00:15.54 |
CIA-28 |
BRL-CAD: 03r_weiss * r47849
10/brlcad/trunk/src/librt/primitives/nmg/nmg_inter.c: Update to
function nmg_crackshells in file 'nmg_inter.c'. Improved bounding
box checking to determine which faces will be intersected. This
improves the speed of boolean operations on nmg objects. |
00:18.39 |
CIA-28 |
BRL-CAD: 03r_weiss * r47850
10/brlcad/trunk/include/vmath.h: Updated include file 'vmath.h'
adding the macro 'V3RPP_DISJOINT_TOL'. |
00:24.06 |
*** join/#brlcad cjdevlin
(~devlin@d118-75-244-176.try.wideopenwest.com) |
00:50.48 |
CIA-28 |
BRL-CAD: 03brlcad * r47851 10/brlcad/trunk/
(13 files in 4 dirs): (log message trimmed) |
00:50.48 |
CIA-28 |
BRL-CAD: avoid the assumption that tol is a
bn_tol. since they all just call the |
00:50.48 |
CIA-28 |
BRL-CAD: distance tolerance, leave it up to
the caller to provide the value so as an API |
00:50.48 |
CIA-28 |
BRL-CAD: call, it only needs to be a tolerance
value instead of a struct. propagate |
00:50.48 |
CIA-28 |
BRL-CAD: accordingly throughout nmg. also
leave FIXME notes for three macros that are |
00:50.49 |
CIA-28 |
BRL-CAD: apparently using >= and <=
comparisons with floating point values where the '=' |
00:50.50 |
CIA-28 |
BRL-CAD: case is not guaranteed to give stable
behavior. the testing interval must be an |
00:51.24 |
brlcad |
open interval |
00:55.06 |
CIA-28 |
BRL-CAD: 03brlcad * r47852
10/brlcad/trunk/src/libfb/fb_obj.c: oops, libfb dir slipped through
the cracks. apply bu_cmd() signature change. |
01:05.08 |
CIA-28 |
BRL-CAD: 03brlcad * r47853
10/brlcad/trunk/doc/deprecation.txt: V3*_TOL() macros in vmath used
to assume a structure with a ->dist element, which is kinda
silly from an API standpoint. made them be a simple tolerance
value. |
01:05.23 |
CIA-28 |
BRL-CAD: 03brlcad * r47854
10/brlcad/trunk/src/libbu/Makefile.am: image.c is no more here, he
go home now |
01:09.39 |
CIA-28 |
BRL-CAD: 03brlcad * r47855
10/brlcad/trunk/src/libgcv/CMakeLists.txt: soup.h and
tri_intersect.h missing from decls |
01:10.03 |
brlcad |
../../src/libged/.libs/libged.so: undefined
reference to `icv_image_save_open' |
01:10.27 |
brlcad |
tools build, missing some libs decls |
02:22.42 |
CIA-28 |
BRL-CAD: 03starseeker * r47856
10/brlcad/trunk/doc/docbook/ (CMakeLists.txt
books/en/CMakeLists.txt): |
02:22.42 |
CIA-28 |
BRL-CAD: Well, the downloadable Windows
xsltproc isn't succeeding either, and at least |
02:22.42 |
CIA-28 |
BRL-CAD: some of the failure signature looks
similar. Time to figure out why. Also, |
02:22.42 |
CIA-28 |
BRL-CAD: need some fixes to executable calls -
quite in docbook logic to allow for spaces |
02:22.43 |
CIA-28 |
BRL-CAD: in pathnames (there'll probably be a
lot more of this needed later on...) Still |
02:22.43 |
CIA-28 |
BRL-CAD: need to rework the third party exec
macro somewhat to allow for manual |
02:22.44 |
CIA-28 |
BRL-CAD: specification of the executable
path. |
02:30.18 |
*** join/#brlcad cvds_
(~leila@h111030.upc-h.chello.nl) |
03:56.34 |
*** join/#brlcad packrat
(~packrator@c-98-209-146-133.hsd1.mi.comcast.net) |
04:28.51 |
*** join/#brlcad packrat
(~packrator@c-98-209-146-133.hsd1.mi.comcast.net) |
04:50.02 |
*** join/#brlcad packrat
(~packrator@c-98-209-146-133.hsd1.mi.comcast.net) |
05:17.05 |
*** join/#brlcad packrat
(~packrator@c-98-209-146-133.hsd1.mi.comcast.net) |
05:28.55 |
CIA-28 |
BRL-CAD: 03Sean 07http://brlcad.org * r3250
10/wiki/Talk:Main_Page: Reverted edits by
[[Special:Contributions/94.122.66.247|94.122.66.247]] ([[User
talk:94.122.66.247|Talk]]); changed back to last version by
[[User:Sean|Sean]] |
05:29.03 |
CIA-28 |
BRL-CAD: 03Sean 07http://brlcad.org * r0
10/wiki/Special:Log/block: blocked [[User:94.122.66.247]] with an
expiry time of infinite (anonymous users only, account creation
disabled): Spamming links to external sites |
06:03.15 |
*** join/#brlcad packrat
(~packrator@c-98-209-146-133.hsd1.mi.comcast.net) |
06:22.18 |
CIA-28 |
BRL-CAD: 03brlcad * r47857
10/brlcad/trunk/include/bu.h: |
06:22.19 |
CIA-28 |
BRL-CAD: create corollary macros for
BU_GETSTRUCT(), BU_GETUNION(), and BU_GETTYPE() |
06:22.19 |
CIA-28 |
BRL-CAD: named BU_PUTSTRUCT(), BU_PUTUNION(),
and BU_PUTTYPE() respectively. regardless, |
06:22.19 |
CIA-28 |
BRL-CAD: deprecate at least the struct/union
macros in favor of the single type-agnostic |
06:22.19 |
CIA-28 |
BRL-CAD: macro (though a better name might be
simply BU_GET() and BU_PUT() to avoid the |
06:22.19 |
CIA-28 |
BRL-CAD: klunky TT-name) |
06:31.53 |
CIA-28 |
BRL-CAD: 03brlcad * r47858
10/brlcad/trunk/src/libbu/pool.c: ws indent consistency
cleanup |
06:34.56 |
*** join/#brlcad jordisayol
(~jordisayo@unaffiliated/jordisayol) |
07:04.31 |
CIA-28 |
BRL-CAD: 03brlcad * r47859 10/brlcad/trunk/ (5
files in 3 dirs): (log message trimmed) |
07:04.31 |
CIA-28 |
BRL-CAD: rename the new libbu pool routines to
be consistent with the existing libbu api. |
07:04.31 |
CIA-28 |
BRL-CAD: groups functions are bu_GROUP prefix,
so rename bu_get_elem_from_pool() to |
07:04.31 |
CIA-28 |
BRL-CAD: bu_pool_get() and bu_free_elem_pool()
to bu_pool_put() respectively. rename all |
07:04.31 |
CIA-28 |
BRL-CAD: of the non-public implementation
functions with a non-'bu_' and simpler 'pool_' |
07:04.31 |
CIA-28 |
BRL-CAD: prefix. update existing callers
accordingly. added notes on a few show-stopper |
07:04.32 |
CIA-28 |
BRL-CAD: issues in the implementation, namely
semaphore locking issues and not building |
09:58.08 |
*** join/#brlcad d_rossberg
(~rossberg@BZ.BZFLAG.BZ) |
14:17.46 |
*** join/#brlcad piksi
(~piksi@pi-xi.net) |
14:49.50 |
*** join/#brlcad piksi
(piksi@pi-xi.net) |
15:17.22 |
brlcad |
howdy d_rossberg, ltns |
15:34.00 |
d_rossberg |
writing reports as every year's end |
16:13.27 |
brlcad |
fun fun, sorry :) |
16:13.41 |
CIA-28 |
BRL-CAD: 03brlcad * r47860
10/brlcad/trunk/include/bu.h: |
16:13.41 |
CIA-28 |
BRL-CAD: implement more generalized BU_GET()
and BU_PUT() macros intended to replace |
16:13.41 |
CIA-28 |
BRL-CAD: BU_GETSTRUCT(), BU_GETUNION(), and
BU_GETTYPE(). these versions should provided |
16:13.41 |
CIA-28 |
BRL-CAD: added application security by
guaranteeing zero-init and sanity cleanup when |
16:13.41 |
CIA-28 |
BRL-CAD: released/put. used appropriately,
they become convenient places to hook in a |
16:13.41 |
CIA-28 |
BRL-CAD: pool allocator given the allocations
are intrinsically small and defined as |
16:13.42 |
CIA-28 |
BRL-CAD: dynamic. |
16:36.56 |
CIA-28 |
BRL-CAD: 03starseeker * r47861
10/brlcad/trunk/doc/docbook/ (5 files in 2 dirs): |
16:36.57 |
CIA-28 |
BRL-CAD: Rework the docbook logic again, to
not use the catalog file - as near as I can |
16:36.57 |
CIA-28 |
BRL-CAD: tell, it's just using the catalog
file to all 'abbreviated' paths in a couple |
16:36.57 |
CIA-28 |
BRL-CAD: other style files. Rather than do
that, we can just make them .in files and |
16:36.57 |
CIA-28 |
BRL-CAD: have CMake fill in the paths
directly. Hopefully will put less strain on |
16:36.57 |
CIA-28 |
BRL-CAD: xsltproc... |
16:38.32 |
CIA-28 |
BRL-CAD: 03starseeker * r47862
10/brlcad/trunk/doc/docbook/resources/brlcad/ (9 files): Clear out
files not needed by the CMake build (may have to put 'em back for
autotools, so doing this as a separate commit to be easily reverted
if need be.) |
17:10.13 |
``Erik |
heh, forget DRY, let's do WET (write
everything twice)! :D |
18:14.21 |
CIA-28 |
BRL-CAD: 03r_weiss * r47863
10/brlcad/trunk/src/librt/primitives/nmg/nmg_inter.c: Cleanup of
function 'nmg_crackshells' in file 'nmg_inter.c'. |
18:52.52 |
brlcad |
``Erik: heh |
18:53.22 |
CIA-28 |
BRL-CAD: 03r_weiss * r47864
10/brlcad/trunk/src/librt/primitives/nmg/nmg_inter.c: Update to
function 'nmg_crackshells' to be consistent with macro
changes. |
18:53.57 |
brlcad |
this will be valuable to us:
http://www.itworld.com/software/231515/usenix-dartmouth-expanding-diff-grep-unix-tools |
18:54.18 |
brlcad |
not exactly what I had in mind for GS diffing,
but it would probably work just as well |
19:16.23 |
CIA-28 |
BRL-CAD: 03brlcad * r47865 10/brlcad/trunk/
(194 files in 57 dirs): |
19:16.23 |
CIA-28 |
BRL-CAD: convert instances of the ol'
BU_GETSTRUCT() macro over to the new BU_GET() |
19:16.24 |
CIA-28 |
BRL-CAD: macro. callers now have to specify
the full type (gasp, have to type struct .. |
19:16.24 |
CIA-28 |
BRL-CAD: which they still had to type with
bu_getSTRUCT...) which should help reduce |
19:16.28 |
CIA-28 |
BRL-CAD: obfuscation for new devs (and reduces
3 API macros to 1). |
19:20.54 |
starseeker |
www.cs.dartmouth.edu/reports/TR2011-705.pdf |
19:22.31 |
starseeker |
no code yet, I take it |
19:22.35 |
CIA-28 |
BRL-CAD: 03brlcad * r47866 10/brlcad/trunk/
(34 files in 14 dirs): convert instances of the ol' BU_GETUNION()
macro over to the new BU_GET() macro. callers now have to specify
the full type which should help reduce obfuscation for new
devs. |
19:26.30 |
starseeker |
was plotting to look into
libxdiff at some point... http://www.xmailserver.org/xdiff-lib.html |
19:29.58 |
CIA-28 |
BRL-CAD: 03brlcad * r47867
10/brlcad/trunk/src/ (2 files in 2 dirs): last one, only a couple
BU_GETTYPE() callers |
19:36.07 |
CIA-28 |
BRL-CAD: 03starseeker * r47868
10/brlcad/trunk/doc/docbook/resources/brlcad/ (3 files): go from
file:// to file:/// in xsl files, per suggestion from xml/xslt devs
- unix doesn't seem to care, but on Windows it appears to matter...
reindent as well |
19:50.32 |
CIA-28 |
BRL-CAD: 03r_weiss * r47869
10/brlcad/trunk/src/librt/primitives/nmg/nmg_inter.c: Update to
function 'nmg_crackshells' in file 'nmg_inter.c' to quiet compiler
warnings of signed/unsighed compares. |
20:02.06 |
CIA-28 |
BRL-CAD: 03starseeker * r47870
10/brlcad/trunk/doc/docbook/resources/brlcad/brlcad-man-stylesheet.xsl.in:
fully qualify center-table-print path |
20:40.41 |
CIA-28 |
BRL-CAD: 03r_weiss * r47871
10/brlcad/trunk/src/librt/primitives/nmg/nmg_bool.c: Updated
function 'nmg_bool' in file 'nmg_bool.c'. Removed unnecessary
checking for dangling faces. This will improve the speed of nmg
boolean operations. |
21:22.54 |
CIA-28 |
BRL-CAD: 03starseeker * r47872
10/brlcad/trunk/src/other/xsltproc/ (libxml/CMakeLists.txt
libxslt/CMakeLists.txt): Modules aren't behaving well on Windows
for some reason, and we don't seem to need them anyway -
disable |
21:24.38 |
CIA-28 |
BRL-CAD: 03brlcad * r47873
10/brlcad/trunk/src/remrt/remrt.c:
BU_GETSTRUCT->BU_GET |
21:25.34 |
CIA-28 |
BRL-CAD: 03brlcad * r47874 10/brlcad/trunk/
(doc/deprecation.txt include/bu.h): replacement of
BU_GETSTRUCT/UNION/TYPE with BU_GET are all minimally impacting, so
go ahead and remove them now |
21:41.52 |
CIA-28 |
BRL-CAD: 03starseeker * r47875
10/brlcad/trunk/doc/docbook/ (5 files in 3 dirs): |
21:41.52 |
CIA-28 |
BRL-CAD: Take the 'write a cmake script file'
to its logical conclusion and have a |
21:41.52 |
CIA-28 |
BRL-CAD: template that is configured for each
target. Simplifies the 'in CMakeLists.txt' |
21:41.52 |
CIA-28 |
BRL-CAD: logic and avoids duplication, and if
we need more tweakage on running xsltproc |
21:41.52 |
CIA-28 |
BRL-CAD: or fop it's a lot simpler. |
22:00.11 |
CIA-28 |
BRL-CAD: 03starseeker * r47876
10/brlcad/trunk/src/libged/dg_obj.c: Fix some stuff in ifdef win32
code related to the bu tcl changes |
22:03.11 |
CIA-28 |
BRL-CAD: 03bob1961 * r47877
10/brlcad/trunk/src/libbu/tcl.c: Changed free() back to
Tcl_Free(). |
22:22.34 |
CIA-28 |
BRL-CAD: 03brlcad * r47878 10/brlcad/trunk/ (4
files in 2 dirs): remove the obsolete libbu.3 manual page that has
fallen significantly out of sync with the library API. migrate the
(still useful) documentation bits into the bu.h header as doxygen
comments for future doc purposing. |
22:27.47 |
CIA-28 |
BRL-CAD: 03starseeker * r47879
10/brlcad/trunk/src/other/perplex/CMakeLists.txt: Don't assume the
math library - bad assumption on Windows |
22:56.07 |
CIA-28 |
BRL-CAD: 03brlcad * r47880 10/brlcad/trunk/ (7
files in 7 dirs): prefix the cmd.h defines with BU_ since they are
part of the libbu api |
22:56.33 |
brlcad |
note to self, check on archer drawing
geometry: bu_get_value_by_keyword messages |
23:13.43 |
*** join/#brlcad piksi
(piksi@pi-xi.net) |
23:18.17 |
CIA-28 |
BRL-CAD: 03brlcad * r47881
10/brlcad/trunk/src/libdm/dm-rtgl.c: crash reported by Peter Machon
( muhala ) via sf bu report 3454338 that detected a bad magic
failure on segs from rtgl's hit callback (finding a bu_list instead
of a seg). it's not used so just don't perform that
check. |