h, include/bn.h, include/bu.h, ... lots of other files ..., src/vdeck/parsarg.c, src/vdeck/vdeck.c, src/vdeck/vproc.c: more ws consistency cleanup with spaces after commas and K&R style consistency distinguishing language constructs (for|while|if|switch) from functions with a space. 2008-01-22 05:11 brlcad * configure.ac: check for strlcat/strlcpy for the guts of bu_strlcat/bu_strlcpy 2008-01-22 05:13 brlcad * include/db.h: include bu.h, self-contain 2008-01-22 05:13 brlcad * src/rttherm/viewtherm.c: header doc cleanup 2008-01-22 05:15 brlcad * src/canon/canon.h: need bu.h 2008-01-22 05:16 brlcad * src/conv/g5-g4.c: avoid the strncpy but keep it obvious that we're replacing 'plastic' with 'phong ' 2008-01-22 05:17 brlcad * src/conv/comgeom/3d.h: remove duplicate defines already provided by db.h for NAMESIZE/NAMEMOVE, cleanup header 2008-01-22 05:22 brlcad * src/conv/iges/check_names.c, src/conv/iges/do_subfigs.c, src/conv/iges/g-iges.c, src/conv/iges/iges.c, src/conv/iges/iges_struct.h, src/conv/iges/main.c, src/conv/iges/revolve.c, src/conv/iges/showtree.c: consistently treat NAMESIZE as C strings of sizeof NAMESIZE+1 for the null (so the max names are 16) .. only db.h uses buffers instead of C-strings (for backwards compat with struct sizes). use new bu_strlcpy to properly copy and null-terminate safely. 2008-01-22 05:23 brlcad * src/conv/dbupgrade.c: replace the strncpy with explicit setting when replacing 'plastic' with 'phong ' 2008-01-22 05:24 brlcad * src/mged/track.c: consistently treat NAMESIZE as C strings of sizeof NAMESIZE+1 for the null (so the max names are 16) .. only db.h uses buffers instead of C-strings (for backwards compat with struct sizes). use new bu_strlcpy to properly copy and null-terminate safely. 2008-01-22 05:24 brlcad * src/mged/objdir.h: only set NAMESIZE if we need to 2008-01-22 05:27 brlcad * src/mged/concat.c: consistently treat NAMESIZE as C strings of sizeof NAMESIZE+1 for the null (so the max names are 16) .. only db.h uses buffers instead of C-strings (for backwards compat with struct sizes). use new bu_strlcpy to properly copy and null-terminate safely. 2008-01-22 05:28 brlcad * src/burst/burst.h: only define NSIG if we need to 2008-01-22 05:28 brlcad * src/burst/Mm.h: use libbu memory management and string processing 2008-01-22 06:09 brlcad * src/adrt/libcommon/canim.c: ws 2008-01-22 06:17 brlcad * src/conv/patch/patch-g.h: consistently use NAMESIZE+1 for non-db.h struct strings (intentionally treating as strings instead of buffers) 2008-01-22 06:18 brlcad * src/conv/g4-g5.c: avoid the strncpy but keep it obvious that we're replacing 'plastic' with 'phong ' 2008-01-22 06:48 brlcad * src/vdeck/vproc.c: wow, still k&r .. update to ansi, use libbu more (untested) 2008-01-22 06:51 brlcad * src/vdeck/vdeck.c: tricky/risky/untested string changes to use bu_strlcpy, but do it anyways since they seem safe enough (given use and surrounding code) and there are no strcat's involved. 2008-01-22 06:59 brlcad * src/librt/nmg_class.c: meh 2008-01-22 07:14 brlcad * src/util/ttcp.c: meh 2008-01-22 07:25 brlcad * include/raytrace.h, src/adrt/bench/main.c, ... lots of other files ..., src/util/remapid.c: sweeping string management changes to help ensure more secure/consistent null termination and bounds checking. use bu_strlcpy/bu_strlcat where appropriate passing the size of the buffer (expected to be a NULL-terminated C-string) instead of calling strcat/strncat/strcpy/strncpy. this lets us get rid of most of the sanity/manual null-terminations. also update to use sizeof instead of hard-coded constants where useful. hopefully not too many/any bugs getting introduced though the entire commit was painstakingly reviewed at least three times over and passed all manual testing as well as the regress tests. 2008-01-22 07:36 brlcad * src/mged/rtif.c: oops, typo -- line not count 2008-01-22 07:39 brlcad * src/proc-db/fence.c: quell warnings 2008-01-22 07:49 brlcad * include/orle.h, src/liborle/rle.c: decouple liborle from libfb (and in turn from libbu) by having it provide its own RLEpixel and RLEColorMap types as well as defining RED/GRN/BLU if needed. this should fix compilation woes being seen on some platforms (e.g. debian) under certain configurations -- particular an error related to not finding tcl.h (from bu.h from fb.h). 2008-01-22 07:53 brlcad * src/adrt/librender/Makefile.am: need tcl flags (bu.h inclusion) 2008-01-22 07:54 brlcad * src/adrt/libutil/Makefile.am: need tcl flags, opl 2008-01-22 07:55 brlcad * src/adrt/slave/Makefile.am: need tcl flags 2008-01-22 07:59 brlcad * include/orle.h: need COPYRGB defined too 2008-01-22 08:00 brlcad * src/fbed/Makefile.am: include the tcl cppflags via fb.h 2008-01-22 08:04 brlcad * src/libbu/brlcad_path.c, src/libbu/convert.c, src/libbu/image.c, src/libbu/lex.c, src/libbu/units.c, src/libbu/whereis.c, src/libbu/which.c: more conversion over to the new bu_strlcpy/bu_strlcat routines for more consistent null-termination and bounds checking. (vls and parse are still not modified as some of their callers seem to be buffers instead of C-strings) 2008-01-22 08:13 brlcad * include/bu.h, src/libbu/fchmod.c: refactor the bu_fchmod interface to only take a FILE * instead of the filename as that has the same security vulnerability as calling chmod() does. remove the chmod() call in the implementation as that was just done for Windows where fchmod() or a secure equivalent doesn't seem to be available. instead, punt by doing what cygwin does. just ignore the request and pretend it failed. might want to pretend it succeeded, but good as anything for now. 2008-01-22 08:19 brlcad * src/rt/do.c: need string.h for the ansi funcs (for memset), quell warning 2008-01-22 08:25 brlcad * src/rt/reshoot.c: need string.h for the ansi funcs (for memset), quell warning 2008-01-22 08:27 brlcad * src/fb/fb-pix.c, src/fb/fb-png.c, src/mged/rtif.c: bah, second mistake almost in a row... almost completely forgot to update the callers to bu_fchmod to remove the filename parameter (!). time to call it a night before I start making other stupid mistakes at this late hour. 2008-01-22 14:04 bob1961 * src/libdm/dm-wgl.c: When creating a toplevel window make sure it's not iconified. 2008-01-22 14:21 bob1961 * src/mged/utility1.c: Close FILE * before calling editit. 2008-01-22 14:24 bob1961 * src/mged/mater.c: Close FILE * before calling editit. 2008-01-22 14:24 bob1961 * src/mged/chgmodel.c: Close FILE * before calling editit. 2008-01-22 14:28 bob1961 * src/mged/red.c: Check return value from bu_temp_file(). Close FILE * before calling editit. 2008-01-22 14:36 bob1961 * src/mged/tedit.c: Minor formatting. 2008-01-22 18:30 erikgreenwald * src/adrt/libutil/display.c: Need to include bu.h for the bu_strlcpy() macro. 2008-01-22 22:08 erikgreenwald * include/rtgeom.h, include/wdb.h, src/librt/g_metaball.c, src/mged/chgmodel.c, src/mged/edsol.c: make goo value per metaball control point 2008-01-22 22:20 erikgreenwald * src/vas4/vas_io.c: Use termios tab expansion name instaed of sgtty if possible (fbsd pr ports/119096). 2008-01-23 03:39 brlcad * src/other/tcl/unix, src/other/tk/unix: ignore the stub libs 2008-01-23 03:40 brlcad * src/proc-db: ignore breplicator 2008-01-23 04:21 brlcad * src/libbu/parse.c: one more converted over to bu_strlcpy/bu_strlcat. this one was intentionally skipped initially as there's a strncpy going on that intentionally is short-copying a buffer. didn't investigate the cause, whether intentional or not, but leaving alone with a note for now. 2008-01-23 04:34 brlcad * src/mged/tedit.c: if your gonna tweak ws, might as well do the whole file 2008-01-23 04:37 brlcad * NEWS: bob got all of the mged text editing commands (edcodes, edmater, wmater, ted, etc) on the windows platform 2008-01-23 04:40 brlcad * NEWS: Bob also got rtwizard working on windows (at least partially -- he still has to add a few missing commands for some of the extended modes). also made the display managers behave better by automatically deiconifying if needed. 2008-01-23 06:30 brlcad * bench/pixcmp.c, include/dm.h, include/fb.h, include/raytrace.h, ... lots of other files ..., src/vas4/vas_io.c: even MORE K&R ws style consistency cleanup. this time, make sure there's a space between the ){'s as well as around "else" statements. 2008-01-23 07:06 brlcad * include/pkg.h, src/fb/fb-pix.c, src/fb/fb-png.c, src/fb/fbclear.c, src/fb/fbcmap.c, src/fb/fbfade.c, src/fb/fbgamma.c, src/fb/fbgammamod.c, src/fb/fblabel.c, src/fb/fbline.c, src/fb/fbscanplot.c, src/fb/pix-fb.c, src/fb/png-fb.c, src/fbserv/fbserv.c, src/libpkg/pkg.c, src/mged/ged.c, src/rt/main.c: refactor libpkg so that the windows-specific pkg_init and pkg_terminate functions should no longer be necessary. according to the msdn docs, the functions may be called multiple times as they internally reference count the number of calls to each and deallocated on the last/matching WSACleanup that corresponds to the number of calls to WSAStartup. the calls were moved directly into pkg_open and pkg_close (which will get picked up by all callers including fb_open) and removed from all callers. this should fix all of the libpkg-using network apps like all of the framebuffer utilities that still need to be ported as well as more consistently now automatically ties in the cleanup call to the pkg connection closure. hopefully this doesn't cause bob problems (i.e. it's untested.. ;) 2008-01-23 07:33 brlcad * src/burst/Sc.h: quell warnings and remove non-ansi decls by removing the oddly named _STDC_ sections 2008-01-23 07:34 brlcad * src/other/tk/unix: ends up with a libtclstub8.5.a.E in here anyways, so add to ignore list 2008-01-23 07:38 brlcad * src/rt: ignore librttherm.a and libremrt.a 2008-01-23 18:20 bob1961 * src/libpkg/pkg.c: If we're going to perform a WSACleanup() in pkg_close() we also need to call WSAStartup() in pkg_getclient() and _pkg_permserver_impl(). 2008-01-23 18:23 bob1961 * src/tclscripts/rtwizard/lib/FbPage.itk: Mods to get things working on Windows. 2008-01-23 18:25 bob1961 * src/util/bw-pix.c: Tweak for compiling on Windows. 2008-01-23 18:29 bob1961 * src/archer/archer.bat, src/mged/mged.bat: Minor tweaks. 2008-01-23 18:32 bob1961 * include/raytrace.h, include/rtgeom.h: Moved a few metaball function declarations from rtgeom.h to raytrace.h. 2008-01-23 18:38 brlcad * doc/deprecation.txt, include/bu.h, src/libbu/brlcad_path.c: make bu_brlcad_root/bu_brlcad_data both return a const char * instead of a char * so callers are more aware that they should copy the string if they want to modify it 2008-01-23 19:30 bob1961 * src/libfb/if_disk.c: Added a call to setmode() in dsk_open(). This fixes the problem where bad pix files were getting created. 2008-01-23 20:45 brlcad * NEWS: Bob fixed PIX image file creation bug on Windows .. this fixes a problem with many of the pix and framebuffer tools that render directly to a pix file via libfb. the problem was the file was not opened in 'binary mode', which is required on windows or it will perform EOL translations (wreaking all sorts of havoc of course with a pix image) 2008-01-23 20:59 erikgreenwald * include/raytrace.h: include a forward decl for struct defined in rtgeom.h 2008-01-23 21:07 brlcad * src/lgt/extern.h, src/lgt/glob.c, src/lgt/lgt.c: er, not clear why lgt request the data segment address .. don't do it so we can remove the sbrk check 2008-01-23 21:25 brlcad * src/remrt/rtsrv.c: remove the unused beginptr and associated sbrk() calls 2008-01-23 21:52 brlcad * include/config_win.h, src/libbu/malloc.c: move the Windows-specific empty sbrk() hack into config_win.h as just being NULL 2008-01-23 23:13 brlcad * doc/deprecation.txt, include/wdb.h, src/libwdb/Makefile.am, src/libwdb/libwdb.3, src/libwdb/poly.c: finally remove the mk_poly/mk_polysolid/mk_fpoly routines for procedurally creating polysolid primitives. the BoT primitive replaced them a long while ago and have been marked obsolete more than long enough 2008-01-23 23:16 bob1961 * src/tclscripts/rtwizard/lib/PictureTypeE.itcl, src/tclscripts/rtwizard/lib/PictureTypeF.itcl: Mods for running on Windows. Also cleaned up the catch statements. 2008-01-23 23:38 brlcad * src/mged/rtif.c: quell warnings, bu_brlcad_* return const now 2008-01-23 23:42 brlcad * src/proc-db/picket_fence.c, src/proc-db/pix2g.c, src/proc-db/spltest.c: quell warnings 2008-01-23 23:45 brlcad * doc/deprecation.txt, include/wdb.h, src/conv/iges/extrude.c, src/conv/nastran-g.c, src/conv/patch/patch-g.c, src/conv/tankill-g.c, src/librt/nmg_misc.c, src/libwdb/nmg.c, src/util/fix_polysolids.c: finally remove the obsolete write_shell_as_polysolid that was replaced by mk_bot_from_nmg during the 6.0 release 2008-01-23 23:46 brlcad * src/proc-db/contours.c, src/proc-db/pix2g.c: quell warnings 2008-01-24 01:10 brlcad * src/libbu/bomb.c, src/libbu/mappedfile.c, src/libbu/vls.c, src/librt/db_io.c, src/librt/db_open.c, src/rt/do.c, src/rt/view_bot_faces.c, src/tab/script-tab.c: obliterate the old HAVE_UNIX_IO check. instead use the corresponding types/stat/whatever checks that the HAVE_UNIX_IO section protects, or rewrite where appropriate to use c89 routines instead of unix io 2008-01-24 01:11 brlcad * configure.ac: no longer need the HAVE_UNIX_IO and TIME_WITH_SYS_TIME checks. the first was removed throughout the code, the latter is already provided via AC_HEADER_TIME check. 2008-01-24 15:23 erikgreenwald * src/librt/g_metaball.c: Fixed i/o. Started adding "Blinn Blob" evaluator 2008-01-25 06:37 brlcad * TODO: annotate that the ogl framebuffer seems to be recently (last few months) horribly busted on at least a couple platforms (e.g. mac os x) 2008-01-25 06:41 brlcad * TODO: pix-fb -p tested and works as documented, not calling the fb_close and exiting until after the sleep pause 2008-01-25 07:01 brlcad * src/libbu/parse.c: revert the bu_strlcpy/bu_strlcat in here, there are still string truncations going on somewhere 2008-01-25 07:04 brlcad * TODO: classic mged seems to be busted now too.. doesn't read mged_prompt() .. need to debug/check string processing 2008-01-25 07:08 brlcad * TODO: pushing a primitive seems to work just fine. no problems encountered. (this was done in response to recent crash bug with xpush mentioned during a CCB meeting, wanted to test just to be sure push was okay) 2008-01-25 07:36 brlcad * src/mged/cmd.c: quell bu_brlcad_data constness warning and fix dbconcat error reporting 'concat' instead of the command actually invoked 2008-01-25 21:11 erikgreenwald * src/librt/g_metaball.c: fake normal for blobs 2008-01-25 22:37 erikgreenwald * src/other/tcl/ChangeLog, src/other/tcl/ChangeLog.1999, ... lots of other files ..., src/other/tcl/win/tclsh.rc: upgrade from tcl8.5b1 to 8.5.0 release 2008-01-25 22:54 brlcad * src/proc-db/breplicator.cpp: and the test case cube is now working.. apparently something isn't being specified correctly if the right face curves (which were all intentionally specified in reverse order) even after accounting for the edge reversal when making the edge trimming loops. still have to figure out why that wouldn't work, but everything does seem to make sense and even renders correctly when the curves have the intended direction. 2008-01-27 07:51 brlcad * NEWS: erik upgraded the bundled tcl/tk from 8.5a1 to 8.5 final now that the tcl/tk folks have finally released the blessed update 2008-01-27 23:54 brlcad * BUGS: john anderson already/recently fixed the dbconcat of empty combinations bug, just tested and verified here that it's indeed working well now 2008-01-28 00:05 brlcad * NEWS, src/librt/wdb_obj.c, src/mged/cmd.c, src/tclscripts/helplib.tcl: improve and clean up dbconcat's behavior, now allowing users to specify either a prefix, a suffix, no affix, or an auto-incrementing suffix/prefix when importing. if there is a name conflict on import, the user is notified that they didn't get what they asked for. this mod was annotated on sf bug 1816012 from david loman (dbconcat segfaults) that jra already fixed. 2008-01-28 00:12 brlcad * BUGS: the -g/-G aspect rt options were fixed last release, now checking for the appropriate setting 2008-01-28 00:13 brlcad * BUGS: john fixed the units command, was related to 8.5 upgrade not liking callers directly accessing interp->result 2008-01-28 00:35 brlcad * BUGS, NEWS, src/librt/db_scan.c: mged now no longer dies in an assertion failure when the geometry file is bogus. it properly reports that things don't seem right and continues gracefully. this handles the simple test case of 'touch file.g && mged -c file.g units m' that was bombing out. now it detects/reports that it's a bogus geometry file and marks the file read-only. 2008-01-28 00:40 brlcad * BUGS: 'lt object' no longer crashes mged. presumably related to my earlier fix that made mged not crash when given lt on empty objects. 2008-01-28 00:47 brlcad * BUGS: calling xpush on objects with empty trees now works, probably either related to the xpush of primitives fix that erik made or the dbconcat processing of empty combinations fix that john made. either way, tested and now seems to work just fine (it drops the matrix since there's nothing to push it onto). 2008-01-28 06:37 brlcad * src/libbu/avs.c, src/libbu/backtrace.c, src/libbu/brlcad_path.c, src/libbu/crashreport.c: quell various additional compilation warnings 2008-01-28 06:40 brlcad * include/brlcad_version.h: quell compilation warnings about not using brlcad_ident() by making it recursive on a condition that should never be true. 2008-01-28 08:18 brlcad * include/bu.h: quell warnings, use unsigned long magic checking since that's how they are treated 2008-01-28 08:19 brlcad * src/libbu/fchmod.c: include headers for fchmod prototype 2008-01-28 08:20 brlcad * src/libbu/bu_tcl.c: quell warnings, use NEAR_ZERO and match new unsigned long prototype 2008-01-28 08:21 brlcad * src/libbu/badmagic.c: quell extra warnings, use unsigned long for magic checks 2008-01-28 08:22 brlcad * src/libbu/crashreport.c: eek, fix an = vs == bug 2008-01-28 08:22 brlcad * src/libbu/color.c: quell extra warnings, don't test floating point values directly. use NEAR_ZERO 2008-01-28 08:24 brlcad * src/libbu/hash.c: quell extra warnings 2008-01-28 14:02 brlcad * include/common.h: how about this? define O_TEMPORARY and O_BINARY from common since they're generic and would otherwise just be included from an AH_BOTTOM or an AC_DEFINE if put into the config header 2008-01-28 14:02 brlcad * src/libbu/temp.c: no longer need the O_TEMPORARY block in here if common.h provides for it 2008-01-28 14:18 brlcad * src/libbu/image.c, src/libbu/lex.c, src/libbu/list.c: quell extra compilation warnings, be explicit on type conversions 2008-01-28 14:39 d_rossberg * include/config_win.h, src/libbu/mappedfile.c: open a binary database in binary mode (O_BINARY is only in MS Windows different from 0(?)) 2008-01-28 21:51 bob1961 * src/libbu/mappedfile.c, src/libbu/temp.c: Mods to get things working on windows with the latest common.h mods. 2008-01-28 21:54 bob1961 * src/util/bw-pix.c, src/util/bwmod.c, src/util/pix-bw.c, src/util/pixmatte.c: Mods to get things working on windows with the latest common.h mods. 2008-01-28 21:56 bob1961 * src/rt/do.c, src/rt/viewedge.c: Open up the command options on windows. 2008-01-28 21:57 bob1961 * src/fb/fb-pix.c: Mods to get things working on windows. 2008-01-28 21:58 bob1961 * src/other/incrTcl/itcl/generic/itclInt.h, src/other/tcl/generic/tclInt.h: Mods to get things working on windows. 2008-01-28 22:14 bob1961 * misc/win32-msvc8/bw2pix, misc/win32-msvc8/bw2pix/bw2pix.vcproj, misc/win32-msvc8/bwmod, misc/win32-msvc8/bwmod/bwmod.vcproj, misc/win32-msvc8/pix2bw, misc/win32-msvc8/pix2bw/pix2bw.vcproj, misc/win32-msvc8/pixmatte, misc/win32-msvc8/pixmatte/pixmatte.vcproj: Initial check-in. 2008-01-28 22:20 erikgreenwald * misc/win32-msvc8/Makefile.am: Add new files to dist 2008-01-28 22:35 bob1961 * src/tclscripts/rtwizard/lib/PictureTypeF.itcl: Fixed a few of my own sloppy typos. 2008-01-28 22:46 bob1961 * misc/win32-msvc8/tclsh/library/installTree.tcl: Mods to copy rtwizard.bat to installDir. 2008-01-28 22:47 bob1961 * misc/win32-msvc8/libtcl/libtcl.vcproj: Updates related to latest Tcl. 2008-01-28 22:52 bob1961 * misc/win32-msvc8/brlcad/brlcad.sln: Added bw2pix, bwmod, pix2bw and pixmatte. 2008-01-29 07:21 brlcad * src/adrt/libtie/tie_define.h: eep, wth is DOUBLE_PRECISION? quell warning and make it TIE_DOUBLE_PRECISION. 2008-01-29 07:59 brlcad * AUTHORS, src/adrt/bench/bench.c, src/adrt/bench/bench.h, src/adrt/libcommon/cdb.c, src/adrt/libcommon/cdb.h, src/adrt/libcommon/pack.c, src/adrt/libcommon/pack.h, src/adrt/libcommon/unpack.c, src/adrt/libtie/tie_define.h, src/adrt/libtienet/tienet_master.c, src/adrt/libtienet/tienet_master.h, src/adrt/libtienet/tienet_util.c, src/adrt/libtienet/tienet_util.h, src/adrt/libutil/display.c, src/adrt/libutil/image.c, src/adrt/libutil/image.h, src/burst/Hm.c, src/burst/Hm.h, src/burst/prnt.c, src/fbed/char.c, src/fbed/extern.h, src/fbed/getfont.c, src/fbed/getinput.c, src/fbed/prnt.c, src/fbed/try.c, src/lgt/getfont.c, src/lgt/prnt.c, src/nirt/nirt.c, src/nirt/nirt.h, src/nirt/parse_fmt.c, src/remrt/ihost.c, src/remrt/ihost.h: accept markus elfring's sf patch [ 1864112 ] "Improve const-correctness" where data items that are const are so annotated. while there are undoubtedly hundred (if not thousands) of such cases across the codebase, he provided a patch for several dozen that were in adrt, fbed, lgt, nirt, burst, and remrt. patch was half manually applied as too much had already changed since the 7.10.4 source release it was based off of. credit him with thanks for finally making a proper patch instead of vague notional 'questions'. 2008-01-29 15:13 erikgreenwald * src/adrt/bench/bench.c, src/adrt/bench/bench.h, src/adrt/libtienet/tienet_util.c, src/adrt/libtienet/tienet_util.h: fix const-incorrectness (shouldn't mark things const when they're changed) 2008-01-29 15:13 erikgreenwald * src/adrt/libutil/display.h: missing const 2008-01-29 15:19 erikgreenwald * src/lgt/prnt.c: minor casting fixes to quell warnings... 2008-01-29 15:19 erikgreenwald * src/fbed/fbed.c: need the length for snprintf... 2008-01-29 15:22 erikgreenwald * src/nirt/parse_fmt.c: cast to quell warnings 2008-01-29 15:23 erikgreenwald * src/vdeck/vdeck.c: quell cast size mismatch warning on 64b builds 2008-01-29 15:30 bob1961 * src/util/rtwizard.bat: Initial check-in. 2008-01-29 17:04 erikgreenwald * src/adrt/libutil/display.c: prototype fix for building w/o sdl 2008-01-29 17:42 erikgreenwald * configure.ac, src/adrt/adrt.h, src/adrt/libutil/Makefile.am, src/adrt/libutil/display.c, src/adrt/libutil/display.h: remove SDL and python shtuff. 2008-01-29 17:47 erikgreenwald * src/adrt/scripts/shiva.sh: trivial script depends on hosts, other unpublished scripts and old binary names... let's just can it. 2008-01-29 17:48 erikgreenwald * src/adrt/scripts/Makefile.am: trivial script depends on hosts, other unpublished scripts and old binary names... let's just can it. 2008-01-29 17:49 erikgreenwald * src/adrt/Makefile.am: add scripts/ to subdirs 2008-01-29 18:23 erikgreenwald * src/other/tk/ChangeLog, src/other/tk/README, ... lots of other files ..., src/other/tk/win/winMain.c, src/other/tk/xlib/xgc.c: upgrade to tk 8.5.0 release 2008-01-29 19:27 bob1961 * src/tclscripts/rtwizard/lib/MGEDpage.itk, src/tclscripts/rtwizard/lib/PictureTypeE.itcl, src/tclscripts/rtwizard/lib/PictureTypeF.itcl: Mods to fix getting/setting/using ghost intensity. 2008-01-29 19:32 bob1961 * src/tclscripts/rtwizard/rtwizard.tcl: Mods to activate the image and help menus. Also handling WM_DELETE_WINDOW protocol so bwish is not left hanging around. 2008-01-29 19:35 bob1961 * src/tclscripts/rtwizard/lib/Wizard.itk: Initially the Image and Help menus are disabled. Added activeMenu method. 2008-01-29 19:38 bob1961 * src/tclscripts/rtwizard/lib/DbPage.itk: The OK button is now disabled until a file is selected. Added the activateOKButton method. 2008-01-29 19:40 bob1961 * src/util/rtwizard.bat: Mods to handle an argument. 2008-01-29 19:50 erikgreenwald * src/libbu/Makefile.am, src/librt/Makefile.am, src/proc-db/Makefile.am: improved dependancy handling 2008-01-29 19:50 erikgreenwald * configure.ac: set openNURBS to auto 2008-01-29 19:51 erikgreenwald * misc/Makefile.defs: echo dependancy name before recursing 2008-01-29 19:52 erikgreenwald * include/wdb.h: minor re-arrangement of metaball point in anticipation of line controls 2008-01-29 20:03 bob1961 * src/tclscripts/rtwizard/lib/DbPage.itk: Minor improvements to the activateOKButton method. 2008-01-29 22:56 brlcad * src/tclscripts/archer/Archer.tcl: needs to ignore the .svn dir too 2008-01-30 14:14 erikgreenwald * src/other/tk/doc/ttk_dialog.n, src/other/tk/generic/tkInitScript.h, src/other/tk/generic/tkStubImg.c, src/other/tk/library/demos/ttk_demo.tcl, src/other/tk/library/demos/ttk_iconlib.tcl, src/other/tk/library/demos/ttk_repeater.tcl, src/other/tk/library/ttk/dialog.tcl, src/other/tk/library/ttk/icons.tcl, src/other/tk/library/ttk/keynav.tcl, src/other/tk/tests/ttk/misc.test: these files don't exist in tk8.5.0 2008-01-30 15:06 erikgreenwald * src/adrt/slave/load.c, src/adrt/slave/load.h, src/adrt/slave/slave.c: begin abstracting the load path 2008-01-30 17:12 erikgreenwald * src/adrt/librender/component.c: use the adrt.h mesh defines instead of libcommon/adrt_common.h 2008-01-30 17:12 erikgreenwald * src/adrt/libcommon/Makefile.am: don't build libcommon 2008-01-30 17:28 erikgreenwald * src/adrt/slave/Makefile.am: don't use libcommon 2008-01-30 17:30 brlcad * src/rt/rtarea.1: separate names with a newline 2008-01-30 17:33 brlcad * src/other/tk/unix/configure: remove generated configure 2008-01-30 17:54 erikgreenwald * src/adrt/librender/camera.c, src/adrt/librender/cut.c, src/adrt/librender/flos.c, src/adrt/librender/path.c, src/adrt/librender/render_util.c, src/adrt/librender/spall.c, src/adrt/librender/surfel.c, src/adrt/libtexture/texture_blend.c, src/adrt/libtexture/texture_bump.c, src/adrt/libtexture/texture_camo.c, src/adrt/libtexture/texture_checker.c, src/adrt/libtexture/texture_clouds.c, src/adrt/libtexture/texture_gradient.c, src/adrt/libtexture/texture_image.c, src/adrt/libtexture/texture_mix.c, src/adrt/libtexture/texture_perlin.c, src/adrt/libtexture/texture_stack.c, src/adrt/libtie/Makefile.am, src/adrt/libtie/tie.c, src/adrt/libtie/tie_kdtree.c, src/adrt/libtienet/Makefile.am, src/adrt/libtienet/tienet_master.c, src/adrt/libtienet/tienet_util.h, src/adrt/libutil/camera.c, src/adrt/libutil/image.c, src/adrt/libutil/umath.c, src/adrt/master/master.c, src/adrt/slave/load.c: use libbu's memory management routines 2008-01-30 18:00 erikgreenwald * src/adrt/libutil/image.c: allow the save_ppm func to build (used in isst) 2008-01-30 18:43 erikgreenwald * src/adrt/libtie/tie_kdtree.c: fixed a couple random crash bugs, improved memory manage messages, etc 2008-01-30 19:36 erikgreenwald * src/adrt/master/Makefile.am, src/adrt/slave/Makefile.am: rename isst_{master,slave} to adrt_... since it's now generalized 2008-01-30 20:03 erikgreenwald * src/adrt/slave/load.c: now approaching critical ugliness 2008-01-30 21:07 brlcad * NEWS, src/librt/dg_obj.c, src/mged/rtif.c: get/check the exit code of the process so that rt doesn't just say it completed. have it report that it failed if the return status is non-zero. mods include the calls for windows so needs some testing. implemented in response to user confusion over failed rt runs that die early or never start and having mged say they completd. 2008-01-30 21:08 brlcad * configure.ac, src/conv/iges/Makefile.am, src/librt/Makefile.am, src/libwdb/Makefile.am, src/proc-db/Makefile.am: make openNURBS a build requirement, so we have to either build it or use a system openNURBS library now .. will need to test/fix compilations on other platforms but this does hopefully fix some of the linux build problems john was seeing on ubuntu. 2008-01-30 21:52 erikgreenwald * src/libtclcad/Makefile.am: update dependancy 2008-01-30 22:01 brlcad * NEWS: parker has made mods to the intensity handling for ghosted images, the main display no longer displays the extra/premature/misleading menu options (Edit, Steps) 2008-01-30 22:14 brlcad * NEWS: parker also added/ported several missing framebuffer tools including those needed by archer. includes bw-pix, bwmod, pix-bw, pixmatte. probably others. 2008-01-31 17:28 bob1961 * misc/win32-msvc8/brlcad/brlcad.sln: Added the following projects: asc2pix asc2nmg bot2bldxf bot2raw bot_shell2vtk cy2g dxf2g enf2g euclid_format euclid_unformat euclid2g g2acad g2adrt g2dxf g2euclid1 g2euclid g2nff g2nmg g2obj g2shellrect g2tankill g2var g2vrml g2x3d g42g5 g52g4 nastran2g nmg2bot nmg2rib nmg2sgp pix2asc ply2g poly2bot tankill-g vg2 2008-01-31 17:46 bob1961 * misc/win32-msvc8/dxf2g/dxf2g.vcproj, misc/win32-msvc8/g2dxf/g2dxf.vcproj: Upgraded to MSVC 8.0 2008-01-31 17:47 bob1961 * misc/win32-msvc8/asc2nmg, ... lots of other files ..., misc/win32-msvc8/vg2g, misc/win32-msvc8/vg2g/vg2g.vcproj: Initial check-in. 2008-01-31 18:02 bob1961 * misc/win32-msvc8/viewpoint2g, misc/win32-msvc8/viewpoint2g/viewpoint2g.vcproj: Initial check-in. 2008-01-31 19:36 bob1961 * misc/win32-msvc8/Makefile.am: Added a few project files to EXTRA_DIST. 2008-01-31 20:15 bob1961 * include/config_win.h: Included fcntl.h and added a few #define. 2008-01-31 20:16 bob1961 * include/raytrace.h: Added declaration for nmg_vlist_to_eu(). 2008-01-31 21:59 bob1961 * src/conv/asc-nmg.c, src/conv/asc2pix.c, src/conv/bot-bldxf.c, src/conv/bot-raw.c, src/conv/bot_shell-vtk.c, src/conv/conv-vg2g.c, src/conv/cy-g.c, src/conv/dxf-g.c, src/conv/enf-g.c, src/conv/euclid-g.c, src/conv/euclid_format.c, src/conv/euclid_unformat.c, src/conv/g-acad.c, src/conv/g-adrt.c, src/conv/g-dxf.c, src/conv/g-euclid.c, src/conv/g-euclid1.c, src/conv/g-nff.c, src/conv/g-obj.c, src/conv/g-shell.rect.c, src/conv/g-tankill.c, src/conv/g-var.c, src/conv/g-vrml.c, src/conv/g-x3d.c, src/conv/nastran-g.c, src/conv/nmg-rib.c, src/conv/nmg-sgp.c, src/conv/pix2asc.c, src/conv/ply-g.c, src/conv/poly-bot.c, src/conv/tankill-g.c: Mods to get things working on Windows. 2008-01-31 22:03 bob1961 * include/config_win.h: Change uintXX_t #defines to typedefs. 2008-01-31 22:11 bob1961 * src/libtclcad/tclcad.c: On windows, include fcntl.h to quell warnings about O_BINARY macro redefinition. 2008-01-31 22:12 bob1961 * src/libbu/temp.c: Remove unnecessary includes. 2008-01-31 23:13 bob1961 * src/librt/dg_obj.c: Added missing retcode in the windows section. 2008-02-01 15:08 erikgreenwald * src/libbu/image.c: PPM save support 2008-02-01 15:08 erikgreenwald * src/adrt/libutil/Makefile.am, src/adrt/libutil/image.c, src/adrt/libutil/image.h: libbu is now responsible for saving images 2008-02-01 15:33 erikgreenwald * include/bu.h, src/libbu/image.c: change image data type to be unsigned 2008-02-01 15:40 erikgreenwald * src/util/bwdiff.c: use bu_exit() instead of bu_log (I think that's what was intended) 2008-02-01 16:38 brlcad * TODO: the bug of mged outright not working is no longer a problem 2008-02-02 06:55 brlcad * include/config_win.h: cleanup and reorganize. organize/group together and annotate the various sections and the make note of the bogus defines that are really bad/counterproductive hacks. add fcntl.h too since it needed with io.h 2008-02-02 06:58 brlcad * include/bu.h, include/config_win.h: check for _MAX_PATH in bu.h 2008-02-02 07:44 brlcad * include/machine.h: meh 2008-02-02 07:45 brlcad * src/canon/ipustat.c, src/conv/asc2g.c, src/conv/conv-vg2g.c, src/conv/g-obj.c, src/conv/g-shell.rect.c, src/conv/g2asc.c, src/conv/patch/patch-g.c, src/fb/bw-fb.c, src/fb/fb-pix.c, src/fb/fb-png.c, src/fb/pix-fb.c, src/fb/png-fb.c, src/fb/pp-fb.c, src/fbed/pos_pad.c, src/libbu/backtrace.c, src/libbu/bomb.c, src/libbu/mappedfile.c, src/libfb/fbserv_obj.c, src/libfb/if_disk.c, src/libpkg/pkg.c, src/librt/nirt.c, src/librt/wdb_obj.c, src/libtermio/termio.c, src/mged/fbserv.c, src/mged/ged.c, src/mged/rtif.c, src/mged/utility1.c, src/remrt/remrt.c, src/rt/do.c, src/rt/hurt.c, src/rt/main.c, src/util/buffer.c, src/util/bwmod.c, src/util/dpix-pix.c, src/util/files-tape.c, src/util/gencolor.c, src/util/pix-bw.c, src/util/pix-yuv.c, src/util/pixbustup.c, src/util/pixmatte.c, src/util/pixtile.c, src/util/yuv-pix.c: header cleanup, remove fcntl.h inclusions that don't seem to be needed 2008-02-02 08:16 brlcad * configure.ac, src/nirt/nirt.c, src/remrt/rtsrv.c: remove check for fcntl.h 2008-02-02 08:51 brlcad * include/config_win.h, src/adrt/bench/bench.c, src/adrt/bench/main.c, src/adrt/isst/main.c, src/adrt/isst/observer.c, src/adrt/libcommon/canim.c, src/adrt/libcommon/env.c, src/adrt/libcommon/pack.c, src/adrt/librender/camera.c, src/adrt/libtie/tie.c, src/adrt/libtie/tie_kdtree.c, src/adrt/libtienet/tienet_master.c, src/adrt/libtienet/tienet_slave.c, src/adrt/libtienet/tienet_util.c, src/adrt/libutil/camera.c, src/adrt/master/main.c, src/adrt/master/master.c, src/adrt/rise/master/dispatcher.c, src/adrt/rise/master/main.c, src/adrt/rise/master/master.c, src/adrt/rise/observer/main.c, src/adrt/rise/observer/observer.c, src/adrt/rise/slave/main.c, src/adrt/slave/main.c, src/anim/anim_hardtrack.c, src/anim/anim_orient.c, src/anim/chan_add.c, src/anim/chan_mult.c, src/external/Unigraphics/ug-g.c, src/libfb/if_X.c, src/libfb/if_X24.c, src/libfb/if_ab.c, src/libfb/if_ogl.c, src/proc-db/brep_cube.cpp, src/proc-db/brep_simple.cpp, src/proc-db/pix2g.c, src/proc-db/tea.c, src/proc-db/torii.c, src/sig/a-d.c, src/sig/c-d.c, src/sig/d-a.c, src/sig/damdf.c, src/sig/dauto2.c, src/sig/dconv.c, src/sig/ddisp.c, src/sig/dfft.c, src/sig/dpeak.c, src/sig/dsel.c, src/sig/dstat.c, src/sig/fhor.c, src/sig/ihist.c, src/sig/syn.c, src/sig/u-bw.c, src/util/bw3-pix.c, src/util/bwstat.c, src/util/dsp_add.c, src/util/orle-pix.c, src/util/pixhist3d-pl.c, src/util/pixrot.c, src/util/pixscale.c, src/util/plcolor.c, src/util/ttcp.c, src/util/wavelet.c: protect all of the unistd.h inclusions for portability 2008-02-03 06:09 brlcad * NEWS: Bob added the rest of BRL-CAD's geometry converters for the Windows port. looks like it's all of them, or pretty close. he added asc2pix asc2nmg bot2bldxf bot2raw bot_shell2vtk cy2g dxf2g enf2g euclid_format euclid_unformat euclid2g g2acad g2adrt g2dxf g2euclid1 g2euclid g2nff g2nmg g2obj g2shellrect g2tankill g2var g2vrml g2x3d g42g5 g52g4 nastran2g nmg2bot nmg2rib nmg2sgp pix2asc ply2g poly2bot tankill-g viewpoint2g 2008-02-03 13:38 brlcad * include/raytrace.h: quell warning about free shadowing free() 2008-02-03 13:53 brlcad * src/conv/cy-g.c: considerably revamp how the header is processed. make NAME be the name of the db object created. stash header lines as _GLOBAL attributes (probably belong on object, but close enough). improve error handling and supported header element processing. clean up the style and ws while we're at it. 2008-02-03 17:18 brlcad * src/other/tcl/generic/tcl.h: quell compilation shadow warnings 2008-02-03 17:30 brlcad * AUTHORS, src/vas4/vas_io.c: apply a patch to vas4 from Ed Schouten (provided over IRC) that fixes an sgtty vs termios mixup in the # inclusion/define foo. 2008-02-03 17:31 brlcad * src/conv/cy-g.c: quell float comparison and alloc type warnings, print the header line 2008-02-03 17:49 brlcad * doc/deprecation.txt, src/vas4/vas4.1, src/vas4/vas4.c: officially mark vas4 for removal, deprecated as of the 7.12 release. if nobody complains or has a good reason for keeping it, the tool can be removed after a few minor releases. 2008-02-03 17:51 brlcad * src/vas4/vas_io.c: s/OXTABS/TAB3/ in the HAVE_TERMIOS_H section where tab expansion is turned on 2008-02-03 18:14 brlcad * src/conv/cy-g.c: add support for the rest of the cy header entities seen on a sample scan provided by toxygen (via irc). this includes detecting DATE, NAME, LGMIN, LGMAX, LGINCR, LTMIN, LTMAX, RMIN, RMAX, LGSHIFT, SCALE, and RPROP. (in addition to the others that were originally supported: NLT, LTINCR, RSHIFT, SPACE, NLG, and DATA) 2008-02-04 00:11 brlcad * src/conv/cy-g.c: stash everything into a var for now just in case we do try to use it, presume rprop is a radius scale factor and use 'scale' for the delta_z scale factor 2008-02-04 01:02 brlcad * src/tclscripts/mged/openw.tcl: since the directory is being pulled dynamic from bu_brlcad_data, don't pay any attention to what is in the .mgedrc (probably shouldn't write it out). more importantly, don't use the -display option with the $mged_browser so that it'll actually work. need some better way than calling up tcl_platform(os)...freakin fugly. this does, however, make help finally work on mac os x. 2008-02-04 01:06 brlcad * NEWS: fixed mged's browser-based help on Mac OS X. multiple problems causing help to not work including users' .mgedrc pointing to a non-existent previous install as well as trying to invoke with the open command while still using the X11-specific -display options that mozilla obeys. 2008-02-04 16:48 bob1961 * include/config_win.h: Mods to get things compiling again. 2008-02-04 18:39 bob1961 * src/nirt/nirt.c: Added b to the flag parameter of fopen(). 2008-02-04 18:51 bob1961 * src/nirt/parse_fmt.c: ValTab[] should not be const. Also added b to the flag parameter of fopen(). 2008-02-04 19:11 bob1961 * misc/win32-msvc8/tkstub/tkstub.vcproj: tkStubImg.c no longer needed. 2008-02-05 00:30 brlcad * src/adrt/libtienet/tienet_master.c: use bu_fgets for win32 portability 2008-02-05 06:01 brlcad * src/conv/comgeom/cvt.c: mark file as binary for windows 2008-02-05 06:02 brlcad * src/conv/cy-g.c: add some comments about what some of the header fields mean 2008-02-05 06:03 brlcad * src/conv/cy-g.1: be explicit that the .cy file format is what cyberware is calling their Echo file format 2008-02-05 06:05 brlcad * src/conv/fast4-g.c: they're file POINTERS not file descriptors, so use fp instead of fd when stashing fopen() pointers. fd's are for open(). 2008-02-05 08:21 brlcad * bench/pixcmp.c, src/adrt/bench/bench.c, ... lots of other files ..., src/librt/shoot.c, src/librt/wdb_obj.c, src/mged/animedit.c: large conversion/specification of the fopen callers as to whether the file is binary or text in order to help the Windows port along. Unless specifically certain that there is no binary fields or EOL assumptions in the file reading, the files are marked as binary in order to preserve format with other platforms. also, be more consistent on using fp for fopen'd file pointers instead of fd (which would be an open'd file descriptor). 2008-02-05 15:41 brlcad * src/libbu/vls.c: overhaul comments and style throughout so that all routines look and behave similarly. inject spacing throughout for grouping and readability. 2008-02-05 15:48 brlcad * include/bu.h: consistently make all magic numbers unsigned long integers. this helps quell warnings aside from just better representing the nature of the magic. 2008-02-05 15:51 brlcad * src/libbu/badmagic.c, src/libbu/bu_tcl.c, src/libbu/list.c, src/libbu/magic.c: quell warnings, more consistency conversion for using unsigned longs for all magic numbers 2008-02-05 15:53 brlcad * src/libbu/bomb.c: need fcntl.h for open() 2008-02-05 16:52 bob1961 * src/mged/fbserv.c: Need code to increment the socket layer's reference count.\n 2008-02-05 16:52 bob1961 * src/libfb/fbserv_obj.c: Need code to increment the socket layer's reference count.\n 2008-02-05 19:42 brlcad * include/bu.h, src/libbu/temp.c: make bu_temp_file take a size_t 2008-02-05 19:44 brlcad * src/libbu/bu_tcl.c, src/libbu/cmdhist.c, src/libbu/cmdhist_obj.c, src/libbu/hist.c, src/libbu/htester.c, src/libbu/image.c, src/libbu/list.c, src/libbu/malloc.c, src/libbu/observer.c, src/libbu/parallel.c, src/libbu/rb_delete.c, src/libbu/rb_diag.c, src/libbu/rb_extreme.c, src/libbu/rb_order_stats.c, src/libbu/rb_rotate.c, src/libbu/semaphore.c, src/libbu/stat.c, src/libbu/units.c, src/libbu/vfont.c: quell a slew of compilation warnings regarding unused vars, type mismatches, and more. 2008-02-05 20:19 erikgreenwald * src/adrt/libtexture/Makefile.am, src/adrt/libtienet/Makefile.am, src/adrt/master/Makefile.am: add TCL_CPPFLAGS for bu.h 2008-02-05 20:26 bob1961 * src/librt/db_open.c: Initialize dbip->dbi_fd in db_open() to -1. 2008-02-05 22:00 bob1961 * src/conv/g-dxf.c: Fixed a bug that was walking off the end of argv. 2008-02-05 22:13 bob1961 * misc/win32-msvc8/g2euclid/g2euclid.vcproj, misc/win32-msvc8/g2euclid1/g2euclid1.vcproj, misc/win32-msvc8/g2tankill/g2tankill.vcproj, misc/win32-msvc8/g2var/g2var.vcproj: Changed the executable's name. 2008-02-05 22:28 bob1961 * misc/win32-msvc8/fb2cmap/fb2cmap.vcproj, misc/win32-msvc8/fb2rle/fb2rle.vcproj: Changed the executable's name. 2008-02-06 06:23 brlcad * src/conv/cy-g.c: refactor and simplify. make the header scanning case insensitive. refering to a reference implementation, annotate which header lines are required and which are optional. add checks for FILLED, SMOOTHED, INSIDE_OUT, and THETA_RIGHTHAND. 2008-02-06 07:45 brlcad * src/conv/cy-g.c: read in the data files into an in-memory buffer before processing. this limits processing to models that will fit in memory, but does make things a whole lot easier for processing the dataset later for things like generating a mesh. also added a check for VOIDVAL's, but just continuing is probably the wrong thing to do. 2008-02-06 10:00 brlcad * NEWS: heh, and I just ran into that same bug myself.. Bob fixed a bug in g-dxf where it was using bu_optind as the argv[] index incorrectly. this would cause an env var to get set for the main object name (at least on Mac OS X) giving something like TERM_PROGRAM=Apple_Terminal as the object name instead of what was specified on the command line.. resulting in something rather unexpected of course. 2008-02-06 15:59 bob1961 * include/config_win.h: Added macro for strncasecmp. 2008-02-06 20:48 bob1961 * src/conv/tankill-g.c: Another mod for windows. 2008-02-07 20:19 brlcad * TODO: add in a feature request from jennifer lagaz (and the modeling team previously) to have rtarea report local units instead of mm^2 and to have an option to override the local units with a command-line option at run-time. 2008-02-07 20:22 brlcad * TODO: need to reverify that rtarea is indeed producing correct data values per comments/requests/investigations from lagaz, saucier, and mallory that it's reporting differences upwards of several % difference (when it should be < 1% error for most views). also finally fix the naming convention nomenclature for presented/exposed surface areas. 2008-02-07 21:48 starseeker * doc/Makefile.am: Add start of BRL-CAD Bibliography file (BibTeX format). 2008-02-07 21:50 starseeker * doc/BRL-CAD.bib: Add start of BRL-CAD Bibliography file (BibTeX format) - add actual bib file. 2008-02-08 06:39 brlcad * src/adrt/master, src/adrt/slave: targets apparently renamed to adrt_master and adrt_slave instead of isst_* 2008-02-08 07:17 brlcad * TODO: bob indicates he fixed query_ray 2008-02-08 07:31 brlcad * src/librt/db_open.c: fix a bug introduced with the changeover of db_open to only using standard i/o for all platforms instead of unix i/o with an uninitialized dbi_fd. the problem was related to the failure case where null is returned, which mged intentionally uses to determine if a file exists, where since with unix i/o a fp and an fd are independently opened (instead of deriving one from the other) so the cleanup code was closing the other. the close code was only checking for non-negative but with dbi_fd unitialized, this was fd 0 (i.e. stdin) which in turn caused Bad Things (tm) to happen. the bug was actually only exposed through the spdi regression test since it directs input which is read in with bu_vls_gets (i.e. via gets and stdin), which was of course failing after the db_open call that closed fd 0. most interestingly, subsequent calls to db_open that succeed would create a valid stream and assign it fd 0. add some file comments while we're in here. 2008-02-08 07:48 brlcad * include/raytrace.h, src/librt/db_inmem.c, src/librt/db_io.c, src/librt/db_open.c: awesome, dbi_fd is already in a struct sections marked as private (for librt only) and subject to change .. so change^Wremove it. there are surprisingly few uses even in librt, so it seems really safe to remove the fd and avoid repeating the same bug of corrupting file pointer streams by closing their associated descriptor. go ahead and remove it, callers can always call fileno(dbi_fp) if they really need to do unix i/o. 2008-02-08 07:51 brlcad * TODO: reprioritize to the items that are really the most critical for release, and the ones that are not.. live with the rt crash and incrtcl configuration issue for another release, but really need to fix the mged issues before doing a rollout. 2008-02-08 08:12 brlcad * src/libbu/vls.c: rewrite bu_vls_gets() to use bu_fgets() instead of manually doing a getc() iteration loop by itself. this lets the routine get the cross-platform benefits of bu_fgets() for handling Windows files stream carriage returns aside from just simplifying the implementation. 2008-02-08 08:30 brlcad * src/libbu/units.c: ws 2008-02-08 08:33 brlcad * include/bu.h, src/libbu/vls.c: quell all compilation warnings and convert bu_vls_strncat, bu_vls_strncpy, and bu_vls_strncmp to all use a size_t parameter instead of an int len. 2008-02-08 18:12 erikgreenwald * NEWS, src/mged/ged.c, src/mged/mged.1: foreground mode is now default for MGED. -b option added to background. 2008-02-08 18:48 erikgreenwald * NEWS, src/other/tcl/ChangeLog, src/other/tcl/README, ... lots of other files ..., src/other/tk/win/ttkWinTheme.c: upgrade tcl/tk to 8.5.1 2008-02-08 19:18 erikgreenwald * NEWS, src/other/libpng/ANNOUNCE, src/other/libpng/CHANGES, ... lots of other files ..., src/other/libpng/scripts/pngw32.rc: update libpng to 1.2.24 2008-02-08 19:20 erikgreenwald * TODO: scratch off the foreground/background issue 2008-02-08 23:33 starseeker * doc/BRL-CAD.bib: Add start of BRL-CAD Bibliography file (BibTeX format). 2008-02-09 00:12 starseeker * doc/BRL-CAD.bib: More references. 2008-02-09 12:23 brlcad * src/adrt/Makefile.am, src/other/tcl/Makefile.am: distcheck changes 2008-02-09 14:21 brlcad * src/util/Makefile.am: include rtwizard.bat in the dist 2008-02-09 14:23 brlcad * Makefile.am: update the distcheck hook for finding missing files to work with Subversion 2008-02-10 16:32 brlcad * src/other/URToolkit/cnv/rletoabA62/rletoabA62.c, src/other/URToolkit/tools/clock/rleClock.c: quell compilation warnings 2008-02-10 16:37 brlcad * src/other/URToolkit/tools/rleccube.c, src/other/URToolkit/tools/to8.c: quellage 2008-02-10 20:52 brlcad * src/libbn/vert_tree.c: gah, don't include raytrace.h in libbn 2008-02-10 22:35 brlcad * include/bn.h, include/nmg.h, include/pkg.h, include/plastic.h, include/raytrace.h, include/rtgeom.h, include/shadefuncs.h, include/spm.h, src/librt/g_nmg.c, src/librt/g_vol.c, src/librt/nmg_bool.c, src/librt/nmg_ck.c, src/librt/nmg_class.c, src/librt/nmg_eval.c, src/librt/nmg_extrude.c, src/librt/nmg_fcut.c, src/librt/nmg_fuse.c, src/librt/nmg_index.c, src/librt/nmg_info.c, src/librt/nmg_inter.c, src/librt/nmg_misc.c, src/librt/nmg_mk.c, src/librt/nmg_mod.c, src/librt/nmg_plot.c, src/librt/nmg_pr.c, src/librt/nmg_pt_fu.c, src/librt/nmg_rt_isect.c, src/librt/nmg_rt_segs.c, src/librt/nmg_visit.c: convert all of the magic numbers from long integers to unsigned long integers. that in turn cascades similar changes where there is struct aliasing and pointers being stashed/used as object types 2008-02-10 22:52 brlcad * src/burst/burst.c, src/burst/ui.c: use bu_temp_file() for opening a temp file instead of tmpnam(), specify all i/o as binary for Windows 2008-02-11 00:17 bharder * doc/book/VolumeIV.xml: Fixed mis-reading (or was it backwards day?) of a table where all entries were reversed from what they needed to be, and a minor typo. 2008-02-11 02:52 brlcad * src/rt/rt.1: document the SIGINFO/SIGUSR1 progress signals that rt will respond to 2008-02-11 02:55 brlcad * NEWS: add a note for rt's documentation update just so more folks are aware that rt will in fact provide progress status if it receives a SIGINFO/SIGUSR1 signal 2008-02-11 03:25 brlcad * src/libfb/if_ogl.c: make sure the opengl context was initialized 2008-02-11 19:14 bob1961 * src/other/libpng/png.h, src/other/tcl/generic/tclInt.h: Minor mods for getting things to compile on Windows. 2008-02-11 22:36 bob1961 * src/tclscripts/archer/Archer.tcl: Uncomment the raytrace control panel menu item. 2008-02-12 20:17 bob1961 * src/archer/archer: Mods to set max width and height. 2008-02-12 20:17 bob1961 * src/tclscripts/archer/ArcherCore.tcl: Only erase ground plane if it's being drawn. 2008-02-12 20:18 bob1961 * src/tclscripts/archer/Archer.tcl: Make updateTheme and Load public methods. 2008-02-12 20:57 erikgreenwald * src/nirt/command.c: re-write backout code to actually work 2008-02-12 20:58 erikgreenwald * TODO: nirt backout is fixed 2008-02-13 17:42 brlcad * src/lgt/resource.c: ws and header 2008-02-13 17:43 brlcad * src/lgt/do_options.c: cleanup function, unprotect since bu_avail_cpus() does the protection for us. 2008-02-13 17:47 brlcad * src/lgt/do_options.c: if there's only one proc, don't bother asking 2008-02-13 17:50 brlcad * src/lgt/resource.c: wtf, the functions in resource.c arent' even used. get rid of it. 2008-02-13 17:51 brlcad * src/lgt/Makefile.am: remove the unused resource.c 2008-02-13 18:21 brlcad * src/libbu/parallel.c: make bu_avail_cpus() behave neatly if we're not compiling PARALLEL so that it just returns 1 2008-02-13 18:42 brlcad * include/machine.h, src/lgt/do_options.c, src/libbu/parallel.c, src/libfb/if_ogl.c, src/rt/hurt.c, src/rt/main.c, src/rt/opt.c, src/util/pixmorph.c: remove/refactor DEFAULT_PSW from machine.h so that now it's really just up to rt_avail_cpus() to report the max and up to the user to decide how many. the default is now consistently set the same for everyone to the number of processors available. if someone needs to re-port to an old CRAY and specifying -P1 from userland isn't sufficient, we can refactor a bu_default_cpus() routine with better logic. one step closer to removing machine.h 2008-02-13 19:10 brlcad * include/machine.h, src/libbu/convert.c: get rid of IEEE_FLOAT .. the code didn't actually use it except for one case in 'cv' and was set to 1 for all platforms. if something needs to know, there are configure tests for ieee float. 2008-02-13 21:31 brlcad * include/machine.h, src/libbu/malloc.c: remove MALLOC_NOT_MP_SAFE from machine.h, moving it to src/libbu/malloc.c and enabling it for all platforms. there were only a few old platforms that were not specifically set as malloc-not-mp-safe and it's not clear which of those were actually verified as safe so just turn it on for all for now. 2008-02-13 21:54 brlcad * src/rt/Makefile.am, src/rt/scat.h: scat.h seems to be completely unused, remove it 2008-02-13 23:09 brlcad * src/libbu/parallel.c: ansify 2008-02-13 23:10 brlcad * include/machine.h, src/libtermio/termio.c, src/libtermio/termio_win32.c: remove FOPEN_MAX from machine.h, moving it over to the only single place it's used over in libtermio. 2008-02-13 23:16 brlcad * include/bu.h, include/machine.h: with the exception of the convex (where it was 'long long'), bitv_t was set to 'long' for all platforms. so move the bitv_t typedef out of machine.h and into bu.h 2008-02-14 00:09 brlcad * include/bu.h, include/machine.h, include/vmath.h, src/burst/burst.h, src/burst/gridrotate.c, src/conv/iges/brlabs.c, src/conv/iges/splinef.c, src/liboptical/turb.c: move genptr_t and fastf_t from machine.h over to bu.h including the various FASTF define limits. might later consider moving them to their own header if/when they become configurable or if bu.h gets broken up. 2008-02-14 00:17 brlcad * include/machine.h: IBM_FLOAT is unused 2008-02-14 00:20 brlcad * include/common.h, include/machine.h: move HIDDEN over to common.h since it's generic. could be set by configure. 2008-02-14 00:30 brlcad * include/bu.h, include/machine.h: move BITV_MASK over from machine.h to bu.h; remove the hypot() decl/protections outright even if they didn't make it to c89 (it made c99).. if we have a live system that has the problem, the appropriate configure foo can be added to provide the decl/impl 2008-02-14 00:48 brlcad * include/bu.h, include/machine.h: migrate the FAST comments 2008-02-14 01:02 brlcad * include/bu.h, include/machine.h: move MAX_PSW from machine.h over to bu.h, fixing the value at 1024 for now (the max value present in machine.h). this makes that the hard limit on number of procs/threads that may be spawned at a minor (trivial) memory expense when not in full use. ray-trace performance does not seem to be affected by the size of the tables. 2008-02-14 02:24 brlcad * include/bu.h, include/machine.h, src/libbu/bitv.c, src/librt/prep.c: convert to BU_BITV_SHIFT and BU_BITV_MASK for consistency 2008-02-14 03:53 brlcad * src/other/Makefile.am: if building with --enable-only-benchmark or --enable-only-librtserver we STILL need to traverse into zlib, libpng, libregex, opennurb, and tnt since they're dependencies of other required libs. 2008-02-14 06:22 brlcad * BUGS, TODO: annotate the results of the OpenGL framebuffer bug hunt. the problem seems to be directly related to SMP processing with the ogl framebuffer interface. with PARALLEL turned off, it works just fine; with it on, the app crashes during the first OpenGL write call into the glx context likely indicating that either the context initialization is invalid (perhaps needing to occur per-thread) or data needs to be semaphore protected. either way, enough progress was made to push the interface back another release since the specific cause is still unclear. 2008-02-14 07:10 brlcad * include/bu.h, include/machine.h, src/libbu/bitv.c: rewrite how the BU_BITV_SHIFT is calculated, instead of using hard-coded constants in machine.h it now determines the value automatically at compile-time. this was achieved using an inlined integer log2 function that computes the size based on sizeof(bitv_t), using a technique from the U. of Kentucky's Aggregate Magic Algorithms collection. a cursory performance impact analysis was performed testing both unoptimized and optimized compilations (on Mac OS X 10.4, gcc 4.0.1) with results looking quite reasonable. performance did decrease moderately for unoptimized (by approx 10%) but also interestingly increased (by approx 10%) for optimized, undoubtedly due to cache/alignment and branching effects. this approach beats a configure-time test since it will acquire the correct shift at run-time from Universal/multi-architecture binaries. 2008-02-14 07:27 brlcad * include/machine.h, include/orle.h: obliterate BIG_ENDIAN and LITTLE_ENDIAN defines.. which actually weren't used anywhere except one use in orle. Most of the current code presently uses NATURAL_IEEE and REVERSED_IEEE instead. header cleanup too, make it clear that machine.h is now fully deprecated. 2008-02-14 07:29 brlcad * doc/deprecation.txt: document that the contents of machine.h was marked as deprecated during 7.10. the file was 'officially' emptied out in 7.12 and should be ready for removal after a few subsequent releases. 2008-02-14 07:53 brlcad * INSTALL, configure.ac, include/machine.h: Woo hoo.. machine.h is now gone. Refactor the last bit for defining PARALLEL by making a configure --enable-parallel option (which defaults to yes for now) with aliases for --enable-parallel-build, --enable-smp, and --enable-smp-build. only other bit remaining was sgi/mips settings, but at least irix6 does set them both with gcc and mipspro so they should be fine. 2008-02-14 08:14 brlcad * include/conf/PATCH: haven't been diligent in updating the patch on this minor as big changes have occurred, but at least the machine.h change begs for it to be bumped. 2008-02-14 08:14 brlcad * include/Makefile.am: don't install conf.h or machine.h 2008-02-14 08:41 brlcad * include/bn.h, include/brep.h, include/brlcad.h, include/bu.h, ... lots of other files ..., src/vdeck/parsarg.c, src/vdeck/vdeck.c, src/vdeck/vproc.c: this day has been coming for so long.. machine.h is dead, all hail machine.h's death. uninclude it everywhere and add a preprocessor warning for anyone else that is still including/using it. 2008-02-14 14:37 bob1961 * src/tclscripts/mged/text.tcl: Replaced Tcl's gets with a version that works with MGED's command window. 2008-02-14 14:43 bob1961 * src/mged/ged.c: Mods to get puts working wrt stdout and stderr. 2008-02-14 15:40 erikgreenwald * src/fb/fb-rle.c, src/fb/rle-fb.c, src/util/bw-rle.c, src/util/pix-rle.c, src/util/rle-pix.c: undefine USE_PROTOTYPES before including rle.h to avoid redefining 2008-02-14 17:04 bob1961 * misc/win32-msvc8/librt/librt.vcproj: Remove parse.c and comb.c 2008-02-14 18:26 erikgreenwald * src/libbu/avs.c, src/libbu/list.c, src/libbu/parallel.c, src/libbu/rb_delete.c, src/libbu/rb_diag.c, src/libbu/rb_order_stats.c, src/libbu/rb_rotate.c, src/libbu/vls.c: move from casting pointers to ints and printing %x to just printing %p 2008-02-14 19:05 brlcad * src/libfb/if_ogl.c: didn't intentionally leave debug turned on, turn it off 2008-02-14 20:29 bob1961 * include/config_win.h: Add define for inline. 2008-02-14 21:21 bob1961 * misc/win32-msvc8/asc2g/asc2g.vcproj: Replaced 7.11.0 with a variable. 2008-02-14 21:58 bob1961 * include/bu.h: For some reason inline causes problems with some of the apps on windows. So, for the moment, it gets if'ed out. 2008-02-14 22:06 bob1961 * include/bu.h: Oops! I checked in the wrong version. 2008-02-16 15:58 brlcad * src/nirt/if.c: the backout() callbacks are no longer needed/used since there isn't custom "reshooting" going on. we just backup the original ray origin now. 2008-02-16 16:10 brlcad * src/nirt/bsphere.c, src/nirt/command.c: initialize negative so we can do some sanity checking, re-add the DEBUG_BACKOUT code 2008-02-16 16:14 brlcad * src/nirt/bsphere.c, src/nirt/nirt.h: reduce the style inconsistency insanity via s/ -> /->/g 2008-02-16 16:17 brlcad * src/nirt/conversion.c, src/nirt/dist_def.c, src/nirt/if.c, src/nirt/nirt.h, src/nirt/read_mat.c: use vmath's DEG2RAD instead of the weak decl it was using 2008-02-16 16:38 brlcad * NEWS: Erik made nirt's -b backout option use the model's bounding sphere as the backout distance instead of the hard-coded 1000 define and also made it use a much simpler backout method by just backing up the original ray being fired by the model's bounding sphere size. vaguely recall there being a request regarding nirt's backout behavior, but don't see any hint of it. think it might have been a verbal request/inquiry. 2008-02-16 17:59 brlcad * NEWS, TODO: per recent commits, Bob indicated that he fixed or at least hack-patched several of the mged I/O bugs where gets in tcl mged wasn't working from stdin as well as stdout/stderr going to the command window instead of to the console. 2008-02-16 18:01 brlcad * src/nirt/command.c, src/nirt/interact.c: more s/ -> /->/g on files that weren't saved before the commit 2008-02-16 20:40 brlcad * sh/header.sh: leave authorship to the revision control system and AUTHORS file 2008-02-16 21:26 louipc * misc/archlinux/PKGBUILD.in: dependency and configure tweaks 2008-02-17 01:35 brlcad * include/bu.h, src/libbu/Makefile.am, src/libbu/endian.c: initial implementation of run-time byte order detection via bu_byteorder() that returns the current endian type. type endian type is one of BU_BIG_ENDIAN BU_LITTLE_ENDIAN and BU_PDP_ENDIAN for now, and uses a simple union test for finding the ordering. profiling showed that it was actually faster to compute than to lookup from a static iff the first test is the right one (which makes sense, avoids a load/read). we need *run-time* detection in order for Universal binaries to work correctly on Mac OS X. 2008-02-17 01:38 brlcad * src/conv/g-var.c: use the new bu_byteorder() instead of WORDS_BIGENDIAN 2008-02-17 01:40 brlcad * misc/win32-msvc7/libbu/libbu.vcproj, misc/win32-msvc8/libbu/libbu.vcproj, src/libbu/libbu.dsp: add new endian.c file to the windows build 2008-02-17 01:55 brlcad * src/libbu/endian.c: oop, save before commit. comment that stashing a static was tested. 2008-02-17 05:52 brlcad * src/fbed/font.h, src/fbed/getfont.c, src/lgt/char.c, src/lgt/font.h, src/lgt/getfont.c: eww, vfont handling is replicated in lgt and fbed. clean things up a bit by at least making them the same. probably need to 'restore' vfont back into the src hierarchy now that it's clear that there are actually sources that belong to it. 2008-02-17 05:53 brlcad * src/conv/ply-g.c: use bu_byteorder() instead of manually guessing the endian 2008-02-17 05:53 brlcad * src/conv/stl-g.c: header cleanup 2008-02-17 17:13 brlcad * src/lgt/getfont.c: don't need assert.h 2008-02-17 17:15 brlcad * src/fbed/fbed.c: ws and style consistency 2008-02-17 20:34 brlcad * configure.ac, misc/Makefile.am, misc/vfont, src/Makefile.am, src/lgt/Makefile.am, src/lgt/char.c, src/lgt/font.h, src/lgt/getfont.c, src/vfont, src/vfont/Makefile.am, src/vfont/font.h, src/vfont/getfont.c: move vfont back into the source tree so that code duplicated in fbed and lgt can be refactored into a library interface. this just moves the files into place and keeps things working. 2008-02-17 22:41 brlcad * src/lgt/char.c, src/lgt/ir.c, src/lgt/lgt.h, src/vfont/getfont.c: modify get_Font() to be get_font() and to take a logging callback (so we can pass it fb_log for fbed) 2008-02-17 22:48 brlcad * src/fbed/Makefile.am, src/fbed/char.c, src/fbed/execshell.c, src/fbed/extern.h, src/fbed/fbed.c, src/fbed/fill_buf.c, src/fbed/font.h, src/fbed/getfont.c, src/fbed/getinput.c, src/fbed/glob.c, src/fbed/loadfont.c, src/fbed/pos_pad.c, src/fbed/prnt.c, src/fbed/sgi_dep.c, src/fbed/try.c: convert fbed over to the new libvfont_nil 2008-02-17 22:56 brlcad * src/vfont/getfont.c: cleanup, use bu_brlcad_data to find the installed font resource files 2008-02-18 05:57 brlcad * doc/deprecation.txt, include/bu.h, src/libbu/log.c, src/libbu/vls.c: make the format specifier const for bu_log, bu_flog, bu_vls_printf, bu_vls_sprintf 2008-02-18 06:44 brlcad * src/fbed/Makefile.am, src/fbed/char.c, src/fbed/extern.h, src/fbed/fbed.c, src/fbed/fill_buf.c, src/fbed/loadfont.c, src/lgt/char.c, src/lgt/extern.h, src/lgt/ir.c, src/vfont/font.h, src/vfont/getfont.c: make get_font() return a struct vfont structure containing the data that was previously in globals. it's small enough that a struct copy should be fine, caller managers their own data instead of it coming from the library. 2008-02-18 07:27 brlcad * src/fbed/empty.c, src/fbed/fbed.c, src/fbed/fill_buf.c, src/fbed/getinput.c, src/fbed/pos_pad.c, src/fbed/sgi_dep.c, src/fbed/try.c: quell all extra compilation warnings 2008-02-18 08:23 brlcad * src/lgt/char.c, src/lgt/do_options.c, src/lgt/error.c, src/lgt/fb.c, src/lgt/grid_rotate.c, src/lgt/hmenu.c, src/lgt/ir.c, src/lgt/lgt.c, src/lgt/lgt.h, src/lgt/lgt_db.c, src/lgt/mat_db.c, src/lgt/octree.c, src/lgt/prnt.c, src/lgt/reflect.c, src/lgt/sgi_dep.c, src/lgt/texture.c, src/lgt/trie.c, src/lgt/vecmath.h: quell all extra compilation warnings save for assert() 2008-02-18 08:25 brlcad * src/vfont/font.h, src/vfont/getfont.c: quell warnings, name callback vfont_log() to avoid shadowing the log() math func 2008-02-18 08:28 brlcad * src/fbed/char.c, src/fbed/fbed.c, src/fbed/fill_buf.c, src/lgt/char.c, src/lgt/ir.c, src/vfont/font.h, src/vfont/getfont.c, src/vfont/vfont.h: rename font.h back to its original vfont.h name just to be consistent with it being an interface (albeit still considered 'private', hence not in include/) 2008-02-18 08:35 brlcad * src/vfont/Makefile.am, src/vfont/vfont.h: get rid of the WORDS_BIGENDIAN use in favor of bu_byteorder() run-time detection 2008-02-18 08:52 brlcad * doc/html/ReleaseNotes/email4.4.html, doc/html/manuals/Install.html: obliterate some more phone numbers, even in older historic docs just in case 2008-02-18 18:34 brlcad * AUTHORS: paul randal with one ell? 2008-02-18 18:51 brlcad * AUTHORS: his first name was apparently cormac, he just went my corey 2008-02-18 22:45 brlcad * db/db.php, include/anim.h, include/common.h, include/db5.h, ... lots of other files ..., src/util/wavelet.c, src/util/yuv-pix.c, src/vas4/vas4.1: categoric removal of the source, author, and function list sections in files, particularly the source file header sections. those are all sections of data that are more often incomplete or out-of-date than they are correct and are a maintenance burden that has no direct benefit. the 'source' section (i.e. the address) is mostly irrelevant with the open sourcing. the authorship is better comprehensively preserved through the SCM revision control system, and was the most common to be incomplete data (aside from the matter of not having clear criteria for knowing when someone could add their name to the list). the function lists were rarely ever updated, often listing non-existent functions or missing more than it documented. the header is meant to be a description of the file's purpose more than its content -- functions should have their own doxygen comment header. this commit modifies all files with a 21005 zip code. additionally, there is an interesting assertion by brian fitzpatrick and ben collins-sussman in their poisonous people talk regarding how having authors listed in source files is perceived by potential new contributors. namely that it can become a barrier to their contributing (whether due to ownership, politics, affiliations, personalities, "I'm not worthy", or some other reason). it's an interesting thought, though the real reason remains that there are bigger maintenance burden and incompleteness problems for long term usability. 2008-02-18 23:09 brlcad * src/remrt/rtsrv.c: match bu_log's new const goodness 2008-02-18 23:30 brlcad * sh/footer.sh: weird.. don't use single-line comments for headers, use comment blocks like we do for .c files 2008-02-18 23:32 brlcad * src/adrt/libtie/tie.h, src/adrt/libtie/tie_define.h, src/adrt/libtie/tie_kdtree.h, src/adrt/libtie/tie_struct.h, src/burst/HmGetc.c, src/bwish/consoleMain.c, src/conv/iges/g-iges.c, src/fbed/empty.c, src/fbed/glob.c, src/fbed/pos_pad.c, src/fbed/try.c, src/libbu/mro.c, src/librt/cut.c, src/mged/joints.h, src/remrt/remrt.c: remove some remaining references to RCS vars 2008-02-18 23:35 brlcad * src/burst/HmGetc.c, src/burst/HmGlob.c, src/lgt/ir.c: de-LINT 2008-02-19 01:30 brlcad * src/adrt/librender/path.c, src/anim/anim_cascade.c, ... lots of other files ..., src/vdeck/parsarg.c, src/vdeck/vdeck.c, src/vdeck/vproc.c: start unraveling the horrible whitesmith-pico indentation style that put statements on the same line as open parens (and still indenting the paren). put the statement on the next line. also caught a few that just had comments following the open paren, but move them down a line too. applying auto-style formatting should clean up their indentation later. one step closer towards global consistency.. 2008-02-19 01:57 brlcad * HACKING: add more details about the indentation/whitespace style as being BSD KNR and/or K&R style with 4-char indents. 2008-02-19 01:59 brlcad * src/adrt/libtexture/texture_perlin.h: stragglers not saved 2008-02-19 01:59 brlcad * src/librt/comb.c: stragglers not saved 2008-02-19 09:20 d_rossberg * include/config_win.h: some C++ libraries (e.g. STL) need the inline specifier 2008-02-19 14:59 erikgreenwald * src/Makefile.am: vfont must be build before fbed and lgt 2008-02-19 14:59 erikgreenwald * src/librt/Makefile.am: minor cleanup of the DEPENDS stuff 2008-02-19 15:00 erikgreenwald * src/fbed/Makefile.am, src/lgt/Makefile.am, src/libbu/Makefile.am: dependancy fixes 2008-02-19 15:36 brlcad * sh/header.sh: one line too many on wrapped headers 2008-02-19 16:32 brlcad * include/RtServerImpl.h, include/bn.h, include/brlcad_version.h, include/bu.h, include/cmd.h, include/db.h, include/db5.h, include/dm-X.h, include/dm-glx.h, include/dm-ogl.h, include/dm-pex.h, include/dm-plot.h, include/dm-ps.h, include/dm-tk.h, include/dm-wgl.h, include/dm.h, include/fbserv_obj.h, include/light.h, include/mater.h, include/nmg.h, include/nurb.h, include/opennurbs_ext.h, include/optical.h, include/orle.h, include/photonmap.h, include/plastic.h, include/raytrace.h, include/regex.h, include/rtgeom.h, include/rtserver.h, include/shadefuncs.h, include/shadework.h, include/solid.h, include/spm.h, include/vector.h, include/vector_fpu.h, include/vector_x86.h, include/vfont-if.h, include/vmath.h, include/wdb.h: the start of applying a consistent style to the sources using the (emacs-based) source formatting scripts. the inlined neatly aligned tabs are a real pain but hopefully is a one-time shift/indent change (to 4-char indents, tab stops at 8). still testing but looks like it's working well. 2008-02-19 16:33 brlcad * misc/batch-indent-region.el: force case-label lines to indent since that's not set with stroustrup or k&r. probably need to provide our own KNF c-style-alist settings, but good enough for now to deal with case indents. 2008-02-19 16:37 brlcad * sh/indent.sh: ignore the m4 files. gets them pretty much totally wrong. 2008-02-19 18:22 bob1961 * misc/win32-msvc8/btclsh/btclsh.vcproj: Set the "Compile As" option to default. 2008-02-19 18:27 bob1961 * include/bu.h: Mods to accomodate a few apps on Windows that still have problems with inline. 2008-02-19 18:28 bob1961 * include/tclcad.h: Fix Tclcad_Init declaration. 2008-02-19 18:32 bob1961 * src/other/tcl/generic/tclInt.h: Add more checks before defining inline. 2008-02-19 18:34 bob1961 * src/other/blt/src/bltInt.h: Add more checks before defining inline. 2008-02-19 18:35 bob1961 * src/bwish/cadAppInit.c: Cad_AppInit returns an int. 2008-02-19 18:36 bob1961 * src/libtclcad/tclcad.c: Tclcad_Init returns an int. 2008-02-19 19:16 erikgreenwald * src/vfont/Makefile.am: need to be able to find tcl.h (for bu.h) 2008-02-19 20:02 bob1961 * src/fbed/fbed.c: I believe we only need SIGCLD or SIGCHLD. 2008-02-19 20:18 erikgreenwald * NEWS, src/other/libpng/ANNOUNCE, src/other/libpng/CHANGES, src/other/libpng/INSTALL, src/other/libpng/KNOWNBUG, src/other/libpng/LICENSE, src/other/libpng/README, src/other/libpng/Y2KINFO, src/other/libpng/configure, src/other/libpng/configure.ac, src/other/libpng/libpng-1.2.24.txt, src/other/libpng/libpng-1.2.25.txt, src/other/libpng/libpng.3, src/other/libpng/libpngpf.3, src/other/libpng/png.5, src/other/libpng/png.c, src/other/libpng/png.h, src/other/libpng/pngconf.h, src/other/libpng/pngpread.c, src/other/libpng/pngread.c, src/other/libpng/pngrtran.c, src/other/libpng/pngrutil.c, src/other/libpng/pngset.c, src/other/libpng/pngtest.c, src/other/libpng/pngwrite.c, src/other/libpng/scripts/CMakeLists.txt, src/other/libpng/scripts/libpng-config-head.in, src/other/libpng/scripts/libpng.pc-configure.in, src/other/libpng/scripts/libpng.pc.in, src/other/libpng/scripts/makefile.32sunu, src/other/libpng/scripts/makefile.64sunu, src/other/libpng/scripts/makefile.aix, src/other/libpng/scripts/makefile.beos, src/other/libpng/scripts/makefile.cygwin, src/other/libpng/scripts/makefile.darwin, src/other/libpng/scripts/makefile.dec, src/other/libpng/scripts/makefile.elf, src/other/libpng/scripts/makefile.gcmmx, src/other/libpng/scripts/makefile.hp64, src/other/libpng/scripts/makefile.hpgcc, src/other/libpng/scripts/makefile.hpux, src/other/libpng/scripts/makefile.linux, src/other/libpng/scripts/makefile.mingw, src/other/libpng/scripts/makefile.ne12bsd, src/other/libpng/scripts/makefile.netbsd, src/other/libpng/scripts/makefile.nommx, src/other/libpng/scripts/makefile.openbsd, src/other/libpng/scripts/makefile.sco, src/other/libpng/scripts/makefile.sggcc, src/other/libpng/scripts/makefile.sgi, src/other/libpng/scripts/makefile.so9, src/other/libpng/scripts/makefile.solaris, src/other/libpng/scripts/makefile.solaris-x86, src/other/libpng/scripts/pngos2.def, src/other/libpng/scripts/pngw32.def: update libpng to 1.2.25 (coverity fixes) 2008-02-19 21:01 bob1961 * misc/win32-msvc8/tclsh/library/installTree.tcl: Mods to accomodate the location change of vfont. 2008-02-19 22:40 bob1961 * src/other/tkimg/pngtcl/pngtclDecls.h, src/other/tkimg/pngtcl/pngtclStubInit.c: No longer using png_read_destroy and png_write_destroy. 2008-02-20 02:25 brlcad * sh/any-png.sh, sh/cadbug.sh, sh/copyright.sh, sh/cray.sh, sh/elapsed.sh, sh/enumerate.sh, sh/facetall.sh, sh/linkroot.sh, sh/pixinfo.sh, sh/pixread.sh, sh/pixwrite.sh, sh/prior.sh, sh/sgisnap.sh, sh/show.sh, sh/tracker.sh, sh/ws.sh: update the indentation ws style of the shell scripts. 2008-02-20 02:30 brlcad * sh/indent.sh: ignore sh/ dir scripts. it gets almost as many wrong as it gets right due to quirky scriptisms. 2008-02-20 02:52 brlcad * sh/indent.sh: know what, just screw sh files. not worth it -- they don't generally matter. 2008-02-20 03:14 brlcad * bench/lgt.sh: ws indent style 2008-02-20 06:20 brlcad * include/opennurbs_ext.h: C++ mode 2008-02-20 06:21 brlcad * sh/footer.sh: C or C++ files should have the c-file-style set to stroustrup 2008-02-20 11:18 brlcad * Makefile.am: nifty, the top-level makefile will obey a DISTCHECK_CONFIGURE_FLAGS directive. make --enable-all one of the things it does during distcheck so that all files are included in the dist and the compilation of everything is tested. 2008-02-20 11:19 brlcad * include/conf/PATCH: bump to a .2 patch revision for ajem release testing (this is 7.12.0 prerelease testing) 2008-02-20 11:22 brlcad * include/conf/PATCH: make the ajem test rev unique, bump it up again 2008-02-20 11:23 brlcad * src/other/libpng/Makefile.am: extra dist mismatch 2008-02-20 11:33 d_rossberg * include/bu.h, include/config_win.h: this should solve the MS Visual Studio inline problem: define (the C) inline as the Microsoft specific __inline in config_win.h 2008-02-20 11:34 brlcad * src/other/iwidgets/Makefile.am, src/other/libpng/Makefile.am, src/other/libz/Makefile.am, src/other/tkimg/base/Makefile.am: few additional files detected missing from the dist 2008-02-20 18:09 starseeker * src/gtools/g_qa.1: Fix formatting for brlman command. 2008-02-21 13:36 bob1961 * include/brep.h: Having extern "C" around vmath.h and bu.h brings about the use of "C" linkage with templates when __cplusplus is defined. 2008-02-21 13:40 bob1961 * misc/win32-msvc8/bwish/bwish.vcproj: Added BRLCAD_DLL, HAVE_CONFIG_H and BRLCADBUILD. Removed IGNORE_CONFIG_H. 2008-02-21 13:40 bob1961 * misc/win32-msvc8/btclsh/btclsh.vcproj: Added BRLCAD_DLL, HAVE_CONFIG_H and BRLCADBUILD. Removed IGNORE_CONFIG_H. 2008-02-21 13:46 bob1961 * src/bwish/cadAppInit.c: Clean up the headers a bit. 2008-02-21 19:18 bob1961 * include/config_win.h: Define NOMINMAX before including windows.h to avoid clobbering std::numeric_limits<*>::min/max in C++ 2008-02-21 22:19 bob1961 * include/config_win.h: define fmax using __max instead of max 2008-02-21 22:20 bob1961 * src/bwish/cadAppInit.c: Include winsock2.h before any possible include of windows.h 2008-02-21 22:24 bob1961 * misc/win32-msvc8/asc2g/asc2g.vcproj, misc/win32-msvc8/brlcad/brlcad.sln, misc/win32-msvc8/librt/librt.vcproj, misc/win32-msvc8/libwdb/libwdb.vcproj: Mods to compile brep. 2008-02-21 22:29 bob1961 * src/librt/g_brep.cpp: Remove the ON_CLASS specifier (i.e. __declspec) for RT_MemoryArchive. 2008-02-22 00:19 brlcad * src/other/tcl/unix/Makefile.in: reapply the patch that installs libtcl.so and libtclstub.so, and conditionally installs the manual pages. 2008-02-22 00:20 brlcad * src/other/tk/unix/Makefile.in: reapply the patch that installs libtk.so and libtkstub.so, and conditionally installs the manual pages. 2008-02-22 14:38 bob1961 * misc/win32-msvc8/brlcad/brlcad.sln: Added breplicator. 2008-02-22 14:40 bob1961 * misc/win32-msvc8/breplicator, misc/win32-msvc8/breplicator/breplicator.vcproj, misc/win32-msvc8/opennurbs, misc/win32-msvc8/opennurbs/opennurbs.vcproj: Initial check-in. 2008-02-22 14:42 brlcad * src/conv/patch/rpatch.f: ws 2008-02-22 14:46 brlcad * src/java/mil/army/arl/brlcad/BrlcadGeometry.java, src/java/mil/army/arl/brlcad/BrlcadGeometryServer.java, src/java/mil/army/arl/brlcad/Partition.java: add local variables footer 2008-02-22 14:47 brlcad * sh/footer.sh, sh/header.sh: add support for .java files 2008-02-22 14:50 brlcad * src/adrt/scripts/makemovie.pl, src/archer/plugins/Wizards/tankwizardIA/TankWizardIA.tcl, src/archer/plugins/utility.tcl, src/archer/plugins/wizards.tcl, src/external/Cubit/g-sat.cxx, src/librt/raydebug.tcl, src/rttherm/ssampview.tcl, src/util/morphedit.tcl: ws style updates for non-C source files that have been processed with indent.sh, making indentation consistent 2008-02-22 15:42 brlcad * src/tclscripts/ami.tcl, src/tclscripts/ampi.tcl, ... lots of other files ..., src/tclscripts/vmath.tcl: needs some run-time testing, but apply the style changes to the tcl sources. indent.sh did surprisingly well at about 90% correctness though it was often confused by embedded #'s in the sources. seeing other examples, it looks like it's valid to quote the # (e.g. for upvar and #auto) and/or quote stringlists that have embeeded #'s but use {}'s. still needs testing. 2008-02-22 20:25 erikgreenwald * src/mged/cmd.c, src/mged/cmd.h, src/mged/rtif.c: collapse some duplicate functions 2008-02-23 19:25 brlcad * src/adrt/adrt.h, src/adrt/adrt_struct.h, src/adrt/bench/bench.c, ... lots of other files ..., src/adrt/slave/slave.c: style and ws update using the indent.sh emacs formatter script. make the sources consistent with header/footer blocks and our BSD KNF / K&R indent style. 2008-02-23 23:27 brlcad * src/anim/anim_cascade.c, src/anim/anim_fly.c, ... lots of other files ..., src/vdeck/vproc.c: major style and ws consistency cleanup. apply the automatic emacs-based indent.sh formatting to the sources so they consistently are formatted to our BSD KNF / K&R indentation style (see HACKING) using 4 char indents with tab stops at 8. 2008-02-23 23:30 brlcad * src/bwish/cadAppInit.c: add footer, indent 2008-02-23 23:31 brlcad * src/conv/g-adrt.c: footer, indent 2008-02-24 00:34 brlcad * src/libfft/ditsplitc.c, src/libfft/splitditc.c: format the sources being generated in a more readable style and include a description header 2008-02-24 00:45 brlcad * src/tab/script.l: include a header and footer 2008-02-24 00:52 brlcad * src/burst/grid.c: get rid of the SINGLE_PRECISION sections, format it up 2008-02-24 00:57 brlcad * src/lgt/do_options.c: more SINGLE_PRECISION removal and cleanup 2008-02-24 01:00 brlcad * src/proc-db/tube.c: cleanup 2008-02-25 04:36 brlcad * src/other/tk/unix/Makefile.in: need SHLIB_SUFFIX 2008-02-25 04:45 brlcad * src/other/tk/unix: 2008-02-25 04:46 brlcad * src/other/tcl/unix: 2008-02-25 05:03 brlcad * src/lgt/extern.h, src/lgt/fb.c, src/lgt/lgt.c: get rid of SGI_WINCLOSE_BUG. it's something libfb should be handling if it's necessary. 2008-02-25 10:04 d_rossberg * include/config_win.h: at least MSVC 6.0 needs the min/max templates (because min/max is not contained in the MSVC-STL because windef.h already defines it as a macro) 2008-02-25 13:41 brlcad * src/burst/burst.h, src/burst/fb.c: oop, more SGI_WINCLOSE_BUG and SINGLE_PRECISION mods that didn't get committed. 2008-02-25 14:56 brlcad * configure.ac, doc/deprecation.txt, include/config_win.h, include/orle.h, m4/epsilon.m4, src/libbu/htond.c, src/libbu/htonf.c, src/libdm/dm_obj.c: NATURAL_IEEE and REVERSE_IEEE are no longer used. instead, try to utilize the bu_byteorder() run-time checks now where possible or WORDS_BIGENDIAN where not feasible. while this doesn't change the behavior of htond and htonf, care needs to be taken for the floating point format not being what we expect. make the configure test warn more verbosely if we encounter something that may not be ieee 754 floating point format (which in turn might imply htond and/or htonf need modification, the other known floating types aren't even used). 2008-02-25 15:09 brlcad * src/other/tcl/generic/tclInt.h: shouldn't need the inline hack any longer 2008-02-25 15:53 brlcad * src/libpkg/pkg.c, src/libsysv/Makefile.am, src/libsysv/bsdselect.c: remove bsdselect, no longer needed or in use. 2008-02-25 19:40 bob1961 * src/libfb/if_ogl.c: Pick the first visual with the greatest depth. Mods to disallow use of DirectColor visuals. 2008-02-25 19:41 brlcad * src/conv/conv-vg2g.c, src/conv/g-stl.c, src/conv/g2asc.c, src/fb/fb-pix.c, src/fb/pix-fb.c, src/libfb/if_disk.c, src/libpkg/pkg.c, src/librt/dg_obj.c, src/librt/nirt.c, src/mged/ged.c, src/nirt/nirt.c, src/rt/main.c, src/util/bwmod.c, src/util/pix-bw.c, src/util/pixmatte.c: windows O_BINARY setmode cleanup 2008-02-25 21:11 erikgreenwald * src/conv/conv-vg2g.c: include fcntl.h so O_RDONLY is properly defined 2008-02-25 21:38 brlcad * include/sysv.h: no longer checking for bsdselect, no longer providing it or using it either 2008-02-25 23:08 bob1961 * include/config_win.h: Only define min/max if MSVC 6.0 and older. 2008-02-26 00:56 brlcad * misc/win32-msvc8/Makefile.am: need breplicator and opennurbs included in the dist 2008-02-26 22:37 bob1961 * src/tclscripts/archer/ArcherCore.tcl: Mods to update the raytrace control panel's size entry. Other mods to keep the raytrace control panel on top. 2008-02-26 22:38 bob1961 * misc/win32-msvc8/libtcl/libtcl.vcproj, misc/win32-msvc8/libtclcad/libtclcad.vcproj: Added inline definition. 2008-02-27 06:31 brlcad * include/fb.h: looks like unistd.h is actually not needed any more 2008-02-27 20:37 starseeker * doc/BRL-CAD.bib: Add technical report reference. 2008-02-28 08:49 brlcad * bench/pixcmp.c, include/Makefile.am, include/bio.h, ... lots of other files ..., src/vdeck/parsarg.c, src/vdeck/vdeck.c: Add a new bio.h private wrapper header for consistently defining native and standard I/O interfaces provided via stdio.h, unistd.h, io.h, and fcntl.h. remove the 300+ instances of a HAVE_UNISTD_H block throughout with a simple include of bio.h. this change was also motivated by the need to remove all of the header includes from the windows-specific config_win32.h (so that winsock2.h can be included before windows.h cleanly). of course, needs windows testing and perhaps a few file tweaks given how many files were modified. 2008-02-28 14:43 bob1961 * src/libfb/if_ogl.c: Finished the pixel selection feature. 2008-02-28 21:16 brlcad * NEWS: Bob fixed a bug in the X11 framebuffer interface where it would cause the entire X11 display to go black or not depending on whether the framebuffer window had focus or not. this issue was related to a problem with the visual type and colormapping where a bad visual was being selected. now it selects a truecolorvisual (instead of directcolor iirc). 2008-02-28 21:18 brlcad * NEWS: like the X framebuffer type, Bob added the same functionality to the ogl framebuffer where the user can query the color value underneath the cursor by right-clicking in the framebuffer window. 2008-02-29 01:20 brlcad * src/tclscripts/lib/Dm.tcl: gah, merge a fix made for 7.10.4 on the stable branch that didn't make its way back to the trunk. this fix makes the Dm class not blindly try to open :0 but instead checks what env(DISPLAY) is set to. 2008-02-29 21:05 starseeker * NEWS, src/librt/dg_obj.c, src/librt/wdb_obj.c, src/tclscripts/helplib.tcl, src/tclscripts/lib/Db.tcl, src/tclscripts/mged/help.tcl, src/tclscripts/mged/tree.tcl: Add -d option to MGED tree command for limiting depth of tree printing. 2008-03-02 08:22 brlcad * db/Makefile.am: keep getting parallel build problems where the exec test fails saying asc2g doesn't exist yet. try syncing the filesystem and if it does still fail, print what ls shows 2008-03-02 08:45 brlcad * db/Makefile.am: the more I think about it, the failure makes sense. the asc2g wrapper is kicking off a compilation which causes a failure when another proc tries to use the script mid-compilation. it is either deleted or is getting set -x during compilation. hopefully this prep phase will make it behave. 2008-03-02 08:57 brlcad * Makefile.am: add a distcheck and install check to make sure that the tclIndex and pkgIndex.tcl files aren't empty. there are some configurations generating empty files when the auto_paths are not set up correctly. the installs will seem to be just fine .. but are actually rather busted and will result in a variety of screwy run-time tcl failures when trying to use any of those routines. also make elapsed.sh use top_srcdir for the DATE file since it's a BUILT_SOURCES file. 2008-03-02 16:03 brlcad * TODO: bob, erik, and myself have all failed at reproducing the mged text flood hanging so the problem was either a ScrolledText widget Tk bug that was fixed with the upgrade to Tk 8.5 or it's sensitive to timing issues and isn't easy to provoke on fast machines. needs to be revisited on a slow machine, but that doesn't need to hold up the release. 2008-03-03 14:27 d_rossberg * include/bio.h, include/config_win.h: moved some windows.h related undefs from config_win.h to bio.h 2008-03-03 14:30 d_rossberg * include/cmd.h, src/libbu/image.c, src/libbu/kill.c, src/libbu/magic.c: include the new bio.h where necessary 2008-03-03 14:39 d_rossberg * src/librt/bezier_2d_isect.c, src/librt/bigE.c, ... lots of other files ..., src/libwdb/units.c, src/libwdb/vol.c, src/libwdb/wdb.c: unfortunately raytrace.h needs windows.h (on a Win32 platform) but bio.h has to be included right after the system includes, therefore it can not be included in raytrace.h (this would be to late) this are the changes for the core libraries only, there are certainly more in other libraries necessary 2008-03-04 15:16 d_rossberg * src/irprep/Makefile.am: put irdisp neatly under WITH_X11 condition 2008-03-04 18:46 bob1961 * include/raytrace.h: struct dg_obj and supporting cast has been moved to dg.h 2008-03-04 18:46 bob1961 * include/dg.h: Initial check-in. 2008-03-04 18:48 bob1961 * src/conv/asc-nmg.c, src/conv/asc2pix.c, src/conv/euclid_format.c, ... lots of other files ..., src/mged/vparse.c: Mods related to bio.h and the new dg.h wrt getting things compiled on Windows. 2008-03-04 18:53 bob1961 * include/raytrace.h: Move "drawable geometry" related function declarations to dg.h 2008-03-04 19:02 bob1961 * include/dg.h: Added function declarations. 2008-03-04 21:00 bob1961 * src/librt/opennurbs_ext.cpp: Move include for bio.h after common.h 2008-03-04 21:05 bob1961 * src/libfb/if_wgl.c: Move include for bio.h after common.h 2008-03-05 07:03 brlcad * include/cmd.h, include/dg.h: this will very-likely re-break the build on Windows, but bio.h is a private header and cannot be included in any public headers. (e.g. it shouldn't ever be installed.) 2008-03-05 07:13 brlcad * include/raytrace.h: why are these three ifree functions needed? if they're special, they should probably be documented with a comment. 2008-03-05 07:14 brlcad * include/raytrace.h: give it a name 'just in case' someone really does need these so they can be given a cppflag that will make it all better 2008-03-05 07:20 brlcad * regress/flawfinder.sh: add a regression test for bio.h to make sure nobody adds it to a public header down the road. 2008-03-05 09:36 brlcad * regress/flawfinder.sh: add another coding check for the common.h inclusion ordering. this makes sure common.h always comes first if/when listed in a file. alas the script isn't tweaked to work right if there are spaces in the path name, but then there are probably other scripts like that in regress too 2008-03-05 09:41 d_rossberg * src/libbu/bu_tcl.c, src/libbu/cmd.c, src/libbu/cmdhist.c, src/libbu/cmdhist_obj.c, src/libbu/observer.c, src/librt/vdraw.c, src/librt/view_obj.c: some additional bio.h includes to make the Windows build work again 2008-03-05 09:57 d_rossberg * src/other/openNURBS/Makefile.am: opennurbs_zlib.h needs to be installed because of the z_stream type 2008-03-05 15:14 erikgreenwald * Makefile.am: include/conf/DATE is in builddir, not srcdir 2008-03-05 15:34 erikgreenwald * src/librt/Makefile.am, src/other/libpng/Makefile.am: update dependancy info 2008-03-05 16:13 brlcad * include/conf/PATCH: heh, skipping 3 and 4, going right to 5 2008-03-05 16:15 bob1961 * src/libfb/fb_obj.c, src/libfb/tcl.c: Include bio.h for windows compilation. 2008-03-05 16:55 bob1961 * misc/win32-msvc8/librt/librt.vcproj: Define _RT_DECL_IFREE. This is needed by asc2g. 2008-03-05 16:56 bob1961 * misc/win32-msvc8/asc2g/asc2g.vcproj: Define _RT_DECL_IFREE. 2008-03-05 17:35 bob1961 * src/tclscripts/lib/Dm.tcl: Need to check for the existence of the display variable. 2008-03-05 17:39 bob1961 * src/util/rtwizard.bat: Update the version. There should be a way to do this dynamically. 2008-03-05 20:00 erikgreenwald * src/other/openNURBS/Makefile.am: add deps 2008-03-06 06:51 brlcad * NEWS, src/libfb/if_ogl.c: remove the protections for the 'm' mode in the opengl framebuffer as well as requesting indirect buffers, now always requesting direct and seeing what we get. my always doing 'multiple window mode', the code now (correctly) acquires and releases the opengl context for all of the drawing so that there aren't multithreaded crashes on mac and linux. bob and I tag-teamed tracking down the cause and a fix for this bug over a couple days. interestingly, the crash on linux only occurred after he 'fixed' the linux behavior to always use a direct color map (upon which it began consistently crashing like the Mac). 2008-03-06 06:56 brlcad * src/adrt/libutil/camera.c, src/bwish/cadAppInit.c, src/libtclcad/tclcad.c, src/mged/attach.c, src/mged/points/points_scan.l, src/util/xyz-pl.c: common.h always comes first, there's a regression test that checks for this now too 2008-03-06 14:58 erikgreenwald * src/vas4/Makefile.am: disable build (but leave in dist for this release) 2008-03-06 15:04 erikgreenwald * configure.ac: favor -lpthread to -pthread 2008-03-06 15:31 bob1961 * src/tclscripts/archer/Archer.tcl: Put back a few class variables (i.e. they fell out during the ArcherCore transition) and make a few plugin related methods public. 2008-03-06 15:37 brlcad * regress/flawfinder.sh: make sure we have a python interpreter before proceeding with the flawfinder test 2008-03-06 15:41 brlcad * TODO: the ogl framebuffer problem is crashed, though it still needs some cleanup. it's making way too many calls to get/release the context now so it's dog-slow .. need to try what bob had which was just protecting write() 2008-03-06 15:52 brlcad * src/libfb/if_ogl.c: woo hoo! the performance problems were simply because several of the glXMakeCurrent acquire/release pairings were .. wrong. rather, they're in the wrong place with respect to the logic sometimes only acquiring in a specific scope but then releasing in a different scope. made them all consistent and sure enough it all behaves spiffily now .. yay for the return of the ogl framebuffer. 2008-03-06 15:54 brlcad * src/libfb/if_ogl.c: remove dead code, just makes maintenance a bitch 2008-03-06 16:01 brlcad * src/libfb/if_wgl.c: more dead code removal 2008-03-06 16:07 bob1961 * src/tclscripts/lib/Display.tcl: Modify rt and rtedge methods to stop unnecessarily specifying parameters like framebuffer, width, height and aspect. 2008-03-06 16:11 brlcad * src/libfb/if_wgl.c: make the wgl interface do the same as the ogl interface, fixing the various incorrectly scoped wglMakeCurrent calls so that they actually pair up correctly with the logic. also, actually make them release the context (the code was wrong, they just kept it) so when parallel support is added to windows, it'll actually work. 2008-03-06 16:39 bob1961 * src/libfb/if_ogl.c: Try to get a direct context (speed). 2008-03-06 20:02 erikgreenwald * TODO: note "minimal system" testing 2008-03-06 21:19 erikgreenwald * src/librt/g_metaball.c: fluff up the isopotential method a bit... back to the old way, kinda 2008-03-06 21:42 bob1961 * src/tclscripts/geometree/GeometryBrowser.tcl: Remove extra set of double quotes in format statement of rgbToHex method. 2008-03-06 22:31 erikgreenwald * src/mged/typein.c: set a default "goo" value on metaballs 2008-03-07 02:16 brlcad * doc/deprecation.txt, include/wdb.h: deprecate mk_fastgen_comb() .. should just use mk_comb() or mk_lcomb() 2008-03-07 02:37 brlcad * include/raytrace.h: cleanup comments on LOS 2008-03-07 02:41 brlcad * NEWS, src/conv/fast4-g.c: make the regions being created by fastf-g default to setting the material code to 1 and LOS to 100 so that it actually matches the defaults that the modeler and librt use. this quick tweak implements a request from dwayne kregel (S2 SCR 918) .. which I just heard about it a few min ago. 2008-03-07 03:39 starseeker * db/terra.dsp: Update props on terra.dsp to avoid file size issues when running demo. 2008-03-07 05:03 louipc * misc/archlinux/PKGBUILD.in, misc/archlinux/brlcad.install.in, misc/archlinux/brlcad.sh.in: I'm making Arch Linux packaging scripts mostly static. It seems the most practical thing to do. 2008-03-07 05:06 louipc * misc/archlinux/PKGBUILD, misc/archlinux/PKGBUILD.in, misc/archlinux/brlcad.install, misc/archlinux/brlcad.install.in, misc/archlinux/brlcad.sh, misc/archlinux/brlcad.sh.in: Rename Arch Linux scripts, also fix md5sum in PKGBUILD 2008-03-07 08:03 louipc * configure.ac, misc/archlinux/Makefile.am: Makefile and configure changes for Arch Linux 2008-03-07 14:01 bob1961 * src/tclscripts/geometree/geometree.tcl: Remove dead auto_path code. 2008-03-07 18:58 bob1961 * src/tclscripts/geometree/GeometryBrowser.tcl: Add a few menu bindings to get things working a little better (i.e. right button to bring up menu, as before, and release to invoke menu item. 2008-03-07 21:45 bob1961 * src/tclscripts/geometree/GeometryBrowser.tcl: Added code to redraw drawn objects. 2008-03-07 21:50 bob1961 * include/conf/MINOR, include/conf/PATCH, misc/nsis/brlcad.nsi, misc/win32-msvc8/asc2g/asc2g.vcproj, src/archer/archer.bat, src/mged/mged.bat, src/util/rtwizard.bat: Update version to 7.12.0. Preparing for release. 2008-03-07 23:41 brlcad * src/other/tcl/unix/Makefile.in, src/other/tk/unix/Makefile.in: make sure the libraries have the exec bit set
