Revision: 78215
http://sourceforge.net/p/brlcad/code/78215
Author: starseeker
Date: 2021-02-08 03:43:10 +0000 (Mon, 08 Feb 2021)
Log Message:
-----------
Merge from trunk through r78208
Modified Paths:
--------------
brlcad/branches/extbuild/BUGS
brlcad/branches/extbuild/CMakeLists.txt
brlcad/branches/extbuild/ChangeLog
brlcad/branches/extbuild/NEWS
brlcad/branches/extbuild/README
brlcad/branches/extbuild/TODO
brlcad/branches/extbuild/doc/README.Windows
brlcad/branches/extbuild/doc/docbook/system/mann/CMakeLists.txt
brlcad/branches/extbuild/doc/docbook/system/mann/gdiff.xml
brlcad/branches/extbuild/doc/docbook/system/mann/quit.xml
brlcad/branches/extbuild/doc/docbook/system/mann/stat.xml
brlcad/branches/extbuild/include/conf/PATCH
brlcad/branches/extbuild/include/gcv/api.h
brlcad/branches/extbuild/include/ged/commands.h
brlcad/branches/extbuild/include/rt/db_diff.h
brlcad/branches/extbuild/misc/CMake/distcheck_in_src_dir.cmake.in
brlcad/branches/extbuild/misc/debian/changelog
brlcad/branches/extbuild/misc/macosx/Resources/ReadMe.rtfd/TXT.rtf
brlcad/branches/extbuild/misc/macosx/Resources/Welcome.rtfd/TXT.rtf
brlcad/branches/extbuild/misc/repowork/repowork.cpp
brlcad/branches/extbuild/regress/ged/CMakeLists.txt
brlcad/branches/extbuild/src/archer/CMakeLists.txt
brlcad/branches/extbuild/src/archer/archer.c
brlcad/branches/extbuild/src/conv/3dm/3dm-g.c
brlcad/branches/extbuild/src/conv/gcv/gcv.c
brlcad/branches/extbuild/src/conv/iges/g-iges.c
brlcad/branches/extbuild/src/libanalyze/analyze_private.h
brlcad/branches/extbuild/src/libanalyze/api.c
brlcad/branches/extbuild/src/libanalyze/raydiff.c
brlcad/branches/extbuild/src/libanalyze/util.cpp
brlcad/branches/extbuild/src/libgcv/gcv.c
brlcad/branches/extbuild/src/libgcv/plugins/asc/asc.cpp
brlcad/branches/extbuild/src/libgcv/plugins/decimate/decimate.c
brlcad/branches/extbuild/src/libgcv/plugins/fastgen4/fastgen4_read.c
brlcad/branches/extbuild/src/libgcv/plugins/fastgen4/fastgen4_write.cpp
brlcad/branches/extbuild/src/libgcv/plugins/gdal/gdal.cpp
brlcad/branches/extbuild/src/libgcv/plugins/obj/obj_read.c
brlcad/branches/extbuild/src/libgcv/plugins/obj/obj_write.c
brlcad/branches/extbuild/src/libgcv/plugins/rhino/rhino_read.cpp
brlcad/branches/extbuild/src/libgcv/plugins/stl/stl_read.c
brlcad/branches/extbuild/src/libgcv/plugins/stl/stl_write.c
brlcad/branches/extbuild/src/libgcv/plugins/vol/png_read.c
brlcad/branches/extbuild/src/libgcv/plugins/vrml/vrml_read/vrml_read.cpp
brlcad/branches/extbuild/src/libgcv/plugins/vrml/vrml_write.c
brlcad/branches/extbuild/src/libged/CMakeLists.txt
brlcad/branches/extbuild/src/libged/brep/CMakeLists.txt
brlcad/branches/extbuild/src/libged/exec_mapping.cpp
brlcad/branches/extbuild/src/libged/gdiff/gdiff.c
brlcad/branches/extbuild/src/libged/ged.c
brlcad/branches/extbuild/src/libged/ged_private.h
brlcad/branches/extbuild/src/libged/gqa/gqa.cpp
brlcad/branches/extbuild/src/libged/lint/lint.cpp
brlcad/branches/extbuild/src/libged/tree/tree.c
brlcad/branches/extbuild/src/libged/which/which.cpp
brlcad/branches/extbuild/src/libged/xpush/xpush.c
brlcad/branches/extbuild/src/libtclcad/commands.c
brlcad/branches/extbuild/src/mged/cmd.c
brlcad/branches/extbuild/src/mged/mged.c
brlcad/branches/extbuild/src/mged/setup.c
brlcad/branches/extbuild/src/other/ext/perplex/perplex/parser.y
brlcad/branches/extbuild/src/other/ext/perplex/perplex/perplex.cpp
brlcad/branches/extbuild/src/other/ext/perplex/perplex/perplex.h
brlcad/branches/extbuild/src/other/ext/perplex/perplex/perplex_template.c
brlcad/branches/extbuild/src/other/ext/perplex/perplex/scanner.re
brlcad/branches/extbuild/src/other/ext/perplex/perplex/token_type.h
brlcad/branches/extbuild/src/rtwizard/main.c
brlcad/branches/extbuild/src/tclscripts/mged/mview.tcl
brlcad/branches/extbuild/src/util/bwmod.c
brlcad/branches/extbuild/src/util/pixscale.c
Added Paths:
-----------
brlcad/branches/extbuild/doc/docbook/system/mann/npush.xml
brlcad/branches/extbuild/regress/ged/ppush_tests.g
brlcad/branches/extbuild/regress/ged/push.cpp
brlcad/branches/extbuild/regress/ged/push_tests.g
brlcad/branches/extbuild/regress/ged/regress-ged_push.cmake.in
brlcad/branches/extbuild/regress/ged/xpush_tests.g
brlcad/branches/extbuild/src/libged/npush/
Property Changed:
----------------
brlcad/branches/extbuild/
brlcad/branches/extbuild/NEWS
brlcad/branches/extbuild/doc/
brlcad/branches/extbuild/include/
brlcad/branches/extbuild/regress/
brlcad/branches/extbuild/src/other/
Index: brlcad/branches/extbuild
===================================================================
--- brlcad/branches/extbuild 2021-02-08 00:16:49 UTC (rev 78214)
+++ brlcad/branches/extbuild 2021-02-08 03:43:10 UTC (rev 78215)
Property changes on: brlcad/branches/extbuild
___________________________________________________________________
Modified: svn:mergeinfo
## -1,4 +1,4 ##
-/brlcad/branches/RELEASE:57439,57447-57860,69901-69913,70323-70333,71915-72242,72525-72534,72826-72858,74376-74454,74964-75140,75372-75685,76001-76451,76693-76768,77107-77132,77145-77155
+/brlcad/branches/RELEASE:57439,57447-57860,69901-69913,70323-70333,71915-72242,72525-72534,72826-72858,74376-74454,74964-75140,75372-75685,76001-76451,76693-76768,77107-77132,77145-77155,78190-78207
/brlcad/branches/analyze_cmd:76836-77221
/brlcad/branches/bioh:75720-75736,75740-75742,75860-75891,75894-75986,76088-76153,76354-76506,76577
/brlcad/branches/brep-debug:61373,61375,61404,61427,61429,61470,61544,61567,61576,61999,62018,62094,62098,62107,62117,62406,62416-62519,62521-62584,62593-62614,62623,62658,62660-62674,62681-62771,62876,62901,62907,62910,62925,62928,62931-63025,63027,63051,63054-63056,63069,63071-63073,63122,63160-63161,63165,63171,63184,63187,63189-63190,63193-63196,63200,63202,63205-63210,63213,63219-63225,63232-63233,63236,63238,63338,63350-63353,63481,63618,63669,64173-64174,64176-64177,64229-64233,64242,64244,64360-64362,65165,65245,65249,65334,65833-65834,66370-66375,66931-66932,66934,67012-67015,67018-67019,67021-67022,67406,67740,67746-67748,67950,67952,68144-68145,68636,68640-68643,68820,69081,69109,69168,69206,69289,69346,69460-69461,69582-69583,69719-69721,69857-69859,69927,69995-69996,70148-70149,70347-70349,70377,70526-70527,71006-71007,71009-71022,71046-71047,71049,71096-71100,77226-77229
## -11,4 +11,4 ##
/brlcad/branches/osg:62110-62113
/brlcad/branches/prep-cache:68236-68933
/brlcad/branches/tcltk86:68300-75257
-/brlcad/trunk:77547-78031
\ No newline at end of property
+/brlcad/trunk:77547-78208
\ No newline at end of property
Modified: brlcad/branches/extbuild/BUGS
===================================================================
--- brlcad/branches/extbuild/BUGS 2021-02-08 00:16:49 UTC (rev 78214)
+++ brlcad/branches/extbuild/BUGS 2021-02-08 03:43:10 UTC (rev 78215)
@@ -15,9 +15,6 @@
Recent Bugs
-----------
-* gqa command is generating invalid plot3 files in multithreaded
- mode.
-
* fbclear -c appears to hang indefinitely inside pkg_bwaitfor,
possible protocol mismatch or capability removed/changed in fbserv.
Modified: brlcad/branches/extbuild/CMakeLists.txt
===================================================================
--- brlcad/branches/extbuild/CMakeLists.txt 2021-02-08 00:16:49 UTC (rev
78214)
+++ brlcad/branches/extbuild/CMakeLists.txt 2021-02-08 03:43:10 UTC (rev
78215)
@@ -1464,6 +1464,7 @@
#BRLCAD_INCLUDE_FILE(windows.h HAVE_WINDOWS_H) # for QueryPerformanceCounter()
on Windows
if(MSVC)
CONFIG_H_APPEND(BRLCAD "#define HAVE_WINDOWS_H 1\n")
+ set(HAVE_WINDOWS_H 1)
endif(MSVC)
Modified: brlcad/branches/extbuild/ChangeLog
===================================================================
--- brlcad/branches/extbuild/ChangeLog 2021-02-08 00:16:49 UTC (rev 78214)
+++ brlcad/branches/extbuild/ChangeLog 2021-02-08 03:43:10 UTC (rev 78215)
@@ -1,18799 +1,4067 @@
-2020-09-14 08:45 brlcad
+2021-02-06 16:22 starseeker
- * [r77143] NEWS:
- include a brief summary writeup of highlights
+ * [r78203] include/conf/PATCH, misc/debian/changelog:
+ Update version numbers
-2020-09-14 07:20 brlcad
+2021-02-06 16:18 starseeker
- * [r77142] NEWS:
- document the util fixes in addition to the fb fixes for Windows.
- looks nominally like confirmed binary input mode fix dpix-pix,
- bwshrink, halftone, pixshrink, sun-pix, as well as the plot and
- brep
- plot commands. also possibly fixed by marking input as binary
- includes bw-a,
- bw-png, bw-ps, bwscale, mac-pix, pix-ppm, pix-ps, pix3filter,
- pixbgstrip, pixborder, pixembed, pixfilter, pixhalf, pixrot,
- pixscale,
- plot3-ps, ap-pix, bwcrop, bwhisteq, bwstat, pixcrop, pixdiff,
- pixhist,
- pixhist3d-plot3, pixhist3d, pixstat, plot3-asc, plot3stat,
- asc-plot3, bombardier, bw-pix, bwcrop, bwdif, pix-ppm, pix3filer,
- pixblend, pixclump, pixcount, pixcrop, pixdiff, pixinterp2x,
- pixmerge, pixmorph, pixuntile, plot3-dm, plot3rot
-
-
+ * [r78202] NEWS:
+ g-iges change is in principle user visible - fixed crash running
+ regression test on OSX.
-2020-09-14 05:08 brlcad
+2021-02-06 16:15 starseeker
- * [r77141] NEWS:
- cliff added asc v4/v5 import/export support to gcv, circa r76277
- and commits forward.
+ * [r78201] doc/docbook/system/mann/gdiff.xml:
+ gdiff -S also checks boolean operations
-2020-09-14 04:47 brlcad
+2021-02-06 16:13 starseeker
- * [r77140] src/libged/make/make.c:
- note that cline is intentionally not included
+ * [r78200] NEWS, doc/docbook/system/mann/gdiff.xml:
+ Document the -S option added to libged's gdiff command.
-2020-09-14 04:31 brlcad
+2021-02-06 16:05 starseeker
- * [r77139] NEWS:
- cliff added asc v4/v5 import support to gcv, circa r76277.
+ * [r78199] NEWS:
+ Note addition of -? option to which in r78052
-2020-09-14 04:25 brlcad
+2021-02-06 15:57 starseeker
- * [r77138] TODO:
- there's an article, but need a gcv manual page
+ * [r78198] NEWS:
+ Whoops, redundant item
-2020-09-14 03:31 starseeker
+2021-02-06 15:50 starseeker
- * [r77137] include/brep/defines.h,
- src/libgcv/plugins/rhino/rhino_read.cpp,
- src/librt/primitives/brep/brep.cpp,
- src/other/openNURBS/example_read/example_read.cpp,
- src/other/openNURBS/example_roundtrip/example_roundtrip.cpp,
- src/other/openNURBS/example_write/example_write.cpp,
- src/other/openNURBS/opennurbs_archive.cpp,
- src/other/openNURBS/opennurbs_archive.h,
- src/other/openNURBS/opennurbs_defines.cpp,
- src/other/openNURBS/opennurbs_defines.h,
- src/other/openNURBS/opennurbs_extensions.cpp,
- src/other/openNURBS/opennurbs_extensions.h,
- src/other/openNURBS/opennurbs_string.cpp,
- src/other/openNURBS/opennurbs_userdata.cpp,
- src/other/openNURBS/opennurbs_zlib.cpp:
- Just directly alter the problematic names in the openNURBS
- headers, so we don't have to try and work around it in ours.
+ * [r78197] NEWS:
+ Note fixes to libtclcad command I/O that were causing problems
+ (particularly on OSX, but almost certainly elsewhere as well.)
-2020-09-14 00:57 starseeker
+2021-02-06 15:41 starseeker
- * [r77136] NEWS:
- Technically user visible - document man page update for remrt in
- r76897
-
-2020-09-13 21:11 starseeker
-
- * [r77135] doc/docbook/system/mann/search.xml:
- Mention alphanum search output sorting in man page.
-
-2020-09-13 17:29 starseeker
-
- * [r77134] ., src/archer/CMakeLists.txt, src/brlman/CMakeLists.txt,
- src/bwish/CMakeLists.txt, src/libbu, src/libbu/parse.c,
- src/libtclcad/CMakeLists.txt, src/libtclcad/commands.c,
- src/mged/CMakeLists.txt, src/tclscripts/lib/RtImage.tcl:
- Pull RELEASE-only changes back into trunk, r77106 through r77132
-
-2020-09-13 17:24 starseeker
-
- * [r77132] misc/CMake/NSIS_silent.template.in:
- Update the silent NSIS template to match the main version.
-
-2020-09-13 15:58 starseeker
-
- * [r77130] misc/CMake/NSIS.template.in:
- See if we can manipulate the startup directory of the shortcuts
- via NSIS
-
-2020-09-13 15:47 starseeker
-
- * [r77129] NEWS:
- The 'nburst' refactor of the burst code is available. Mostly a
- general cleanup of the code, which was previously in an early
- coding style quite different from most of BRL-CAD. Although
- primarily intended for batch processing, it also provides a new
- command line interactive mode that replaces the early terminal
- menu system removed in 7.30.2.
-
-2020-09-13 15:27 starseeker
-
- * [r77128] NEWS:
- overlay and screengrab commands gained -F options that let them
- interact with the framebuffer, rather than the display manager
- scene. Useful for capturing and updating the contents of the
- embedded MGED framebuffer.
-
-2020-09-13 14:47 starseeker
-
- * [r77127] doc/docbook/system/man1/CMakeLists.txt,
- doc/docbook/system/man1/bolt.xml:
- bolt man page was added to MGED's category, but we still have the
- command line tool as well.
-
-2020-09-13 14:37 starseeker
-
- * [r77126] NEWS:
- whichair and whichid can now also search subsets of .g file using
- the --root specifier, in addition to the standard search of all
- objects.
-
-2020-09-13 14:34 starseeker
-
- * [r77125] doc/docbook/system/mann/whichair.xml:
- Update whichair man page for new capabilities.
-
-2020-09-13 14:04 starseeker
-
- * [r77124] NEWS:
- Added a new 'bot set' command to match 'bot get' - only supports
- mode and orientation right now, but these should eventually be
- expanded to other bot operations.
-
-2020-09-13 14:03 starseeker
-
- * [r77123] NEWS:
- We're now returning semantically meaningful strings for bot mode
- and orientation with the low level 'bot get' command - users
- don't know how to interpret the raw numbers.
-
-2020-09-13 13:56 starseeker
-
- * [r77122] NEWS:
- Circa r75774, added an extrude subcommand to the GED bot command
- for making ARB6 volumes based on BoT triangles. This is a simple
- extrude operation - it can't guarantee 'smooth' alignment between
- individual ARB6 extrusions at adjacent faces.
-
-2020-09-13 06:37 brlcad
-
- * [r77121] doc/README.other:
- remove the subdir
-
-2020-09-13 06:36 brlcad
-
- * [r77120] doc/CMakeLists.txt, doc/README.AIX, doc/README.IRIX:
- IRIX was retired in 2013 and we subsequently no longer compile
- there due to c++11. AIX is still quite active (particularly on
- HPC) but our README commentary is entirely obsolete.
-
-2020-09-13 06:31 brlcad
-
- * [r77119] doc/CMakeLists.txt, doc/README.OSCON-2014:
- remove the notes for a 2014 oscon presentation. exists in history
- and is inconsistent (i.e., confusing) with the platform READMEs
-
-2020-09-13 06:29 brlcad
-
- * [r77118] doc/CMakeLists.txt, doc/README.AIX, doc/README.BSD,
- doc/README.IRIX, doc/README.OSCON-2014, doc/README.Solaris,
- doc/README.VAX, doc/README.other/README.AIX,
- doc/README.other/README.BSD, doc/README.other/README.IRIX,
- doc/README.other/README.OSCON-2014,
- doc/README.other/README.Solaris, doc/README.other/README.VAX:
- moving platform-specific README's together. platforms should be
- presented equitably or can result in a dismissive perspective
- harmful in the long term. best practice and project ethos has
- always been to handle cross-platform variants on valid footing.
-
-2020-09-13 06:05 brlcad
-
- * [r77117] NEWS:
- whichid and whichair commands now let one specify a range of IDs.
- the changes were prototyped by nick in a script, and more
- recently adopted into libged by cliff
-
-2020-09-12 16:34 starseeker
-
- * [r77109] src/libtclcad/commands.c:
- Check for null before calling Tcl_Close
-
-2020-09-11 19:05 brlcad
-
- * [r77105] bench/run.sh:
- /dev/null is a special device and -f is only for regular files,
- so we really just want to check for existence here.
-
-2020-09-10 04:20 brlcad
-
- * [r77098] include/icv/io.h:
- clarify some of the icv_image_size() parameters.
-
-2020-09-10 00:41 brlcad
-
- * [r77096] TODO:
- search issue was fixed in r77076. it's all good.
-
-2020-09-09 22:50 starseeker
-
- * [r77095] CMakeLists.txt:
- One of the unit tests for bu_file failed on the Mac github runner
- using the system realpath. Might be the test isn't valid, but
- we're right before release - just put back the internal version
- for now.
-
-2020-09-09 20:33 brlcad
-
- * [r77094] TODO:
- search -exec appears to be completely non-functional. I'm getting
- no output. Build from a couple month back appears to be working,
- so maybe related to process/channel/io changes.
-
-2020-09-09 20:32 brlcad
-
- * [r77093] src/libged/TODO:
- elaborate thoughts on an object-awar gsort/tables/filter command.
-
-2020-09-09 13:23 brlcad
-
- * [r77092] doc/docbook/system/man3/CMakeLists.txt,
- doc/docbook/system/man3/libicv.xml:
- applied sf patch 554 (Added libicv man page in
- doc/docbook/system/man3/) from Sumagna Das
-
-2020-09-09 13:18 brlcad
-
- * [r77091] src/libgcv/plugins/vol/png_read.c:
- fix ws indent
-
-2020-09-09 03:55 brlcad
-
- * [r77090] bench/run.sh:
- support /dev/null and the NUL special file construct on Windows
- (for git bash and msys2-based scripting environments)
-
-2020-09-09 03:41 brlcad
-
- * [r77089] bench/run.sh:
- turn off path conversion in msys2 and in git's customized
- msys-based shell. likely getting /dev/null conversion for free,
- so fix for that is next
-
-2020-09-08 20:38 starseeker
-
- * [r77088] NEWS, doc/docbook/system/mann/search.xml,
- src/librt/search.c, src/librt/search.h:
- This came up again... forward port and enable the most basic of
- the search command size filters from r67646. This is just the
- dp->d_len object size filter, but it pairs with the ls options
- that can report size - with this in place, we can use ls -lS to
- get an overall report of object sizes and then search with -size
- and other filters to identify large or small subsets based on
- type. Particular use case was trying to find a large plate mode
- bot in the FAA Generic_Twin.g example: ls -lS to get an idea of
- size distributions and then search -type plate -size >50000 to
- find larger examples. Would be nice to be able to specify sorting
- the search output on size at some point as well...
-
-2020-09-08 20:27 starseeker
-
- * [r77086] TODO:
- Note a quirk of bot -V visualizations that needs attention at
- some point.
-
-2020-09-08 19:29 brlcad
-
- * [r77085] TODO:
- note the mged startup path issues, discussed with a few users.
- need to investigate and document the current behavior for
- command-line, double-click, drag-n-drop file opening, GUI menu
- opening, and GUI CLI opening.
-
-2020-09-08 17:46 brlcad
-
- * [r77083] NEWS:
- looks like we missed documenting a gci contribution from jeffrey
- liu that added support for reading EBM data from binunif database
- objects (previously only supported files)
-
-2020-09-08 17:40 brlcad
-
- * [r77082] NEWS:
- retroactively document all of ali haydar's summer 2019 work on
- annotations. not included, but it's worth noting that his changes
- were realized through additional effort and mentoring by daniel
- roßberg.
-
-2020-09-08 17:30 brlcad
-
- * [r77081] NEWS:
- rishabh has been working all summer on the opencl pipeline in
- brl-cad with several more primitives converted including hyp and
- a bug, with untested/unfinished work on pipe, vol, arbn, and
- cline.
-
-2020-09-08 17:28 brlcad
-
- * [r77080] NEWS:
- daniel actually applied patches from ali haydar affecting display
- of annotations on windows.
-
-2020-09-08 17:27 brlcad
-
- * [r77079] AUTHORS:
- credit 2020 gsoc student rishab suthar with his openCL gpgpu
- pipeline contributions
-
-2020-09-08 16:41 starseeker
-
- * [r77077] src/libged/bot/bot.cpp:
- Teach bot GED command to set orientation and mode, and report
- user-understandable strings for get in those cases rather than
- the raw number.
-
-2020-09-08 16:38 starseeker
-
- * [r77076] src/mged/mged.c:
- Whoops - need to set the interp callbacks here as well.
-
-2020-09-08 16:22 brlcad
-
- * [r77075] AUTHORS:
- credit sumagna das for his work implementing the beginnings of a
- new gcv plugin (png-to-vol).
-
-2020-09-08 15:29 starseeker
-
- * [r77074] include/tclcad.h, src/libtclcad/commands.c:
- Need separate channels for in/out/err (MGED on Windows was
- crashing trying to close due to callbacks for both stdout and
- stderr channel closing.)
-
-2020-09-06 14:33 starseeker
-
- * [r77073] src/libgcv/plugins/vol/png_read.c:
- Add header and footer (caught by regress-licenses)
-
-2020-09-06 06:26 brlcad
-
- * [r77072] src/libgcv/plugins/vol/png_read.c:
- add a bit more debugging to understand what's going on, if any
- are being called
-
-2020-09-06 06:15 brlcad
-
- * [r77071] src/libgcv/plugins/vol/png_read.c:
- vol plugin needs to export gcv_plugin_info as a callable function
-
-2020-09-06 06:14 brlcad
-
- * [r77070] src/libgcv/plugins/gdal/gdal.cpp:
- don't crash if we can't determine the unit type in the gdal
- plugin. still can't handle png and bombs out, but at least aborts
- slightly more gracefully.
-
-2020-09-06 05:50 brlcad
-
- * [r77069] src/libgcv/plugins/vol,
- src/libgcv/plugins/vol/CMakeLists.txt,
- src/libgcv/plugins/vol/png_read.c:
- apply zulip patch from Sumagna Das that stubs in an initial
- vol-from-png plugin. not yet functional, but compiles and exposes
- a few other issues in libgcv.
-
-2020-09-06 05:48 brlcad
-
- * [r77068] src/libgcv/plugins/gdal/gdal.cpp,
- src/libgcv/plugins/gdal/gdal_ll.cpp:
- fit comments and separate if expression from the true block for
- consistent readability
-
-2020-09-04 20:52 starseeker
-
- * [r77066] src/libged/put_comb/put_comb.c:
- Unused string
-
-2020-09-04 20:08 brlcad
-
- * [r77065] src/libged/put_comb/put_comb.c:
- per the usage, the caller can say 'No' and still provide the
- remaining 4 region args, it just won't do anything with them.
- don't make it a fatal condition if that happens, just continue,
- since the caller (as is the case in mged) may just be a script
- that passes defaults. avoids the caller needing to
- conditionalize. undoing the error avoids a failure editing
- non-region combs via the combination editor in mged.
-
-2020-09-04 18:35 starseeker
-
- * [r77064] TODO:
- OK, not a new issue - calling reopen with a read-only database is
- what's doing this.
-
-2020-09-04 15:43 starseeker
-
- * [r77063] TODO:
- It's that the database is read-only, not that it's on Windows.
-
-2020-09-04 15:41 starseeker
-
- * [r77062] CMakeLists.txt:
- See if this lets us use system realpath. If so, we may be able to
- yank the bsd fallback implementation.
-
-2020-09-04 15:30 starseeker
-
- * [r77061] include/bu/defines.h, include/common.h:
- More uniformly define MAXPATHLEN across the code.
+ * [r78196] NEWS:
+ Note work on librt timer code
- Tcl defines MAXPATHLEN, and there was a potential for different
- definitions between their headers and ours. Not certain, but that
- is
- one possible source of the strange realpath crash observed on
- Linux
- but not reproducible in isolation - if code with one definition
- of
- MAXPATHLEN created a buffer and then fed it to a library compiled
- using
- a different definition, those different assumptions are a source
- of
- possible trouble. I think that's what was happening with the
- whoami.c
- code after I switched it to using MAXPATHLEN. Previously that
- code
- called realpath directly rather than bu_file_realpath. After the
- MAXPATHLEN switch, those raw realpath calls triggered a segfault
- on
- Linux - most likely due to a non-PATH_MAX definition of
- MAXPATHLEN
- previously being defaulted to by the bu/defines.h definition.
+ This is borderline - certain timing outputs may change now
+ that we are making more of an effort to use the CPU timers
+ when the API says we are, but it's not clear that there will
+ be significant user visible implications.
- Move the MAXPATHLEN definition to common.h, which is always
- included
- before tcl.h in our code, and add a couple definitions from the
- Tcl
- headers to use as fallbacks. Would be nice if there is some way
- to
- systematically check for this kind of issue...
+ If timing answer change that will be potentially significant,
+ so going ahead and noting in NEWS to provide some visibility.
-2020-09-04 14:08 starseeker
+2021-02-06 15:33 starseeker
- * [r77060] src/libbu/whereami.c:
- realpath -> bu_file_realpath
+ * [r78195] NEWS:
+ User visible - Sean fixed example in the saveview man page.
-2020-09-04 13:46 starseeker
+2021-02-06 15:28 starseeker
- * [r77059] src/libbu/realpath_bsd.c, src/libbu/whereami.c:
- PATH_MAX->MAXPATHLEN
+ * [r78194] NEWS:
+ Note Sumagna Das's patch #556 adding in-database support for vol
+ primitive data.
-2020-09-04 13:39 starseeker
+2021-02-06 15:21 starseeker
- * [r77058] doc/README.Windows:
- WiX .msi is getting more useful - add some notes to
- README.Windows.
+ * [r78193] NEWS:
+ Note change of MGED Overlap GUI to using gchecker tool behind the
+ scenes for more reliable execution on Windows.
-2020-09-04 05:55 brlcad
+2021-02-06 15:03 starseeker
- * [r77057] src/libtclcad/auto_path.c, src/libtclcad/bn.c,
- src/libtclcad/bu.c, src/libtclcad/cmdhist.c,
- src/libtclcad/commands.c, src/libtclcad/dm.c,
- src/libtclcad/eval.c, src/libtclcad/fb.c, src/libtclcad/fbserv.c,
- src/libtclcad/global.c, src/libtclcad/init.c,
- src/libtclcad/mouse.c, src/libtclcad/polygons.c,
- src/libtclcad/rt.c, src/libtclcad/tclcad_private.h,
- src/libtclcad/view/faceplate.c, src/libtclcad/wrapper.c:
- update file header filenames. use ./ relative header path to
- denote local include files.
+ * [r78192] BUGS:
+ semaphore protection appears to fix the gqa invalid plot3 issue
-2020-09-04 05:35 brlcad
+2021-02-06 14:45 starseeker
- * [r77056] src/libtclcad/CMakeLists.txt, src/libtclcad/commands.c,
- src/libtclcad/eval.c, src/libtclcad/global.c,
- src/libtclcad/init.c, src/libtclcad/mouse.c,
- src/libtclcad/polygons.c, src/libtclcad/tclcad_eval.c,
- src/libtclcad/tclcad_global.c, src/libtclcad/tclcad_init.c,
- src/libtclcad/tclcad_mouse.c, src/libtclcad/tclcad_obj.c,
- src/libtclcad/tclcad_obj_wrapper.c,
- src/libtclcad/tclcad_polygons.c, src/libtclcad/wrapper.c:
- rename most tclcad files to consistently not use a filename
- prefix. updating contents separately.
+ * [r78191] TODO:
+ quick check of dbconcat m35 into ktank seems to work?
-2020-09-04 04:34 brlcad
+2021-02-06 14:41 starseeker
- * [r77055] include/common.h:
- go ahead and put the snippets into code/endcode blocks to make
- their callout more explicit and obvious.
+ * [r78190] ChangeLog:
+ Update ChangeLog with commits through 78187.
-2020-09-04 04:14 brlcad
+2021-02-05 23:51 starseeker
+ * [r78187] **many files**:
+ Update copyright statements to year 2021 per sh/copyright.sh script
- * [r77054] include/bn/anim.h, include/bn/complex.h,
- include/bn/noise.h, include/bn/poly.h, include/bn/rand.h,
- include/bu/color.h, include/bu/column.h, include/bu/endian.h,
- include/bu/tbl.h:
- doxygen / comment cleanup
+2021-02-05 20:43 starseeker
-2020-09-04 04:13 brlcad
+ * [r78186] misc/CMake/distcheck_in_src_dir.cmake.in:
+ The in-src-dir distcheck has the same issue as the others -
+ ignore the .ninja_log file as there is nothing we can do about it
+ - ninja is being run to do the build steps, including cleanup,
+ and as long as ninja is running the log file is there.
- * [r77053] include/bu/app.h, include/bu/glob.h:
- convert define and enum comments for doxygen
+2021-02-05 19:42 starseeker
-2020-09-04 04:09 brlcad
+ * [r78185] NEWS:
+ Still testing, but unless proven otherwise gqa plotting in
+ multithreaded mode works now.
- * [r77052] include/bn/plane.h:
- document the struct elements where already documented
+2021-02-05 18:49 starseeker
-2020-09-04 04:01 brlcad
+ * [r78184] TODO, src/gtools/tests/CMakeLists.txt,
+ src/libanalyze/analyze_private.h, src/libanalyze/api.c,
+ src/libged/gqa/gqa.cpp:
+ Add specific semaphore protection for GQA plot I/O.
- * [r77051] include/bn/dvec.h:
- put the cond/endcond doxygen markers in doxy blocks
+2021-02-05 16:27 starseeker
-2020-09-04 03:06 starseeker
+ * [r78183] src/conv/iges/g-iges.c:
+ Avoid duplicating call to nmg_kr() in g-iges
+
+ OSX is getting a crash, and valgrind on Linux reports an issue
+ with db_free_tree called from g-iges.c:646 getting an invalid
+ read. The comment with that db_free_tree line indicates
+ db_free_tree does an nmg_kr call, so it looks like this prior
+ call in the same routine was interfering.
- * [r77050] TODO:
- Only on Windows, getting db_dircheck reports of duplicate objects
- on rtwizard startup.
+2021-02-05 01:07 starseeker
-2020-09-04 01:57 starseeker
+ * [r78182] src/other/CMakeLists.txt:
+ Getting some leftover files on the Mac
- * [r77049] src/mged/CMakeLists.txt:
- Typo
+2021-02-04 21:39 starseeker
-2020-09-04 01:43 starseeker
+ * [r78181] regress/ged/CMakeLists.txt, regress/ged/ppush_tests.g,
+ regress/ged/push.cpp, regress/ged/push_tests.g,
+ regress/ged/regress-ged_push.cmake.in, regress/ged/xpush_tests.g,
+ src/libged/tests/CMakeLists.txt, src/libged/tests/ppush_tests.g,
+ src/libged/tests/push.cpp, src/libged/tests/push_tests.g,
+ src/libged/tests/xpush_tests.g:
+ Modify ged push tests to be regression tests, not unit tests.
- * [r77048] src/archer/CMakeLists.txt, src/archer/archer.ico,
- src/archer/archer.rc, src/mged/CMakeLists.txt, src/mged/mged.ico,
- src/mged/mged.rc, src/rtwizard/CMakeLists.txt,
- src/rtwizard/rtwizard.ico, src/rtwizard/rtwizard.rc:
- If I'm understanding this correctly, the WiX msi installer relies
- on the application to provide the icon, unlike NSIS. Do what
- wish.exe does and try specifying rc files that identify the
- icons.
+2021-02-04 20:46 starseeker
-2020-09-03 22:42 starseeker
+ * [r78180] src/libged/tests/CMakeLists.txt:
+ Linux I/O isn't reliably placing the file for this test...
- * [r77047] CMakeLists.txt:
- Note potentially relevant WiX example in CMake itself.
+2021-02-04 20:14 starseeker
-2020-09-03 22:08 starseeker
+ * [r78179] src/libanalyze/util.cpp:
+ initialize
- * [r77046] src/archer/CMakeLists.txt, src/mged/CMakeLists.txt,
- src/rtwizard/CMakeLists.txt:
- Shouldn't hardcode 'bin' string
+2021-02-04 19:53 starseeker
-2020-09-03 22:06 starseeker
+ * [r78178] src/gtools/tests/CMakeLists.txt:
+ Comment out the multithreaded gqa ovlp test temporarily, so we
+ can do distcheck testing.
- * [r77045] CMakeLists.txt, src/archer/CMakeLists.txt,
- src/mged/CMakeLists.txt, src/rtwizard/CMakeLists.txt:
- Try setting the CPACK install properties.
+2021-02-04 18:38 starseeker
-2020-09-03 21:49 starseeker
+ * [r78177] TODO:
+ We identified which change caused the regression, but not yet
+ whether the problem is with the new semaphore system as used with
+ gqa or with some other issue that was simply being masked by the
+ old semaphore system. Workaround until this is fixed is to use
+ -P1 to generate plot files. Bump down to next patch release.
- * [r77044] CMakeLists.txt, misc/CMake/CMakeLists.txt,
- misc/CMake/WIX.template.in:
- CPACK_START_MENU_SHORTCUTS may let us get the menu set up without
- a custom template.
+2021-02-04 18:35 starseeker
-2020-09-03 19:49 starseeker
+ * [r78176] TODO:
+ Looked at gdiff issue - as near as I can tell resolution would be
+ a significant implementation change either in gdiff or librt's
+ lower level bookkeeping. keep+dbconcat offers a workaround in the
+ interim.
- * [r77043] misc/CMake/CMakeLists.txt:
- correct filename
+2021-02-04 17:16 starseeker
-2020-09-03 19:12 starseeker
+ * [r78175] src/libged/npush/npush.cpp:
+ Add a note addressing tor's issue with non uniform matrices and
+ why we're handling it as we are. (Duplicating push's behavior and
+ letting the low level primitive routines deal with it, and not
+ considering a failure to apply such a matrix a push failure.)
- * [r77042] CMakeLists.txt, misc/wix/brlcad_banner.bmp,
- misc/wix/brlcad_dialog.bmp:
- Tweak WiX images.
+2021-02-04 16:28 starseeker
-2020-09-03 17:46 starseeker
+ * [r78174] src/libged/tests/CMakeLists.txt:
+ Add ctest runs for the new npush tests.
- * [r77041] misc/CMake/FindWix.cmake:
- Remove debug message
+2021-02-04 16:26 starseeker
-2020-09-03 17:36 starseeker
+ * [r78173] src/libged/exec_mapping.cpp:
+ We don't know what calling code may or may not do with their argv
+ array and contents - they own it. Restore the original state of
+ the argv[0] entry before returning.
- * [r77040] CMakeLists.txt, misc/CMake/CMakeLists.txt,
- misc/CMake/WIX.template.in:
- See if we can point to the CMake standard built-in template and
- get the same installer as leaving the template unset.
+2021-02-04 15:23 starseeker
-2020-09-03 17:22 starseeker
+ * [r78172] src/libanalyze/raydiff.c, src/libanalyze/util.cpp,
+ src/libged/tests/push.cpp:
+ valgrind caught a few memory leaks in libanalyze
- * [r77039] misc/CMake/FindWix.cmake:
- Adjust path hunting for WiX
+2021-02-04 15:06 starseeker
-2020-09-03 17:09 starseeker
+ * [r78171] src/libged/npush/npush.cpp:
+ Make sure tidn and oidn are always correctly initialized.
- * [r77038] CMakeLists.txt, misc/CMake/CMakeLists.txt,
- misc/CMake/FindNSIS.cmake, misc/CMake/FindWix.cmake:
- Untested as yet, but see if we can key CPack setup on Windows off
- of what installer generators we have available.
+2021-02-03 20:35 starseeker
-2020-09-03 13:52 starseeker
+ * [r78170] misc/repowork/repowork.cpp:
+ Add ability to update svn:revision labels in Git commit messages
- * [r77037] misc/tools/CMakeLists.txt, misc/tools/inactvhdrs,
- misc/tools/inactvhdrs/inactvhdrs.cpp:
- Toss in a quick utility to try and detect headers in source
- hierarchies that aren't #included by any of the local files.
+2021-02-02 21:48 starseeker
-2020-09-03 13:47 starseeker
+ * [r78169] src/libged/npush/npush.cpp:
+ Use bn_mat_ck to validate the matrix being applied to a solid
+ object.
- * [r77036] src/libgcv/tests/CMakeLists.txt:
- Missing CMakeLists.txt file
+2021-02-02 17:09 starseeker
-2020-09-03 07:13 brlcad
+ * [r78168] src/libged/npush/npush.cpp:
+ Incorporate more full path printing into logging
- * [r77035] src/libgcv/CMakeLists.txt, src/libgcv/test_bottess.c,
- src/libgcv/tests, src/libgcv/tests/test_bottess.c:
- move gcv's only unit test into a subdir like our other libs.
+2021-02-02 15:27 starseeker
-2020-09-03 07:04 brlcad
+ * [r78167] src/libged/npush/npush.cpp:
+ Pass the full path for the tree update walk as well for better
+ reporting.
- * [r77034] include/gcv/api.h:
- oof, probably shouldn't assume it's a model mime type.
+2021-02-02 15:11 starseeker
-2020-09-02 18:08 starseeker
+ * [r78166] src/libged/npush/npush.cpp:
+ Track full paths in the push_walk step.
- * [r77033] CMakeLists.txt:
- Note user report that the WiX msi installer has an ability to
- install silently. Need to explore what it would take to match
- features with the NSIS .exe installer.
+2021-02-02 14:50 starseeker
-2020-09-01 19:30 starseeker
+ * [r78165] src/libged/npush/npush.cpp:
+ report full path for validation failures to provide more context
- * [r77031] include/bn/dvec.h, include/ged.h,
- include/ged/commands.h, include/rt/calc.h:
- Clear the remainder of the doxygen warnings
+2021-02-02 14:30 starseeker
-2020-09-01 19:17 starseeker
+ * [r78164] src/libged/tests/push.cpp:
+ reorganize the test reporting
- * [r77030] include/bu/vls.h, include/wdb.h:
- Clear some more doxygen warnings
+2021-02-02 13:48 starseeker
-2020-09-01 19:09 starseeker
+ * [r78163] src/libged/npush/npush.cpp:
+ We're already catching cyclic paths with another check - update
+ message.
- * [r77029] include/bu/file.h:
- Adjust comment for doxygen
+2021-02-01 18:55 starseeker
-2020-09-01 19:05 starseeker
+ * [r78162] src/libged/npush/npush.cpp:
+ Verify tops object set hasn't changed.
- * [r77028] include/brep/curvetree.h, include/brep/util.h:
- Tell doxygen not to worry about these includes
+2021-02-01 17:10 starseeker
-2020-09-01 18:45 starseeker
+ * [r78161] src/libged/tests/push.cpp:
+ Check the _ext top levels to verify the locality (or not) of push
+ changes.
- * [r77027] misc/doxygen/libbu.dox:
- Add group for bu/process.h
+2021-02-01 16:55 starseeker
-2020-09-01 18:36 starseeker
+ * [r78160] src/libged/npush/npush.cpp:
+ Ah. With the -L flag we're getting a case where checking
+ tree_altered isn't enough - check for an iname as well.
- * [r77026] misc/doxygen/Doxyfile.in:
- Increase DOT_GRAPH_MAX_NODES to avoid warnings about too many
- notes.
+2021-02-01 15:59 starseeker
-2020-09-01 18:28 starseeker
+ * [r78159] src/libged/npush/npush.cpp:
+ Sort by matrices first, then check apply_solid
- * [r77025] include/common.h:
- Indent for doxygen
+2021-02-01 15:25 starseeker
-2020-09-01 18:04 brlcad
+ * [r78158] src/libged/npush/npush.cpp:
+ Add a final check to spot missing objects added to pushed trees.
- * [r77024] include/bn/vectfont.h, include/bn/vlist.h:
- delete extra doxygen closure
+2021-02-01 14:15 starseeker
-2020-09-01 18:01 brlcad
+ * [r78157] src/libged/npush/npush.cpp:
+ Turn on the halting criteria for the -r flag (matrices above
+ regions.)
- * [r77023] include/bu/parallel.h:
- ditto, use addtogroup instead of ingroup for consistency.
+2021-01-31 19:35 starseeker
-2020-09-01 17:59 brlcad
+ * [r78156] src/libged/npush/npush.cpp:
+ Fix depth setting in tree_update walk, set error condition if we
+ can't find what we're looking for, depth limit the update walk if
+ a limit is set.
- * [r77022] include/bu/process.h:
- use addtogroup instead of ingroup, consistent with the rest and
- fixes an unbalanced grouping error.
+2021-01-30 23:50 starseeker
-2020-09-01 17:51 brlcad
+ * [r78155] src/libged/npush/npush.cpp:
+ Make a note - depth limits (-d, -r, etc.) aren't working yet.
- * [r77021] include/bu/dylib.h:
- missing @ sign
+2021-01-30 19:14 starseeker
-2020-09-01 17:47 starseeker
+ * [r78154] src/libged/tests/push.cpp:
+ Use the -S option to gdiff instead of the custom tree walker in
+ the test program.
- * [r77020] CHANGES, TODO:
- Per discussion with Sean, most of these don't qualify as needing
- the deprecation process.
+2021-01-30 15:23 starseeker
-2020-09-01 17:47 brlcad
+ * [r78153] src/libged/gdiff/gdiff.c:
+ Rough in a version of the "tree diff" capability under the gdiff
+ command.
- * [r77019] misc/doxygen/Doxyfile.in, misc/tools/dom2dox/Doxyfile:
- PERL_PATH and MSCGEN_PATH are deprecated, so remove them. we
- weren't using them anyways.
+2021-01-30 15:16 starseeker
-2020-09-01 17:42 starseeker
+ * [r78152] include/rt/db_diff.h:
+ include avs, type is used by db_diff
- * [r77018] TODO:
- This is looking quite tricky - did some experiments in swrast
- branch, but no quick fix popping out so far - bump down.
+2021-01-30 14:08 starseeker
-2020-09-01 15:20 starseeker
+ * [r78151] src/libged/npush/npush.cpp:
+ Dry run only controls writing the internals - all other steps
+ should be active.
- * [r77017] CHANGES:
- List some of the more obscure MGED menu entries for deprecation.
+2021-01-29 22:17 starseeker
-2020-09-01 13:49 starseeker
+ * [r78150] src/libged/tests/push.cpp:
+ more notes
- * [r77014] include/ged.h, include/ged/commands.h,
- include/rt/geom.h, include/rt/primitives/annot.h,
- misc/doxygen/libged.dox:
- More doxygen adjustments
+2021-01-29 22:15 starseeker
-2020-09-01 13:27 starseeker
+ * [r78149] src/libged/tests/push.cpp:
+ TODO - see if the cross-checking ability of this program can be
+ built in to npush itself - have npush with -V enabled call keep
+ to make a temp copy of the tree, process, then dbconcat back in
+ with a prefix, do the checks, then killtree -f the dbconcated
+ copy.
- * [r77013] include/bg/aabb_ray.h, include/bg/chull.h,
- include/bg/lseg.h, include/bg/obr.h, include/bg/pointgen.h,
- include/bg/polygon.h, include/bg/polygon_types.h,
- include/bg/spsr.h, include/bg/tri_pt.h, include/bg/tri_ray.h,
- include/bg/tri_tri.h, include/bg/trimesh.h,
- misc/doxygen/libbg.dox:
- Organize libbg doxygen comments
+2021-01-29 22:04 starseeker
-2020-09-01 12:59 starseeker
+ * [r78148] src/libged/npush/npush.cpp, src/libged/tests/push.cpp:
+ Looking better to keep the current processing than trying to
+ consolidate. Rename to reflect final purpose of various codes.
+ With a test program in place, next step will be to shake down the
+ various options.
- * [r77012] misc/doxygen/libbn.dox:
- Categorize for doxygen
+2021-01-29 20:49 starseeker
-2020-09-01 12:41 starseeker
+ * [r78147] src/libged/tests/push.cpp:
+ Be a bit more verbose with messages.
- * [r77011] include/bu/app.h, include/bu/glob.h,
- include/bu/interrupt.h, include/bu/mime.h,
- misc/doxygen/libbu.dox:
- Add some doxygen adjustments.
+2021-01-29 20:31 starseeker
-2020-09-01 07:21 brlcad
+ * [r78146] src/libged/tests/push.cpp:
+ need per-call argv copy for npush
- * [r77010] NEWS:
- note that our bundld libpng and zlib libraries now compile with
- API prefixes so as not to clash with system versions. the
- libraries are also installed with a _brl suffix to avoid
- accidentally matching a system library.
+2021-01-29 20:14 starseeker
-2020-09-01 06:55 brlcad
+ * [r78145] src/libged/tests/push.cpp:
+ Ray diffing isn't enough - also need to check for matrix
+ differences in the tree.
- * [r77009] NEWS:
- oof, document the biggest change in this release, tcl/tk is
- upgraded
+2021-01-29 18:59 starseeker
-2020-09-01 06:51 brlcad
+ * [r78144] src/libged/tests/push.cpp:
+ Implement concat call
- * [r77008] NEWS:
- posthumously document when the mged.bat script was killed
+2021-01-29 14:58 starseeker
-2020-09-01 06:23 brlcad
+ * [r78143] src/libged/tests/CMakeLists.txt,
+ src/libged/tests/push.cpp:
+ Start rigging a program to do testing of push behaviors.
- * [r77007] NEWS:
- improved nurbs ray tracing performance back in r74817 by
- eliminating unnecessary computation. chould be minimally or not
- evident in production builds, but likely evident as no warnings
- were issued
+2021-01-29 00:23 starseeker
-2020-09-01 06:01 brlcad
+ * [r78142] src/libged/tests/CMakeLists.txt,
+ src/libged/tests/ppush_tests.g, src/libged/tests/xpush_tests.g:
+ Add control .g files with the results of push and xpush
- * [r77006] src/libged/typein/typein.c:
- stray n
+2021-01-27 21:20 starseeker
-2020-09-01 06:00 brlcad
+ * [r78141] src/libged/npush/npush.cpp:
+ Note next step
- * [r77005] NEWS:
- document the util fixes in addition to the fb fixes for Windows.
- look nominally like dpix-pix, bwshrink, halftone, pixshrink as
- well as the plot and brep plot commands
+2021-01-27 20:30 starseeker
-2020-09-01 05:52 brlcad
+ * [r78140] src/libged/npush/npush.cpp:
+ Same deal with solids - can't write until we've staged
+ everything.
- * [r77004] NEWS:
- vdeck should handle windows text files now
+2021-01-27 20:15 starseeker
-2020-09-01 05:36 brlcad
+ * [r78139] src/libged/npush/npush.cpp:
+ Fix inverted check for identity matrix.
- * [r77003] NEWS:
- dm command is published, so documenting the change. sports a new
- type subcommand (replaces set dmtype
+2021-01-27 20:06 starseeker
-2020-09-01 05:35 brlcad
+ * [r78138] src/libanalyze/raydiff.c:
+ Free rti before resources
- * [r77002] src/tclscripts/mged/help.tcl:
- document the three subcommands to dm. bad usage as synopsis is
- indeterminant, and can't be resolved because all three are
- different.
+2021-01-27 19:05 starseeker
-2020-09-01 05:14 brlcad
+ * [r78137] src/libged/npush/npush.cpp:
+ Quirk in sorting was preventing lookup from finding correct
+ instance.
- * [r77001] NEWS:
- cliff removed the Display Manager menu options and associated
- options for switching display managers
+2021-01-27 19:03 starseeker
-2020-08-31 08:06 brlcad
+ * [r78136] src/libged/npush/npush.cpp:
+ Fix matrix sorting, only look at solid flag if dp isn't a comb
- * [r76994] src/libdm/glx/dm-ogl.c, src/libdm/tk/dm-tk.c:
- unwrap
+2021-01-27 18:25 starseeker
-2020-08-31 08:00 brlcad
+ * [r78135] src/libged/npush/npush.cpp:
+ still got a problem somewhere in the walking
- * [r76993] src/librt/db5_scan.c:
- document the claim on db5_diradd's previously unused clientdata.
- if it's set to one, the object will now be flagged as inmem.
+2021-01-27 17:53 starseeker
-2020-08-31 07:57 brlcad
+ * [r78134] src/libged/tests/push_tests.g:
+ Use smaller spheres for tests.
- * [r76992] src/librt/db_lookup.c:
- unclear and undocumented why db_lookup is explicitly turning off
- the inmem flag, so leave a note to investigate.
+2021-01-27 17:43 starseeker
-2020-08-31 05:18 brlcad
+ * [r78133] src/libged/npush/npush.cpp:
+ Do write operations on a separate copy of the tree. This may not
+ be necessary, but until I can prove overwriting data mid-walk can
+ never cause a problem go with the safer option.
- * [r76991] NEWS:
- rtweight now reports just once on the first material ID with no
- density file entry, and doesn't report on those unset.
+2021-01-27 03:02 starseeker
-2020-08-30 18:41 brlcad
+ * [r78132] src/libged/npush/npush.cpp:
+ Closer - object names in tree look more reasonable.
- * [r76990] TODO:
- the libs exist. conversion is ongoing and non-specific
+2021-01-27 02:44 starseeker
-2020-08-30 18:40 brlcad
+ * [r78131] src/libged/npush/npush.cpp:
+ Tree walk still isn't right...
- * [r76989] NEWS:
- adapting weights for modern times, modes for pix-bw
+2021-01-26 23:41 starseeker
-2020-08-30 18:39 brlcad
+ * [r78130] src/libged/npush/npush.cpp:
+ I was afraid of that... can't edit and walk the tree
+ simultaneously. Need to rethink.
- * [r76988] doc/docbook/system/man1/pix-bw.xml:
- document the weights for all the options including the new
- conversion modes
+2021-01-26 22:17 starseeker
-2020-08-30 18:38 brlcad
+ * [r78129] src/libged/npush/npush.cpp:
+ Start working on the actual I/O logic
- * [r76987] src/util/pix-bw.c:
- use more precision to more closely match the standard for ntsc.
- add additional conversion modes for hdtv, hdr, and the default
- simple averaging as an explicit mode. no consideration of linear
- vs non-linear.
+2021-01-26 20:25 starseeker
-2020-08-27 20:56 starseeker
+ * [r78128] src/libged/npush/npush.cpp:
+ checkpoint
- * [r76981] src/qged/TODO.qt:
- Note possible resources for qged console widget improvements.
+2021-01-26 19:32 starseeker
-2020-08-27 15:43 starseeker
+ * [r78127] src/libged/npush/npush.cpp:
+ Adjust walking code
- * [r76979] src/other/CMakeLists.txt:
- Looks like the Creo plugin needs our local versions of these
- headers on Windows to build. See if we can override the
- suppression of the 3rd party header installs in the specific
- cases where it is necessary.
+2021-01-26 18:59 starseeker
-2020-08-27 14:50 starseeker
+ * [r78126] src/libged/npush/npush.cpp:
+ No need to pass dbip around if it is also in the state container.
- * [r76978] src/external/Creo/CMakeLists.txt:
- Couple more naming adjustments for 3rd party libs
+2021-01-26 16:47 starseeker
-2020-08-27 14:44 starseeker
+ * [r78125] doc/docbook/system/mann/stat.xml:
+ typo
- * [r76977] src/external/Creo/CMakeLists.txt:
- fix comment
+2021-01-26 14:16 starseeker
-2020-08-27 13:54 starseeker
+ * [r78124] src/libged/npush/npush.cpp:
+ checkpoint
- * [r76976] src/external/Creo/CMakeLists.txt:
- Looks like not all versions use MDIR - look for it, but try to
- proceed even if it isn't found.
+2021-01-25 21:36 starseeker
-2020-08-27 12:25 starseeker
+ * [r78123] src/libged/npush/npush.cpp:
+ Note if we're applying matrix to the solid parameters
- * [r76975] src/external/Creo/CMakeLists.txt:
- The adjustments to the src/other build logic have changed the lib
- names - quick stab at updating them, but need to test.
+2021-01-25 21:23 starseeker
-2020-08-27 12:17 starseeker
+ * [r78122] src/libged/npush/npush.cpp:
+ checkpoint
- * [r76974] src/libdm/tests/CMakeLists.txt,
- src/libdm/tests/rendered_image.cpp:
- Experiment is in swrast branch.
+2021-01-25 21:08 starseeker
-2020-08-26 23:54 starseeker
+ * [r78121] src/libged/npush/npush.cpp:
+ Various adjustments for shallower nesting of code.
- * [r76971] src/libdm/tests/CMakeLists.txt:
- Whoops - disable, then commit
+2021-01-25 20:43 starseeker
-2020-08-26 23:32 starseeker
+ * [r78120] src/libged/npush/npush.cpp:
+ Cleanup, pass settings in to state
- * [r76970] src/libdm/tests/CMakeLists.txt,
- src/libdm/tests/rendered_image.cpp:
- Experiment with hooking a cut down version of
- https://github.com/zauonlok/renderer into the tcl_img demo code
- playing the role of the 'dm' viewer. Using one of the demo
- scenes, not any sort of CAD code, but better than random noise
- for visual inspection purposes.
+2021-01-25 19:52 starseeker
-2020-08-26 22:04 brlcad
+ * [r78119] src/libged/npush/npush.cpp:
+ checkpoint
- * [r76969] TODO:
- created a lighting environment manually too many times. need a
- studio box routine. note some other models that we be grand to
- have as well.
+2021-01-25 17:24 starseeker
-2020-08-26 15:55 starseeker
+ * [r78118] src/libged/npush/npush.cpp:
+ Step 1 - copy objects that need copying
- * [r76967] src/libdm/tests/tcl_img.cpp:
- Print an indicator of what the next DM/FB update will be.
+2021-01-25 15:50 starseeker
-2020-08-26 15:13 starseeker
+ * [r78117] src/libged/npush/npush.cpp:
+ Look at how each instance needs to be edited.
- * [r76966] TODO:
- Spoke too soon - embedded framebuffer is still an issue.
+2021-01-25 15:42 starseeker
-2020-08-26 14:48 starseeker
+ * [r78116] src/libged/npush/npush.cpp:
+ Cleanup
- * [r76965] src/mged/adc.c, src/mged/axes.c, src/mged/chgview.c,
- src/mged/color_scheme.c, src/mged/dozoom.c, src/mged/grid.c,
- src/mged/menu.c, src/mged/rect.c, src/mged/set.c,
- src/mged/share.c:
- Set the dm dirty flag when we're not using the DMP global as well
+2021-01-25 15:19 starseeker
-2020-08-26 14:37 starseeker
+ * [r78115] src/libged/npush/npush.cpp:
+ More reporting adjustments. Need to tune this for the various
+ cases.
- * [r76964] src/mged/set.c:
- Need to set the dm dirty flag here. Unfortunately, it looks like
- we were getting away with the X dm embedding in Tk by not having
- the embedded framebuffer come up - with this enabled we're back
- to crashing when we try to do an embedded X24 framebuffer in the
- main MGED Tk window.
+2021-01-25 15:10 starseeker
-2020-08-26 14:09 starseeker
+ * [r78114] src/libged/npush/npush.cpp:
+ Improve reporting, check candidate names against database and
+ each other.
- * [r76963] src/mged/mged.c:
- Need to clear out an old GEDP before creating a new one.
+2021-01-25 14:38 starseeker
-2020-08-26 13:11 d_rossberg
+ * [r78113] src/libged/npush/npush.cpp:
+ Make the 'look beyond the specified trees' feature optional (and
+ non-default) so we can better match existing push behavior. (The
+ non-locality feature probably also has significant performance
+ implications in large .g files, but that's currently untested.)
- * [r76962] src/librt/db5_scan.c:
- don't want to duplicate db5_diradd() and db5_diradd_handler()
- because of setting a single flag - used the free client_data
- parameter instead
- maybe there is a better posibility to detect an inmem directory
- handle which I don't know
+2021-01-25 14:25 starseeker
-2020-08-26 13:10 starseeker
-
- * [r76961] TODO:
- Remove note in trunk branch as well.
-
-2020-08-26 12:12 starseeker
-
- * [r76958] misc/repoconv/github_ci_actions.yml:
- Use Ninja where we can.
-
-2020-08-26 11:54 d_rossberg
-
- * [r76957] src/librt/db5_scan.c:
- reverted a bit from previous commit
- looks like it breaked at least the asc2g
-
-2020-08-26 10:54 d_rossberg
-
- * [r76955] include/rt/db_io.h, src/librt/db5_scan.c,
- src/librt/dir.c:
- functions to read the binary database data (as it is contained in
- a .g file) from RAM into an in-memory database instance
- the main function to use is rt_dirbuild_inmem()
- very early alpha status, but can be tested
-
-2020-08-26 03:06 starseeker
-
- * [r76954] regress/licenses/licenses_check.cpp:
- Fix copyright regex
-
-2020-08-26 01:27 starseeker
-
- * [r76953] src/qged/main_window.cpp, src/qged/main_window.h:
- Latest qtads defines a concept of a central widget, which matches
- what we need for the central display of an Archer style
- interface.
-
-2020-08-26 01:18 starseeker
-
- * [r76952] src/libdm/qt/dm-qt.h, src/qged/qtads/CMakeLists.txt,
- src/qged/qtads/DockAreaTabBar.cpp,
- src/qged/qtads/DockAreaTitleBar.cpp,
- src/qged/qtads/DockAreaWidget.cpp,
- src/qged/qtads/DockAreaWidget.h,
- src/qged/qtads/DockContainerWidget.cpp,
- src/qged/qtads/DockContainerWidget.h,
- src/qged/qtads/DockFocusController.cpp,
- src/qged/qtads/DockFocusController.h,
- src/qged/qtads/DockManager.cpp, src/qged/qtads/DockManager.h,
- src/qged/qtads/DockOverlay.cpp, src/qged/qtads/DockSplitter.cpp,
- src/qged/qtads/DockSplitter.h, src/qged/qtads/DockWidget.cpp,
- src/qged/qtads/DockWidget.h, src/qged/qtads/DockWidgetTab.cpp,
- src/qged/qtads/FloatingDragPreview.cpp,
- src/qged/qtads/images/maximize-button-focused.svg,
- src/qged/qtads/images/maximize-button.svg,
- src/qged/qtads/images/restore-button-focused.svg,
- src/qged/qtads/images/restore-button.svg:
- Update qtads with latest upstream changes.
-
-2020-08-25 22:37 brlcad
-
- * [r76951] src/libdm/glx/dm-ogl.c:
- initialize the window's background color in case it's defaulting
- to nothing (i.e., transparent). seems to matter on direct visuals
- with an alpha with latest X11.
-
-2020-08-25 20:59 starseeker
-
- * [r76950] NEWS:
- Definitely user visible - fix transparent window selection
- problem with some Linux desktop environments for DM and FB
- windows.
-
-2020-08-25 20:58 starseeker
-
- * [r76949] CMakeLists.txt, src/libdm/glx/dm-ogl.c,
- src/libdm/glx/if_ogl.c:
- Fix transparent display and framebuffer windows.
+ * [r78112] src/libged/npush/npush.cpp:
+ Remove combtree_i structure from npush - unneeded.
- Some configurations of Linux desktops (including default Gnome
- setups
- on Ubuntu and Redhat) have started using transparency by default.
- This
- has resulted in an odd semi-transparent window for default ogl
- display
- manger and framebuffer windows.
-
- Fix this by adding a filter the X11 visual choosing functions
- that
- rejects any window reporting an alpha mask. With this filter the
- selection returns opaque windows and restores the expected
- rendering
- appearance.
+ Was making things too complicated by assigning IDN matrices
+ to dp_i instances on the way down the push tree - combs will
+ need IDN matrices, but only when writing the comb instances
+ in the .g file. In the meantime, we need to keep the "current"
+ matrix state in order to recognize situations where we're going
+ to need to copy higher level combs with the dp_i instances.
+ Once non IDN matrices are recorded rather than thrown away, we
+ don't need any detailed look at the comb tree itself - the
+ matrix encapsulates the uniqueness (or non-uniqueness) of the
+ comb.
-2020-08-25 18:44 starseeker
+2021-01-23 13:48 starseeker
- * [r76948] doc/legal/embedded/CMakeLists.txt,
- doc/legal/embedded/strnstr.txt:
- Fix regress-licenses
+ * [r78111] src/libged/npush/npush.cpp:
+ comment adjustments
-2020-08-25 18:39 starseeker
+2021-01-22 22:02 starseeker
- * [r76947] src/tclscripts/mged/help.tcl,
- src/tclscripts/mged/openw.tcl:
- Remove the dmtype menu items, help entry
+ * [r78110] src/libged/npush/npush.cpp:
+ checkpoint
-2020-08-25 17:37 starseeker
+2021-01-22 20:33 starseeker
- * [r76946] regress/repository/CMakeLists.txt,
- regress/repository/repocheck.cpp, regress/repository/strnstr.c:
- Checkpoint some work on the repocheck tool.
-
- brlcad.org's BSD is suddenly producing segfaults with the
- repocheck
- tool, which aren't reproduced on Linux. Debugging, it looks as if
- the bu_mapped_file strings aren't being accepted by the C++
- string
- and strstr routines. The most likely cause is non-NULL
- termination
- of the bu_mapped_file buffer character arrays (which aren't
- guaranteed).
- Using the std::string constructor we can supply buflen from the
- mapped
- file to explicitly manage the content length, but strstr doesn't
- have
- such an option. The speedup from the strstr preliminary check is
- considerable, but what strnstr implementations do exist are
- fairly
- platform specific. Instead, took a stab at creating a fast,
- portable
- "bu_strnstr" by altering OpenBSD's strstr implementation to
- accept a
- length limit. For now, just adding in with the repocheck since
- this
- is experimental, but it may be that this is a concern more
- generally
- with strstr operations on bu_mapped_files, and if so it may
- belong
- in libbu proper...
+ * [r78109] src/libged/npush/npush.cpp:
+ Not sure these are being assembled correctly...
-2020-08-25 16:10 starseeker
+2021-01-22 19:54 starseeker
- * [r76945] regress/repository/repocheck.cpp:
- When constructing the string, use the explicit buflen from the
- mapped file.
+ * [r78108] src/libged/npush/npush.cpp:
+ comments
-2020-08-25 15:41 starseeker
+2021-01-22 19:49 starseeker
- * [r76944] src/mged/attach.c, src/tclscripts/mged/CMakeLists.txt,
- src/tclscripts/mged/dmtype.tcl:
- Runtime display manager type switching with dmtype never worked
- all that well, and currently doesn't work at all. Move the type
- reporting function (dmtype with no args reported the currently
- running dm type) to a subcommand of the dm command where it
- should have been in the first place (dmtype -> dm type). Remove
- the runtime dm type setting - to change MGED display manager
- types, alter the mged_default(dm_type) property in .mgedrc
+ * [r78107] src/libged/npush/npush.cpp:
+ Add some notes about how to proceed
-2020-08-25 14:27 starseeker
+2021-01-22 19:27 starseeker
- * [r76942] src/mged/buttons.c, src/mged/chgmodel.c,
- src/mged/chgview.c, src/mged/cmd.c, src/mged/edarb.c,
- src/mged/edsol.c, src/mged/mater.c, src/mged/overlay.c,
- src/mged/usepen.c:
- Tag the dm as dirty when update_views is set as well.
+ * [r78106] src/libged/npush/npush.cpp:
+ checkpoint
-2020-08-25 13:43 starseeker
+2021-01-22 16:02 starseeker
- * [r76941] regress/asc/asc2dsp.sh:
- Quote pathname
+ * [r78105] src/libged/npush/npush.cpp:
+ print full msg
-2020-08-25 12:51 starseeker
+2021-01-22 16:00 starseeker
- * [r76939] regress/coverage/CMakeLists.txt:
- List files for distclean
+ * [r78104] src/libged/npush/npush.cpp:
+ print matrix
-2020-08-25 12:27 starseeker
+2021-01-22 15:47 starseeker
- * [r76938] src/mged/dm-generic.c:
- uncomment
+ * [r78103] src/libged/npush/npush.cpp:
+ checkpoint
-2020-08-25 11:57 starseeker
+2021-01-21 22:50 starseeker
- * [r76937] src/mged/adc.c, src/mged/set.c:
- couple more dm sets for the dirty flag
+ * [r78102] doc/docbook/system/mann/npush.xml:
+ Lay out an example that requires parent comb redefinition. Still
+ not sure how to implement this yet, but there are clearly cases
+ where it can be necessary.
-2020-08-25 03:03 starseeker
+2021-01-21 21:47 starseeker
- * [r76936] src/mged/attach.c, src/mged/cmd.c,
- src/mged/dm-generic.c, src/mged/doevent.c, src/mged/dozoom.c,
- src/mged/fbserv.c, src/mged/mged.c, src/mged/predictor.c:
- Needs more testing, but adjusting the setting of the dirty flags
- gives saner behavior with the classic X libdm backend. Still
- can't embed the X dm backend in Tk with 8.6 - that seems to be a
- different category of failure - but this allows the X dm to
- function in classic mode. Checkpointing a state that shows
- progress.
+ * [r78101] src/libged/tests/push_tests.g:
+ Add another test case
-2020-08-25 02:29 starseeker
+2021-01-21 16:42 starseeker
- * [r76935] src/mged/adc.c, src/mged/attach.c, src/mged/axes.c,
- src/mged/buttons.c, src/mged/chgview.c, src/mged/cmd.c,
- src/mged/color_scheme.c, src/mged/doevent.c, src/mged/dozoom.c,
- src/mged/edsol.c, src/mged/fbserv.c, src/mged/grid.c,
- src/mged/menu.c, src/mged/mged.c, src/mged/mged_dm.h,
- src/mged/rect.c, src/mged/set.c, src/mged/share.c:
- Switch MGED's display set from using bu_list to using a bu_ptbl.
+ * [r78100] src/libged/npush/npush.cpp:
+ Adjust printing messages for npush
-2020-08-25 01:51 brlcad
+2021-01-21 16:06 starseeker
- * [r76934] TODO:
- It's always been a bit strange to me that one can't just put a
- matrix over any object (comb or primitive). Probably a carryover
- from comgeom days, but one we really should rectify in v6.
+ * [r78099] NEWS, src/libged/ged.c, src/libged/ged_private.h,
+ src/libged/tree/tree.c:
+ Add verbosity option to tree command to label comb instances with
+ non-IDN matrices.
-2020-08-24 22:22 starseeker
+2021-01-21 16:04 starseeker
- * [r76933] src/mged/adc.c, src/mged/axes.c, src/mged/buttons.c,
- src/mged/chgview.c, src/mged/cmd.c, src/mged/color_scheme.c,
- src/mged/edsol.c, src/mged/grid.c, src/mged/rect.c,
- src/mged/set.c:
- dmlp->m_dmp
+ * [r78098] src/libged/npush/npush.cpp:
+ Add comments
-2020-08-24 22:21 starseeker
+2021-01-20 17:33 starseeker
- * [r76932] src/mged/adc.c, src/mged/attach.c, src/mged/axes.c,
- src/mged/buttons.c, src/mged/chgview.c, src/mged/cmd.c,
- src/mged/color_scheme.c, src/mged/dm-generic.c,
- src/mged/doevent.c, src/mged/dozoom.c, src/mged/edsol.c,
- src/mged/fbserv.c, src/mged/grid.c, src/mged/menu.c,
- src/mged/mged.c, src/mged/mged_dm.h, src/mged/overlay.c,
- src/mged/predictor.c, src/mged/rect.c, src/mged/set.c,
- src/mged/share.c, src/mged/titles.c, src/mged/usepen.c:
- dml_->dm_
+ * [r78097] src/libged/npush/npush.cpp:
+ Add more comments
-2020-08-24 22:15 starseeker
+2021-01-20 16:28 starseeker
- * [r76931] src/mged/adc.c, src/mged/attach.c, src/mged/buttons.c,
- src/mged/chgview.c, src/mged/cmd.c, src/mged/color_scheme.c,
- src/mged/doevent.c, src/mged/dozoom.c, src/mged/edsol.c,
- src/mged/fbserv.c, src/mged/menu.c, src/mged/mged.c,
- src/mged/mged_dm.h, src/mged/overlay.c, src/mged/predictor.c,
- src/mged/set.c, src/mged/share.c, src/mged/titles.c,
- src/mged/usepen.c:
- curr_dm_list -> mged_curr_dm
+ * [r78096] src/libged/npush/npush.cpp:
+ Comment state variables
-2020-08-24 22:11 starseeker
+2021-01-19 19:30 starseeker
- * [r76930] src/mged/adc.c, src/mged/attach.c, src/mged/axes.c,
- src/mged/buttons.c, src/mged/chgview.c, src/mged/cmd.c,
- src/mged/color_scheme.c, src/mged/doevent.c, src/mged/dozoom.c,
- src/mged/edsol.c, src/mged/fbserv.c, src/mged/grid.c,
- src/mged/menu.c, src/mged/mged.c, src/mged/mged.h,
- src/mged/mged_dm.h, src/mged/rect.c, src/mged/sedit.h,
- src/mged/set.c, src/mged/share.c:
- Rename dm_list container to better reflect what it is - an mged
- superset of struct dm
+ * [r78095] src/libged/npush/npush.cpp:
+ Take s_i out of the push_state - not used in tree walking
-2020-08-24 22:04 starseeker
+2021-01-19 19:27 starseeker
- * [r76929] src/mged/adc.c, src/mged/attach.c, src/mged/axes.c,
- src/mged/buttons.c, src/mged/chgview.c, src/mged/cmd.c,
- src/mged/color_scheme.c, src/mged/dozoom.c, src/mged/edsol.c,
- src/mged/fbserv.c, src/mged/grid.c, src/mged/menu.c,
- src/mged/mged.c, src/mged/mged_dm.h, src/mged/rect.c,
- src/mged/set.c, src/mged/share.c:
- rename
+ * [r78094] src/libged/npush/npush.cpp:
+ Adjust combtree, add more comments
-2020-08-24 22:00 starseeker
+2021-01-15 20:00 starseeker
- * [r76928] src/mged/adc.c, src/mged/attach.c, src/mged/cmd.c,
- src/mged/dm-generic.c, src/mged/doevent.c, src/mged/fbserv.c,
- src/mged/mged.c, src/mged/mged_dm.h, src/mged/predictor.c,
- src/mged/set.c:
- Make it a little easier to grep for the MGED dirty flag. The
- attached X dm seems to be hot spinning (refreshing continually
- when nothing has changed, maxing one CPU core), need to figure
- out why...
+ * [r78093] src/libged/tests/push_tests.g:
+ It's not best practice, but it is allowed by librt - make an
+ example with three of the exact same comb in the same group,
+ differentiated only by their instance matrices.
-2020-08-24 21:05 starseeker
+2021-01-15 19:55 starseeker
- * [r76927] src/libdm/dm_plugins.cpp:
- Note these functions will always fall back to 'nu'
+ * [r78092] src/libged/npush/npush.cpp,
+ src/libged/tests/push_tests.g:
+ Add example of comb present in different levels of a push tree.
-2020-08-24 20:53 starseeker
+2021-01-15 19:07 starseeker
- * [r76926] src/mged/dm-generic.c:
- Check mvars and vparse returns once, up front, before use
+ * [r78091] src/libged/npush/npush.cpp:
+ Rough in logic to keep the tree walk going post-leaf for
+ information.
-2020-08-24 18:30 starseeker
+2021-01-15 18:59 starseeker
- * [r76925] src/mged/attach.c:
- System Tcl/Tk doesn't like calling Tk_DeleteGenericHandler in
- classic mode with a non-Tk dm like txt. Only do Tk event handler
- logic if the dm we open reports it is graphical.
+ * [r78090] src/libged/npush/npush.cpp:
+ Pass the survey state, so we can change how we're recording more
+ easily mid-tree-walk.
-2020-08-24 18:22 starseeker
+2021-01-15 18:13 starseeker
- * [r76924] src/libdm/dm-generic.c:
- Protect a couple more pointer returning dm functions with magic
- checks.
+ * [r78089] src/libged/npush/npush.cpp:
+ checkpoint
-2020-08-24 15:45 starseeker
+2021-01-15 16:40 starseeker
- * [r76923] include/dm.h, src/libdm/dm-generic.c:
- Don't actually use dm_get_vp - remove
+ * [r78088] src/libged/npush/npush.cpp:
+ Not sure how yet, but going to have to develop ability to
+ propagate comb tree changes back up the hierarchy if necessary.
-2020-08-24 15:40 starseeker
+2021-01-15 15:44 starseeker
- * [r76922] src/libdm/dm-generic.c:
- Adjust several dm functions to never return NULL.
+ * [r78087] src/libged/npush/npush.cpp:
+ Print when we have differing trees.
-2020-08-24 15:30 starseeker
+2021-01-15 15:38 starseeker
- * [r76921] include/bu/magic.h, include/dm/defines.h,
- src/libdm/X/dm-X.c, src/libdm/dm-generic.c,
- src/libdm/glx/dm-ogl.c, src/libdm/null/dm-Null.c,
- src/libdm/osgl/dm-osgl.cpp, src/libdm/plot/dm-plot.c,
- src/libdm/postscript/dm-ps.c, src/libdm/qt/dm-qt.cpp,
- src/libdm/tk/dm-tk.c, src/libdm/txt/dm-txt.c,
- src/libdm/wgl/dm-wgl.c:
- Add a BU_CKMAG check to ensure vls returning function calls can
- never return NULL.
+ * [r78086] src/libged/npush/npush.cpp:
+ Halting pushes above primitives has complex implications for
+ combs.
-2020-08-24 15:30 starseeker
+2021-01-15 15:37 starseeker
- * [r76920] src/mged/mged.c:
- libdm initialization should vls init this, not the app
+ * [r78085] src/libged/tests/push_tests.g:
+ Add more tests
-2020-08-24 15:07 starseeker
+2021-01-15 01:52 starseeker
- * [r76919] src/mged/cmd.c:
- Few more get_pathname adjustments
+ * [r78084] src/libged/npush/npush.cpp:
+ TODO note... this needs to be more robust.
-2020-08-24 13:49 starseeker
+2021-01-15 01:49 starseeker
- * [r76918] src/mged/cmd.c:
- Fix dm pointer validation target, check result.
+ * [r78083] src/libged/npush/npush.cpp:
+ Not a renaming candidate unless the matrix is non-IDN
-2020-08-24 13:40 brlcad
+2021-01-14 23:46 starseeker
- * [r76917] src/libdm/dm_plugins.cpp:
- checking null is unnecessary/redundant with BU_STR_EQUIV() as it
- handles them (they're equivalent to an empty string) so there's
- no risk of crashing.
+ * [r78082] src/libged/npush/npush.cpp:
+ checkpoint
-2020-08-24 13:39 starseeker
+2021-01-14 21:43 starseeker
- * [r76916] src/mged/attach.c:
- Variety of dm NULL checks, local vars to improve readability.
+ * [r78081] src/libged/npush/npush.cpp:
+ checkpoint
-2020-08-24 13:17 starseeker
+2021-01-14 19:44 starseeker
- * [r76915] src/libtclcad/view/util.c:
- Check dm_get_pathname result in to_is_viewable
+ * [r78080] src/libged/npush/npush.cpp:
+ Check list of objects against the active trees - don't try to
+ push if an object that is below a specified object is also
+ specified. Particularly if depth limitations are specified, I
+ don't know whether this is safe (or guaranteed to produce
+ intuitive results) in all cases.
-2020-08-24 13:17 starseeker
+2021-01-14 18:48 starseeker
- * [r76914] src/libtclcad/tclcad_polygons.c:
- More adjustments to dm_get_pathname calls
+ * [r78079] src/libged/npush/npush.cpp:
+ Make a note - we'll have to do each specified push fully before
+ processing the next, if multiple comb inputs are specified to the
+ push command.
-2020-08-24 13:11 starseeker
+2021-01-14 18:43 starseeker
- * [r76913] src/libtclcad/tclcad_obj.c:
- Set local struct bu_vls * var to dm_get_pathname output - improve
- readability, check for NULL and zero length.
+ * [r78078] src/libged/npush/npush.cpp:
+ checkpoint
-2020-08-24 12:49 starseeker
+2021-01-14 15:05 starseeker
- * [r76912] src/libdm/dm_plugins.cpp, src/libtclcad/tclcad_mouse.c,
- src/libtclcad/tclcad_obj.c:
- Start going through Sean's list of dm functions - check their
- behavior in the calling code if NULL is returned. Ways to go
- yet...
+ * [r78077] src/libged/npush/npush.cpp:
+ Leaves in a push aren't necessarily solids - will need to handle
+ differently.
-2020-08-24 05:46 brlcad
+2021-01-13 20:40 starseeker
- * [r76911] NEWS, TODO, doc/docbook/system/man1/rtweight.xml:
- per todo, improve rtweight docs explaining that it reports the
- geometric centroid of objects
+ * [r78076] src/libged/npush/npush.cpp:
+ Start working on how to check comb trees for uniqueness in a .g
+ file
-2020-08-24 04:59 brlcad
+2021-01-13 19:11 starseeker
- * [r76910] regress/weight/weight.ref,
- regress/weight/weight.test2.ref, src/rt/viewweight.c:
- put a header line on the rtweight density table output.
+ * [r78075] doc/docbook/system/mann/npush.xml,
+ src/libged/tests/push_tests.g:
+ Add another example that would require comb copying with a depth
+ limit set.
-2020-08-24 03:42 brlcad
+2021-01-13 18:05 starseeker
- * [r76909] src/rt/viewarea.c:
- explain what set_factor was doing, had wrong comment, and renamed
- for future clarity.
+ * [r78074] src/libged/npush/npush.cpp:
+ Include string header
-2020-08-23 17:48 starseeker
+2021-01-13 16:43 starseeker
- * [r76908] NEWS, doc/docbook/system/man1/remrt.xml,
- src/remrt/remrt.c:
- Report the TCP port, which is what the user is most likely to
- actually need when setting up a remrt run.
+ * [r78073] src/libged/npush/npush.cpp:
+ Make a note - we need to think about comb trees in all this,
+ since they define comb objects which may not be localized to just
+ the pushed tree...
-2020-08-23 16:44 brlcad
+2021-01-13 16:29 starseeker
- * [r76907] src/libdm/X/dm-X.c, src/libdm/glx/dm-ogl.c,
- src/libdm/include/calltable.h, src/libdm/null/dm-Null.c,
- src/libdm/osgl/dm-osgl.cpp, src/libdm/plot/dm-plot.c,
- src/libdm/postscript/dm-ps.c, src/libdm/qt/dm-qt.cpp,
- src/libdm/tk/dm-tk.c, src/libdm/txt/dm-txt.c,
- src/libdm/wgl/dm-wgl.c:
- consolidate the embedded bu_vls strings to encourage
- initializing/releasing together.
+ * [r78072] src/libged/tests/push_tests.g:
+ Add mixed depth hierarchy
-2020-08-23 16:18 brlcad
+2021-01-13 15:27 starseeker
- * [r76906] src/libdm/tests/dm_test.c:
- protect against null dereferencing
+ * [r78071] src/libged/tests/push_tests.g:
+ Add a deeper hierarchy with matrices
-2020-08-23 16:15 brlcad
+2021-01-13 15:06 starseeker
- * [r76905] src/libdm/dm-generic.c:
- should manually initialize all embedded structs on creation,
- particularly vls so the magic is set correctly
+ * [r78070] doc/README.Windows:
+ Elaborate more on the RelWithDebInfo CMake mode as it pertains to
+ Visual STStudio.
-2020-08-23 16:13 brlcad
+2021-01-12 21:10 starseeker
- * [r76904] src/mged/edsol.c, src/mged/mged.c, src/mged/mged_dm.h:
- eliminate the preprocessor indirection, check for null
- dereferencing
+ * [r78069] NEWS, doc/docbook/system/mann/quit.xml, src/mged/cmd.c,
+ src/tclscripts/mged/mview.tcl:
+ This needs to be done with a more informative dialog, if we're
+ going to go this route - revert r77990
-2020-08-23 16:12 brlcad
+2021-01-12 18:54 starseeker
- * [r76903] src/libtclcad/tclcad_obj.c, src/libtclcad/view/util.c:
- make sure we're not dereferencing null
+ * [r78068] doc/docbook/system/mann/CMakeLists.txt,
+ doc/docbook/system/mann/npush.xml, src/libged/npush/npush.cpp:
+ Start roughing out a man page. Need to clearly articulate what
+ some of these ideas are supposed to do in order to know what has
+ to be implemented, so might as well document the meaning while
+ we're at it...
-2020-08-23 16:10 brlcad
+2021-01-12 15:33 starseeker
- * [r76902] src/libdm/dm_plugins.cpp:
- ws consistency cleanup
+ * [r78067] src/libged/lint/lint.cpp:
+ Fix crash if lint -F option is passed invalid search args. (lint
+ isn't documented yet, so not really user visible.)
-2020-08-23 16:09 brlcad
+2021-01-12 15:25 starseeker
- * [r76901] src/libdm/dm_plugins.cpp:
- simplify and make sure we're not printing a null pointer. seems
- unlikely logic is correctly / gracefully handling there being no
- dm.
+ * [r78066] src/libged/npush/npush.cpp:
+ checkpoint
-2020-08-23 13:35 brlcad
+2021-01-12 14:37 starseeker
- * [r76900] src/libanalyze/density.cpp:
- trigger an anything greater than our capacity for safety, not
- just the full value.
+ * [r78065] src/libged/npush/npush.cpp:
+ Also use the leaf matrix here, if there is one.
-2020-08-23 05:50 brlcad
+2021-01-12 14:36 starseeker
- * [r76899] src/remrt/rtsrv.c:
- init more, remove some continuations.
+ * [r78064] src/libged/tests/push_tests.g:
+ Add a test with object usage outside tree to push
-2020-08-23 03:49 brlcad
+2021-01-12 14:10 starseeker
- * [r76898] BUGS:
- fbclear -c appears to hang indefinitely. don't know when it
- broke.
+ * [r78063] src/libged/tests/CMakeLists.txt,
+ src/libged/tests/push_tests.g:
+ Start assembling some basic test cases for matrix pushing.
-2020-08-23 02:12 starseeker
+2021-01-11 22:01 starseeker
- * [r76897] doc/docbook/system/man1/remrt.xml:
- Attempt to add a start-to-finish example of a Hellow World remrt
- usage.
+ * [r78062] src/libged/npush/npush.cpp:
+ Report both need-copy and push-matrix-only.
-2020-08-23 00:42 starseeker
+2021-01-11 19:22 starseeker
- * [r76896] NEWS:
- Note support for help options added to g2asc and asc2dsp
+ * [r78061] src/libged/npush/npush.cpp:
+ Checkpoint. Slowly working through the conceptual implications of
+ various forms of push.
-2020-08-22 23:49 starseeker
+2021-01-08 19:08 starseeker
- * [r76895] src/libdm/if_stack.c:
- Check implementation, not wrapping pointer
+ * [r78060] src/libged/npush/npush.cpp:
+ checkpoint
-2020-08-22 22:23 starseeker
+2021-01-08 18:03 starseeker
- * [r76894] src/libdm/dm_init.cpp, src/libdm/dm_plugins.cpp:
- Don't special case in fb_open, just add the built-in fb
- interfaces to the map.
+ * [r78059] src/libged/npush/npush.cpp:
+ checkpoint
-2020-08-22 16:17 starseeker
+2021-01-08 16:10 starseeker
- * [r76893] misc/tools/debug2c/debug2c.cxx:
- Handle help options in debug2c so it won't write spurious -? file
- in usage.sh execution.
+ * [r78058] src/libged/npush/npush.cpp:
+ Add a verbosity option - I suspect this type of output may prove
+ useful in the long run, so let's promote it beyond
+ development/debugging throwaway code to a feature.
-2020-08-22 16:14 starseeker
+2021-01-08 16:00 starseeker
- * [r76892] regress/usage.sh:
- IIF we're doing one process at a time, check for -? files
+ * [r78057] src/libged/npush/npush.cpp:
+ Checkpoint
-2020-08-22 16:03 starseeker
+2021-01-08 15:33 starseeker
- * [r76891] src/conv/asc/asc2dsp.c, src/conv/asc/g2asc.c:
- Spotted a couple of programs in usage.sh output that are treating
- '-h' and '-?' as filenames and creating a '-?' ouput file. Add
- basic option handling so they interpret -h and -? as help
- requests.
+ * [r78056] src/libged/npush/npush.cpp:
+ Checkpoint
-2020-08-22 15:53 starseeker
+2021-01-07 19:01 brlcad
- * [r76890] src/libbu/opt.c:
- The bare '-' character may be used to denote arguments for
- programs (for example, to use stdin or stdout in lieu of a file)
- - update bu_opt to recognize that the isolated '-' char isn't an
- option. (Note - need to think about whether the "--" string is in
- the same category... probably...
+ * [r78055] TODO:
+ dbconcat appears to be not working even on a simple case
-2020-08-22 15:31 starseeker
+2021-01-07 17:41 starseeker
- * [r76889] regress/asc/CMakeLists.txt, regress/asc/asc2dsp.sh:
- The 'old' test doesn't run - input file reports invalid
- characters - but the new one does run. Rather than completely
- turn off the test, run the conversion that does work.
+ * [r78054] src/libged/npush/npush.cpp:
+ Rough in various npush options
-2020-08-22 14:40 starseeker
+2021-01-07 16:36 starseeker
- * [r76888] src/sig/dauto.c:
- This is the source of the bu_malloc(0) report and dauto crash log
- in the usage.sh run. Need to validate the numerical conversion
- before trying to malloc memory. Still not sure why I was seeing
- intermittent failure to print the usage message, but so far can't
- reproduce that failure after fixing this.
+ * [r78053] src/libged/npush/npush.cpp:
+ Whoops, commited too many files. Get npush building
-2020-08-22 06:22 brlcad
+2021-01-07 16:33 starseeker
- * [r76887] regress/comgeom/comgeom.sh:
- still could be broken/wrong. need to simplly use the variable.
+ * [r78052] src/libged/npush/npush.cpp, src/libged/which/which.cpp:
+ Missed ? when enabling help option for which GED cmd.
-2020-08-22 06:15 brlcad
+2021-01-07 15:27 starseeker
- * [r76886] sh/cmp.sh:
- set mged commands to a variable so they can be automatically
- found
+ * [r78051] include/ged/commands.h, src/libged/CMakeLists.txt,
+ src/libged/exec_mapping.cpp, src/libged/npush,
+ src/libged/npush/CMakeLists.txt, src/libged/npush/npush.cpp,
+ src/libged/xpush/xpush.c, src/libtclcad/commands.c,
+ src/mged/setup.c:
+ Rough in a command for experimenting with push/xpush style logic.
-2020-08-22 06:06 brlcad
+2021-01-07 14:04 starseeker
- * [r76885] src/fb/fbstretch.c, src/sig/dauto.c:
- revert c76872. bu_exit() is not the problem. bug masking is a
- straight up antipattern.
+ * [r78050] src/libanalyze/raydiff.c:
+ Clean up rtip
-2020-08-22 05:54 brlcad
+2021-01-07 01:47 starseeker
- * [r76884] src/sig/dauto.c:
- unable to reproduce, bu_exit() is working as intended on mac.
+ * [r78049] src/archer/CMakeLists.txt:
+ Fix Archer shortcuts
-2020-08-22 05:19 brlcad
+2021-01-06 23:34 brlcad
- * [r76883] src/libbu/bomb.c:
- it's correct to include setjmp.h because the implementation
- directly calls longjmp(), even if headers include it for other
- reasons
+ * [r78048] src/util/pixscale.c:
+ not a fix, but this looks like it's off by one
-2020-08-22 04:43 brlcad
+2021-01-06 23:33 brlcad
- * [r76882] CHANGES:
- document removal of facetall.sh
+ * [r78047] src/util/bwmod.c:
+ we reopen file descriptors, so really shouldn't be using the
+ numbers direct.
-2020-08-22 04:25 brlcad
+2021-01-06 23:32 brlcad
- * [r76881] NEWS:
- reword for clarity (Remove the old facetall.sh script - MGED
- facetize command should now be a superset of this functionality.)
+ * [r78046] TODO:
+ pixscale aborts premature
-2020-08-21 23:50 starseeker
+2021-01-06 17:29 starseeker
- * [r76880] src/libtclcad/tclcad_init.c:
- Cleanup, provide Tclcad_Init for .so file loading
-
-2020-08-21 22:35 starseeker
-
- * [r76879] src/libtclcad/tclcad_init.c, src/rtwizard/main.c:
- tclcad_init is the call to use. Also not directly initing Itcl
- from rtwizard.
-
-2020-08-21 22:33 starseeker
-
- * [r76878] src/libdm/dm_plugins.cpp:
- We need to do better than exact string lookup - replicate the old
- _if_list[i]->if_name bu_strncmp check in the plugin test. Fixes
- rtwizard preview and render.
-
-2020-08-21 22:18 starseeker
-
- * [r76877] src/tclscripts/rtwizard/rtwizard:
- No longer need to do this.
-
-2020-08-21 20:45 starseeker
-
- * [r76876] NEWS, src/librt/primitives/vol/vol.c:
- Sumagna Das reported that drawing a VOL primitive with missing
- data file was crashing MGED. Plotting routine wasn't validating
- the map in vip before trying to access it. Check first, and
- simply return if no map is present (nothing to draw.)
-
-2020-08-21 20:27 starseeker
-
- * [r76875] regress/CMakeLists.txt, regress/coverage,
- regress/coverage/CMakeLists.txt, regress/coverage/covgcheck.cpp,
- regress/coverage/regress-coverage.cmake.in:
- Checkpoint work on logic to read through the regression tests and
- spot the execution of programs. Patterns undoubtedly need
- improvement, but getting semi-sane results commiting with the
- regression test itself commented out.
-
-2020-08-21 16:50 starseeker
-
- * [r76874] CMakeLists.txt, misc/CMake/BRLCAD_Targets.cmake:
- Have the build system report out the BRLCAD_ADDEXEC programs that
- aren't test programs or otherwise excluded from installation.
- These are the programs of primary interest for testing.
-
-2020-08-21 16:02 starseeker
-
- * [r76873] NEWS, sh/CMakeLists.txt, sh/facetall.sh:
- Remove the old facetall.sh script - MGED facetize command should
- now be a superset of this functionality.
-
-2020-08-21 15:43 starseeker
-
- * [r76872] src/fb/fbstretch.c, src/sig/dauto.c:
- Still can't determine why bu_exit is an issue in these cases
- (only commonality seems to be bu_exit called right after isatty -
- not sure why that would matter...) but bu_exit isn't really
- necessary to print the help msgs. Just bu_log and return.
- usage.sh can complete its run without throwing bombs now on
- linux, although one or more programs are producing a '-?' file
- rather than intepreting ? as a help option.
-
-2020-08-21 15:37 starseeker
-
- * [r76871] src/libbn/sphmap.c:
- Add some sanity checking for the sphmap routines called by
- pix-spm.
-
-2020-08-21 15:20 starseeker
-
- * [r76870] src/libbu/bomb.c:
- No, that doesn't do it either. Failure doesn't seem to be 100%
- reliable - usage.sh can trigger it, but testing in isoluation
- only produces occasional failures. Almost as if some global state
- is involved...
-
-2020-08-21 15:09 starseeker
-
- * [r76869] regress/usage.sh, src/libbu/bomb.c, src/sig/dauto.c:
- Ah! There we go. Adding setjmp.h to bomb.c's system headers
- works. Should already have been coming in from bu/parallel.h
- though...
-
-2020-08-21 15:05 starseeker
-
- * [r76868] src/sig/dauto.c:
- Including bu/parallel.h gets bu_exit behaving again, but not
- clear why.
-
-2020-08-21 14:55 starseeker
-
- * [r76867] src/sig/dauto.c:
- bu_exit doesn't seem to succeed here - BU_SETJMP is set somehow,
- and bu_bomb is called by bu_exit. Just log the message and
- return, until it's clearer what's going on.
-
-2020-08-21 14:27 starseeker
-
- * [r76866] src/libdm/dm_plugins.cpp:
- rtwizard check caught it - /dev/mem and /dev/stack interfaces are
- built in, not loaded as plugins - check accordingly.
-
-2020-08-21 14:09 starseeker
-
- * [r76865] regress/solids/solids.sh:
- scrub unnecessary logging lines from solids.sh
-
-2020-08-21 14:02 starseeker
-
- * [r76864] regress/comgeom/comgeom.sh:
- Fix logfile name being grepped by comgeom regression test for
- errors.
-
-2020-08-21 12:30 starseeker
-
- * [r76861] CMakeLists.txt, src/adrt/CMakeLists.txt,
- src/archer/CMakeLists.txt, src/art/CMakeLists.txt,
- src/burst/CMakeLists.txt, src/burst/old/CMakeLists.txt,
- src/fb/CMakeLists.txt, src/fbserv/CMakeLists.txt,
- src/libdm/CMakeLists.txt, src/libdm/X/CMakeLists.txt,
- src/libdm/glx/CMakeLists.txt, src/libdm/osgl/CMakeLists.txt,
- src/libdm/plot/CMakeLists.txt,
- src/libdm/postscript/CMakeLists.txt, src/libdm/qt/CMakeLists.txt,
- src/libdm/tk/CMakeLists.txt, src/libdm/txt/CMakeLists.txt,
- src/libdm/wgl/CMakeLists.txt, src/libged/CMakeLists.txt,
- src/mged/CMakeLists.txt, src/remrt/CMakeLists.txt,
- src/rt/CMakeLists.txt, src/rtwizard/CMakeLists.txt,
- src/sig/CMakeLists.txt, src/util/CMakeLists.txt:
- Make a couple of build system adjustments (set target folders,
- mark as advanced). Major change is to do for libdm what we did
- for libged and make a single target for apps to use as a
- dependency lister, rather than calling out individual backends -
- that's not robust to changes in backend target names and would be
- fragile in the long run.
-
-2020-08-19 14:54 starseeker
-
- * [r76849] src/libbu/process.c:
- Stash a note - not sure where the number 20 came from here...
-
-2020-08-19 14:18 starseeker
-
- * [r76848] src/mged/mged.c:
- Sean reports negative experience with Tcl_Sleep - not clear how
- much benefit it provides here, so remove it until there's a clear
- need.
-
-2020-08-19 13:39 starseeker
-
- * [r76847] src/libbu/process.c, src/mged/mged.c:
- Valgrind pegged which fd was getting leaked. If we close this,
- also shouldn't process Tcl events in cmdline mode after the
- commands finish.
-
-2020-08-19 08:21 brlcad
-
- * [r76846] src/rt/viewweight.c:
- more thread-sane, print as one block
-
-2020-08-19 08:18 brlcad
-
- * [r76845] src/rt/viewweight.c:
- track and report on the first few IDs that don't have a density
- defined, still silently ignoring the rest. update the message as
- being a warning instead of advisotry. don't report on objects
- without an ID.
-
-2020-08-19 02:06 starseeker
-
- * [r76844] src/librt/test_root3-subd.cpp:
- Dust off the root3 subdivision test program.
-
-2020-08-19 01:44 starseeker
-
- * [r76843] include/bg/tri_pt.h:
- Fix comment
-
-2020-08-19 01:36 starseeker
-
- * [r76842] include/bg/lseg.h, src/libbg/lseg_lseg.c,
- src/libbg/lseg_pt.c, src/libbg/tests/lseg_lseg.c,
- src/libbg/tests/lseg_pt.c, src/libbrep/opennurbs_ext.cpp,
- src/libged/snap.c:
- Rename to match the convention used by related functions in libbn
- (which need to migrate to libbg)
-
-2020-08-19 01:24 starseeker
-
- * [r76841] CHANGES:
- Experience with CDT proved it is better to implement convex hull
- functions in such a way that the decision of whether to allocate
- new arrays for vertices is left to the caller. Will need to
- change function signatures, and provide separatew utility
- functions that package up that bookkeeping.
-
-2020-08-19 01:05 starseeker
-
- * [r76840] doc/legal/embedded/CMakeLists.txt,
- doc/legal/embedded/nurbs_fit.txt, src/libbrep/CMakeLists.txt,
- src/libbrep/opennurbs_fit.cpp, src/libbrep/opennurbs_fit.h,
- src/libbrep/surface_tree_queue_tests.patch:
- Remove a few more old, unused files
-
-2020-08-19 00:59 starseeker
-
- * [r76839] doc/legal/embedded/CMakeLists.txt,
- doc/legal/embedded/point_in_polyhedron.txt, include/bg/trimesh.h,
- include/brep/CMakeLists.txt, include/brep/cdt2.h,
- src/libbg/CMakeLists.txt, src/libbg/tests/CMakeLists.txt,
- src/libbg/tests/trimesh_pt_in.c, src/libbg/trimesh_pt_in.c,
- src/libbrep/CMakeLists.txt, src/libbrep/cdt/omesh.cpp,
- src/libbrep/cdt/overt.cpp, src/libbrep/cdt/ovlps.cpp,
- src/libbrep/cdt/ovlps_grps.cpp, src/libbrep/cdt/tri_isect.cpp,
- src/libbrep/cdt/util.cpp, src/libbrep/cdt2:
- Strip out disable/incomplete code from the CDT implementation.
- VCS has it if an opportunity to revisit it appears.
-
-2020-08-19 00:37 starseeker
-
- * [r76838] src/libbg/tests/CMakeLists.txt:
- Arrrgh. Looks like I recorded the wrong triangle - probably
- followed the ray backwards instead of forwards, from the looks of
- things. Not a useful test, will have to track down another
- example.
-
-2020-08-18 20:07 starseeker
-
- * [r76834] include/analyze/pnts.h, src/libanalyze/CMakeLists.txt,
- src/libanalyze/pnt_inside.c, src/libanalyze/polygonizer.c,
- src/libged/CMakeLists.txt, src/libged/cg,
- src/libged/pnts/pnts.cpp, src/libged/pnts_util.c,
- src/libged/pnts_util.h, src/libged/tests/CMakeLists.txt,
- src/libged/tests/cg_tests.g:
- Back out the exposure of the point inside/outside test - needs
- more design thought.
-
-2020-08-18 16:26 starseeker
-
- * [r76833] src/libged/tests/CMakeLists.txt,
- src/libged/tests/cg_tests.g:
- Add a test .g file for the cg pnts operation - this has three
- points on or inside the sphere and three outside it. Not sure yet
- if we're ready to actually wire up a test.
-
-2020-08-18 16:03 brlcad
-
- * [r76832] src/libged/which/which.cpp:
- back_inserter needs iterator.h
-
-2020-08-18 15:58 starseeker
-
- * [r76831] src/libged/pnts/pnts.cpp:
- Remove the 'inside' subcommand from pnts
-
-2020-08-18 15:57 starseeker
-
- * [r76830] src/libged/cg/CMakeLists.txt, src/libged/cg/cg.cpp,
- src/libged/cg/eval_pnts.cpp, src/libged/cg/ged_cg.h:
- Add the point inside/outside test to cg
+ * [r78045] NEWS, src/archer/archer.bat, src/archer/archer.c,
+ src/mged/CMakeLists.txt, src/mged/mged.bat, src/mged/mged.c,
+ src/rtwizard/CMakeLists.txt, src/rtwizard/main.c,
+ src/rtwizard/rtwizard.bat:
+ Adjust GUI apps default working dir to HOME.
- This supports two operations - intersection and subtraction -
- between
- a point cloud on the left and a volumetric object on the right.
- It
- will find the subset of the points in the left hand object that
- are
- either inside and on (i.e. intersecting) or outside the test
- volume.
- Command form is:
+ When launched from a system installed location via icon, the
+ default
+ working directory of MGED, Archer and rtwizard gets set to the
+ bin
+ directory that holds the executable. This is typically a
+ read-only
+ location, which results in a variety of problems when trying to
+ run
+ various commands.
- cg eval "pnts.s + sph.s" inside.pnts
+ The most typical case where this is a problem is Windows desktop
+ and
+ start menu shortcuts. The NSIS installer has a way to adjust the
+ directory, but the MSI installer generated by WiX is a tougher
+ nut to
+ crack. Various experiments have so far failed to identify a
+ working
+ approach to adjusting the shortcut startup directories with
+ CMake/WiX.
+ Using Windows bat files as launchers works, but using them with
+ the
+ MSI installer results in incorrect icons for the desktop
+ shortcuts
+ with no obvious way to override them.
- cg eval "pnts.s - sph.s" outside.pnts
+ Rather than continue to jump through hoops, just set the working
+ dir
+ directly to the user's home directory when these three
+ applications
+ are invoked without arguments. This should preserve all existing
+ behaviors for any scripting cases, but will avoid a read-only
+ working
+ directory when using the no-args launch produced by desktop
+ shortcuts.
-2020-08-18 14:43 starseeker
+2021-01-06 13:54 starseeker
- * [r76829] src/libged/CMakeLists.txt, src/libged/cg,
- src/libged/cg/CMakeLists.txt, src/libged/cg/cg.cpp:
- Stub in a cg LIBGED command - no functionality yet, just a basic
- command logic framework.
+ * [r78044] src/archer/CMakeLists.txt, src/archer/archer.bat:
+ Try an alternative with Archer (untested)
-2020-08-18 14:13 starseeker
+2021-01-06 13:25 starseeker
- * [r76828] src/libged/brep/brep.cpp:
- C++ file, not C file - fix footer
+ * [r78043] src/mged/CMakeLists.txt, src/rtwizard/CMakeLists.txt,
+ src/rtwizard/main.c, src/rtwizard/rtwizard.bat:
+ Set up similar wrapper for rtwizard to adjust working dir when
+ invoking from MSI shortcut. Archer is trickier, since all of its
+ option parsing is in Tcl/Tk right now.
-2020-08-18 13:16 starseeker
+2021-01-06 09:40 brlcad
- * [r76827] src/rt/viewweight.c:
- Very simple cap on the number of times this message is printed.
- bu_logging from the hit callback is generally not a great idea to
- begin with, since you'll have lots of threads simultaneously
- hitting bu_log most of the time... even if that's not an issue
- for the application it's fairly hard to interpret. Better would
- be to store any information we want to report about this issue
- per-thread and then summarize it at the dispatcher level...
+ * [r78042] TODO:
+ gdiff appears to be misbehaving
-2020-08-18 12:21 starseeker
+2021-01-05 20:50 starseeker
- * [r76826] src/mged/mged.c:
- Throttle the event checking a bit to try and avoid an all-out hot
- spin waiting for output.
+ * [r78041] src/mged/CMakeLists.txt, src/mged/mged.bat:
+ Grr. Windows bat file test.
-2020-08-18 11:56 starseeker
+2021-01-05 20:04 starseeker
- * [r76825] src/mged/mged.c:
- When we're in command execution mode, clear out all the Tcl
- events (if any) from subprocesses *before* we exit, in order to
- print their output.
+ * [r78040] src/mged/mged.c:
+ Experiment with an option to set working dir.
+
+ Trying to get the MSI installer from WiX to change the working
+ directory for shortcuts has proven challenging. An alternative
+ is to allow for the possibility of defining a shortcut that tells
+ the application itself to change the working directory - untested
+ as yet.
-2020-08-18 11:29 starseeker
+2021-01-04 14:55 starseeker
- * [r76824] src/libged/ged_util.c:
- STDOUT, not STDERR - already have a STDERR handler
+ * [r78039] CMakeLists.txt:
+ Need this variable set in CMake, not just the header. Noticed it
+ because the Windows icons for WiX weren't being set.
-2020-08-18 11:07 starseeker
+2020-12-20 14:58 starseeker
- * [r76823] src/libged/exec_mapping.cpp,
- src/libged/tests/plugins.cpp:
- Rather than bu_log at the library level, let the calling code
- decide what to do about validating argv[0] with the information
- returned by the GED_UNKNOWN flag and argv[0] value itself.
+ * [r78037] src/libged/brep/CMakeLists.txt:
+ add distclean rule
-2020-08-18 10:57 starseeker
+2020-12-19 22:11 starseeker
- * [r76822] src/libged/exec_mapping.cpp,
- src/libged/tests/plugins.cpp:
- Rather than fatally aborting when we find a command mismatch on a
- named ged function call, warn and override with the "root"
- function name. This should work in every case the command would
- originally have worked - the only failure would be if there is a
- command that the plugin is supposed to be providing that didn't
- get set up, which would be a legitimate bug on libged's end.
+ * [r78031] regress/mged/mged.sh:
+ MGED now recognizes that a completely empty file isn't a valid .g
+ file - remove this part of mged.sh
-2020-08-18 02:28 starseeker
+2020-12-19 18:00 starseeker
- * [r76821] src/libged/ged_util.c:
- Listen for both stdout and stderr output from the rt tools - this
- gets the summary in MGED GUI mode, but we still don't get
- anything at all in classic mode. Not sure why yet.
- tclcad_create_io_handler is being called, but
- Tcl_CreateFileHandler is never triggered so
- _ged_rt_output_handler isn't called. Apparently this was also
- true in 72189, as the output is printed without the callback ever
- being triggered there as well...
+ * [r78028] NEWS, src/mged/set.c:
+ Wrong level of indirection passing the struct to parsing was
+ breaking reporting for MGED 'vars' command. Also report rc
+ setting.
-2020-08-17 17:47 starseeker
+2020-12-19 17:41 starseeker
- * [r76820] misc/repoconv/github_ci_actions.yml:
- Didn't mean to delete that comment
+ * [r78027] src/libdm/dm_init.cpp, src/libged/ged_init.cpp:
+ More memory cleanup
-2020-08-17 17:34 starseeker
+2020-12-19 17:34 starseeker
- * [r76819] misc/repoconv/github_ci_actions.yml:
- Not ready for Ninja yet, go with just a make all/check rather
- than distcheck
+ * [r78026] src/libtclcad/auto_path.c:
+ Do some memory cleanup
-2020-08-17 15:19 starseeker
+2020-12-19 15:35 starseeker
- * [r76818] include/analyze/pnts.h, src/libanalyze/pnt_inside.c,
- src/libanalyze/polygonizer.c, src/libged/pnts/pnts.cpp,
- src/libged/pnts_util.c, src/libged/pnts_util.h:
- Expose point inside/outside testing using the raytracer via the
- pnts command. Not documented yet, needs more testing.
+ * [r78025] TODO:
+ Note history found to date to serve as starting point for further
+ debugging.
-2020-08-17 13:08 starseeker
+2020-12-19 15:32 starseeker
- * [r76817] include/analyze/pnts.h, src/libanalyze/CMakeLists.txt,
- src/libanalyze/pnt_inside.c, src/libanalyze/polygonizer.c:
- Extract the inside/outside pnt test with the raytracer created
- for the polygonizer.
+ * [r78024] NEWS, TODO, src/mged/mged.c:
+ If a file doesn't report a valid .g header, don't keep opening it
+ in MGED.
-2020-08-17 12:50 starseeker
+2020-12-19 15:14 starseeker
- * [r76816] include/bu/magic.h:
- revert r76808 - github runner didn't like that approach either...
+ * [r78023] TODO:
+ Ditto bu_brlcad_root - commits need review, but should be removed
+ now.
-2020-08-17 03:43 starseeker
+2020-12-19 15:11 starseeker
- * [r76815] src/other/CMakeLists.txt, src/other/gdal.dist,
- src/other/gdal/CMakeLists.txt, src/other/gdal/alg/CMakeLists.txt,
- src/other/gdal/alg/gdalpansharpen.cpp,
- src/other/gdal/apps/CMakeLists.txt,
- src/other/gdal/frmts/CMakeLists.txt,
- src/other/gdal/frmts/EnvisatFile.c,
- src/other/gdal/frmts/EnvisatFile.h,
- src/other/gdal/frmts/IdrisiDataset.cpp,
- src/other/gdal/frmts/IngrTypes.cpp,
- src/other/gdal/frmts/IngrTypes.h,
- src/other/gdal/frmts/IntergraphBand.cpp,
- src/other/gdal/frmts/IntergraphBand.h,
- src/other/gdal/frmts/IntergraphDataset.cpp,
- src/other/gdal/frmts/IntergraphDataset.h,
- src/other/gdal/frmts/JpegHelper.cpp,
- src/other/gdal/frmts/JpegHelper.h, src/other/gdal/frmts/README,
- src/other/gdal/frmts/TODO_Projections.txt,
- src/other/gdal/frmts/aaigrid,
- src/other/gdal/frmts/aaigrid/CMakeLists.txt,
- src/other/gdal/frmts/aaigrid/aaigriddataset.cpp,
- src/other/gdal/frmts/aaigrid/aaigriddataset.h,
- src/other/gdal/frmts/aaigriddataset.cpp,
- src/other/gdal/frmts/aaigriddataset.h,
- src/other/gdal/frmts/ace2dataset.cpp, src/other/gdal/frmts/adrg,
- src/other/gdal/frmts/adrg/CMakeLists.txt,
- src/other/gdal/frmts/adrg/adrgdataset.cpp,
- src/other/gdal/frmts/adrg/srpdataset.cpp,
- src/other/gdal/frmts/adrgdataset.cpp,
- src/other/gdal/frmts/adsrange.cpp,
- src/other/gdal/frmts/adsrange.hpp,
- src/other/gdal/frmts/aigccitt.c,
- src/other/gdal/frmts/aigdataset.cpp,
- src/other/gdal/frmts/aigopen.c, src/other/gdal/frmts/aigrid,
- src/other/gdal/frmts/aigrid.h,
- src/other/gdal/frmts/aigrid/CMakeLists.txt,
- src/other/gdal/frmts/aigrid/aigccitt.c,
- src/other/gdal/frmts/aigrid/aigdataset.cpp,
- src/other/gdal/frmts/aigrid/aigopen.c,
- src/other/gdal/frmts/aigrid/aigrid.h,
- src/other/gdal/frmts/aigrid/gridlib.c,
- src/other/gdal/frmts/airsar,
- src/other/gdal/frmts/airsar/CMakeLists.txt,
- src/other/gdal/frmts/airsar/airsardataset.cpp,
- src/other/gdal/frmts/airsardataset.cpp, src/other/gdal/frmts/arg,
- src/other/gdal/frmts/arg/CMakeLists.txt,
- src/other/gdal/frmts/arg/argdataset.cpp,
- src/other/gdal/frmts/argdataset.cpp,
- src/other/gdal/frmts/atlsci_spheroid.cpp,
- src/other/gdal/frmts/atlsci_spheroid.h, src/other/gdal/frmts/blx,
- src/other/gdal/frmts/blx.c, src/other/gdal/frmts/blx.h,
- src/other/gdal/frmts/blx/CMakeLists.txt,
- src/other/gdal/frmts/blx/blx.c, src/other/gdal/frmts/blx/blx.h,
- src/other/gdal/frmts/blx/blxdataset.cpp,
- src/other/gdal/frmts/blxdataset.cpp, src/other/gdal/frmts/bmp,
- src/other/gdal/frmts/bmp/CMakeLists.txt,
- src/other/gdal/frmts/bmp/bmpdataset.cpp,
- src/other/gdal/frmts/bmpdataset.cpp, src/other/gdal/frmts/bpg,
- src/other/gdal/frmts/bpg/bpgdataset.cpp,
- src/other/gdal/frmts/bpgdataset.cpp,
- src/other/gdal/frmts/btdataset.cpp, src/other/gdal/frmts/cals,
- src/other/gdal/frmts/cals/CMakeLists.txt,
- src/other/gdal/frmts/cals/calsdataset.cpp,
- src/other/gdal/frmts/calsdataset.cpp, src/other/gdal/frmts/ceos,
- src/other/gdal/frmts/ceos.c, src/other/gdal/frmts/ceos.h,
- src/other/gdal/frmts/ceos/CMakeLists.txt,
- src/other/gdal/frmts/ceos/ceosdataset.cpp,
- src/other/gdal/frmts/ceos/ceosopen.c,
- src/other/gdal/frmts/ceos/ceosopen.h, src/other/gdal/frmts/ceos2,
- src/other/gdal/frmts/ceos2/CMakeLists.txt,
- src/other/gdal/frmts/ceos2/ceos.c,
- src/other/gdal/frmts/ceos2/ceos.h,
- src/other/gdal/frmts/ceos2/ceosrecipe.c,
- src/other/gdal/frmts/ceos2/ceossar.c,
- src/other/gdal/frmts/ceos2/link.c,
- src/other/gdal/frmts/ceos2/sar_ceosdataset.cpp,
- src/other/gdal/frmts/ceosdataset.cpp,
- src/other/gdal/frmts/ceosopen.c, src/other/gdal/frmts/ceosopen.h,
- src/other/gdal/frmts/ceosrecipe.c,
- src/other/gdal/frmts/ceossar.c, src/other/gdal/frmts/coasp,
- src/other/gdal/frmts/coasp/CMakeLists.txt,
- src/other/gdal/frmts/coasp/coasp_dataset.cpp,
- src/other/gdal/frmts/coasp_dataset.cpp,
- src/other/gdal/frmts/cosar,
- src/other/gdal/frmts/cosar/CMakeLists.txt,
- src/other/gdal/frmts/cosar/cosar_dataset.cpp,
- src/other/gdal/frmts/cosar_dataset.cpp,
- src/other/gdal/frmts/cpgdataset.cpp,
- src/other/gdal/frmts/ctable2dataset.cpp,
- src/other/gdal/frmts/ctg,
- src/other/gdal/frmts/ctg/CMakeLists.txt,
- src/other/gdal/frmts/ctg/ctgdataset.cpp,
- src/other/gdal/frmts/ctgdataset.cpp,
- src/other/gdal/frmts/ddffield.cpp,
- src/other/gdal/frmts/ddffielddefn.cpp,
- src/other/gdal/frmts/ddfmodule.cpp,
- src/other/gdal/frmts/ddfrecord.cpp,
- src/other/gdal/frmts/ddfsubfielddefn.cpp,
- src/other/gdal/frmts/ddfutils.cpp, src/other/gdal/frmts/derived,
- src/other/gdal/frmts/derived/CMakeLists.txt,
- src/other/gdal/frmts/derived/deriveddataset.cpp,
- src/other/gdal/frmts/derived/derivedlist.c,
- src/other/gdal/frmts/derived/derivedlist.h,
- src/other/gdal/frmts/deriveddataset.cpp,
- src/other/gdal/frmts/derivedlist.c,
- src/other/gdal/frmts/derivedlist.h, src/other/gdal/frmts/dimap,
- src/other/gdal/frmts/dimap/CMakeLists.txt,
- src/other/gdal/frmts/dimap/dimapdataset.cpp,
- src/other/gdal/frmts/dimapdataset.cpp,
- src/other/gdal/frmts/dipxdataset.cpp,
- src/other/gdal/frmts/doq1dataset.cpp,
- src/other/gdal/frmts/doq2dataset.cpp, src/other/gdal/frmts/dted,
- src/other/gdal/frmts/dted/CMakeLists.txt,
- src/other/gdal/frmts/dted/dted_api.c,
- src/other/gdal/frmts/dted/dted_api.h,
- src/other/gdal/frmts/dted/dted_create.c,
- src/other/gdal/frmts/dted/dted_ptstream.c,
- src/other/gdal/frmts/dted/dteddataset.cpp,
- src/other/gdal/frmts/dted_api.c, src/other/gdal/frmts/dted_api.h,
- src/other/gdal/frmts/dted_create.c,
- src/other/gdal/frmts/dted_ptstream.c,
- src/other/gdal/frmts/dteddataset.cpp,
- src/other/gdal/frmts/e00compr.h, src/other/gdal/frmts/e00grid,
- src/other/gdal/frmts/e00grid/CMakeLists.txt,
- src/other/gdal/frmts/e00grid/e00compr.h,
- src/other/gdal/frmts/e00grid/e00griddataset.cpp,
- src/other/gdal/frmts/e00grid/e00read.c,
- src/other/gdal/frmts/e00griddataset.cpp,
- src/other/gdal/frmts/e00read.c,
- src/other/gdal/frmts/ecrgtocdataset.cpp,
- src/other/gdal/frmts/ehdrdataset.cpp,
- src/other/gdal/frmts/ehdrdataset.h,
- src/other/gdal/frmts/eirdataset.cpp, src/other/gdal/frmts/elas,
- src/other/gdal/frmts/elas/CMakeLists.txt,
- src/other/gdal/frmts/elas/elasdataset.cpp,
- src/other/gdal/frmts/elasdataset.cpp,
- src/other/gdal/frmts/envidataset.cpp,
- src/other/gdal/frmts/envidataset.h, src/other/gdal/frmts/envisat,
- src/other/gdal/frmts/envisat/CMakeLists.txt,
- src/other/gdal/frmts/envisat/EnvisatFile.c,
- src/other/gdal/frmts/envisat/EnvisatFile.h,
- src/other/gdal/frmts/envisat/adsrange.cpp,
- src/other/gdal/frmts/envisat/adsrange.hpp,
- src/other/gdal/frmts/envisat/envisatdataset.cpp,
- src/other/gdal/frmts/envisat/records.c,
- src/other/gdal/frmts/envisat/records.h,
- src/other/gdal/frmts/envisat/timedelta.hpp,
- src/other/gdal/frmts/envisat/unwrapgcps.cpp,
- src/other/gdal/frmts/envisatdataset.cpp,
- src/other/gdal/frmts/ers,
- src/other/gdal/frmts/ers/CMakeLists.txt,
- src/other/gdal/frmts/ers/ersdataset.cpp,
- src/other/gdal/frmts/ers/ershdrnode.cpp,
- src/other/gdal/frmts/ers/ershdrnode.h,
- src/other/gdal/frmts/ersdataset.cpp,
- src/other/gdal/frmts/ershdrnode.cpp,
- src/other/gdal/frmts/ershdrnode.h,
- src/other/gdal/frmts/fastdataset.cpp, src/other/gdal/frmts/fit,
- src/other/gdal/frmts/fit.cpp, src/other/gdal/frmts/fit.h,
- src/other/gdal/frmts/fit/CMakeLists.txt,
- src/other/gdal/frmts/fit/fit.cpp, src/other/gdal/frmts/fit/fit.h,
- src/other/gdal/frmts/fit/fitdataset.cpp,
- src/other/gdal/frmts/fit/gstEndian.h,
- src/other/gdal/frmts/fit/gstTypes.h,
- src/other/gdal/frmts/fitdataset.cpp,
- src/other/gdal/frmts/fujibasdataset.cpp,
- src/other/gdal/frmts/gdal_vrt.h,
- src/other/gdal/frmts/genbindataset.cpp,
- src/other/gdal/frmts/geotiff.cpp, src/other/gdal/frmts/gff,
- src/other/gdal/frmts/gff/CMakeLists.txt,
- src/other/gdal/frmts/gff/gff_dataset.cpp,
- src/other/gdal/frmts/gff_dataset.cpp,
- src/other/gdal/frmts/grcdataset.cpp,
- src/other/gdal/frmts/grddataset.cpp,
- src/other/gdal/frmts/gridlib.c,
- src/other/gdal/frmts/gs7bgdataset.cpp,
- src/other/gdal/frmts/gsagdataset.cpp,
- src/other/gdal/frmts/gsbgdataset.cpp,
- src/other/gdal/frmts/gscdataset.cpp, src/other/gdal/frmts/gsg,
- src/other/gdal/frmts/gsg/CMakeLists.txt,
- src/other/gdal/frmts/gsg/gs7bgdataset.cpp,
- src/other/gdal/frmts/gsg/gsagdataset.cpp,
- src/other/gdal/frmts/gsg/gsbgdataset.cpp,
@@ Diff output truncated at 100000 characters. @@
This was sent by the SourceForge.net collaborative development platform, the
world's largest Open Source development site.
_______________________________________________
BRL-CAD Source Commits mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/brlcad-commits
