01:31.59 |
``Erik |
huh, gamekit got to 80% before
failing |
03:52.58 |
*** join/#brlcad louipc
(~louipc@archlinux/trusteduser/louipc) |
04:14.28 |
*** join/#brlcad _yukonbob
(~svs@S010600235a187d92.ok.shawcable.net) |
07:12.46 |
*** join/#brlcad d_rossberg
(~rossberg@BZ.BZFLAG.BZ) |
08:56.03 |
*** join/#brlcad mafm_
(~mafm@83.37.7.245) |
09:46.32 |
*** join/#brlcad merzo
(~merzo@smartbussiness.mobicom.net.ua) |
12:06.30 |
starseeker |
``Erik: what was the failure? |
12:08.17 |
starseeker |
(if nothing else, gamekit is interesting for
its implementation of a reader for .blend files |
12:37.28 |
d-lo |
Mernin |
13:09.31 |
*** join/#brlcad Zaebos
(~irc@pd95b7f5e.dip0.t-ipconnect.de) |
13:20.02 |
``Erik |
missing config.h in both the unix and xcode
variants |
13:20.50 |
``Erik |
thought ogre had a plugin to
export their .mesh/.skeleton models from blender (among
others) |
13:30.43 |
CIA-42 |
BRL-CAD: 03davidloman * r40168
10/rt^3/trunk/src/libPkgCpp/: Adding libpkgcpp.so to
svn:include |
13:31.07 |
d-lo |
doh! Not quite awake on that one. |
13:35.41 |
d-lo |
question for *nix guys: is /etc/environment
universal among all shells? e.g. if I set a PATH var there, it
will show up in all shells (bash, csh, tch, etc) ? |
13:35.53 |
starseeker |
Blender has a plugin to export OGRE meshes,
but that's GPL |
13:37.29 |
``Erik |
um, /usr/bin/env <progIwant> |
13:37.43 |
``Erik |
#!/usr/bin/env python for
example |
13:38.14 |
``Erik |
oh, wait, misunderstood |
13:38.38 |
``Erik |
no, /etc/environment doesn't soudn familiar
and isn't on mac or fbsd |
13:38.45 |
starseeker |
I'd be surprised if all the shells agreed on
something like that |
13:39.22 |
``Erik |
csh and sh have totally different notions on
how to even set variables, so I don't think there's any way tto be
universal |
13:40.07 |
``Erik |
most sh's are one or the other, and most of
the known universe uses sh (I had never even heard of anyone
actually using csh/tcsh until I came here) |
13:40.24 |
starseeker |
``Erik: the blend "format" is a scary beast,
and I had considered trying to talk the Blender guys into LGPLing
the parser aspects of their code until I found out more of the
details of how they came to be GPL |
13:40.42 |
``Erik |
afaik, the blend format is a swizzled memory
image, not a format :D |
13:40.50 |
starseeker |
bingo |
13:41.13 |
starseeker |
fortunately, it sounds like the gamekit folk
have done a lot of the hard work of figuring out how to read that
beast |
13:41.33 |
``Erik |
until blender changes again? or do they link
to blender libs? |
13:41.45 |
starseeker |
probably until blender changes again |
13:42.03 |
starseeker |
but that's as good as it could get for someone
like us, who won't be linking to the libs anytime soon |
13:42.08 |
``Erik |
what's the issue with a gpl'd exporter for
blender? |
13:42.33 |
starseeker |
hmm? there isn't an issue with Blender - I'm
thinking about code we could use to base a blend-g convertor
on |
13:42.48 |
``Erik |
if you're making a self contained thing like a
game, you crank the handle when preparing the resources and ship
the results... |
13:42.51 |
``Erik |
hrmmm |
13:43.27 |
``Erik |
a python asc exporter/importer for blender
would probably be the least painful |
13:43.52 |
starseeker |
from their description, the gamekit guys have
more or less done an independent reader for the actual .blend
files |
13:44.14 |
starseeker |
``Erik: yeah, that's one option |
13:44.39 |
starseeker |
but you need Blender to pull it off, which
kinda defeats the point of a convertor |
13:44.54 |
``Erik |
proe-g needs the proe-libs to pull it
off |
13:45.06 |
``Erik |
q requires it's libs, too |
13:45.24 |
starseeker |
right, because no one has ever reverse
engineered the proe format and told the world about it |
13:46.03 |
starseeker |
from what brlcad has said in the past, that
would be a murderously difficult task |
13:46.06 |
``Erik |
because it's almost as scary and change prone
as blenders :D *duck* |
13:46.11 |
starseeker |
right |
13:46.47 |
starseeker |
they have no particular incentive to make it
easy for anyone else to read their files |
13:47.21 |
starseeker |
unfortunately, they also have the same lack of
incentive to export ALL their information (parametrics, metadata,
etc) in something like STEP |
13:47.22 |
``Erik |
pro/e has disencentive, blender's opinion is
probably "just install blender and write a python plugin" |
13:47.29 |
starseeker |
right |
13:47.35 |
``Erik |
disincentive |
13:48.45 |
starseeker |
we'll never have the resources to do anything
about reverse-engineering binary cad formats, but if open source
cad does start to take off we might see other projects start to try
and tackle it |
13:49.12 |
``Erik |
as with most advancements, I think we'll see
it happen in games, first |
13:49.23 |
``Erik |
e.g. gamekit :D *duck* |
13:49.32 |
starseeker |
um, doubt it - CAD models are overkill for
proe and friends |
13:49.34 |
starseeker |
heh |
13:49.50 |
``Erik |
well, for .blend at least... pro/e is very
niche |
13:49.57 |
starseeker |
nods |
13:50.19 |
``Erik |
if you have pro/e or uni or whatever files you
care about, then you probably have the software and can click
'export' |
13:50.40 |
starseeker |
nowadays that's true |
13:50.59 |
starseeker |
'cept maybe in countries where a pro/e license
is half a year's wages |
13:51.07 |
d-lo |
and if you dont have the software, there's
always bittorrents :P |
13:51.12 |
``Erik |
the bearded hacker chugging mountain dew in
the campus computer lab probably doesn't have access to any of
those, and doesn't care |
13:51.19 |
starseeker |
d-lo: +1 cynical |
13:51.25 |
d-lo |
:D |
13:51.26 |
``Erik |
she'd be more interested in writing a new
cad |
13:51.42 |
``Erik |
(yes. bearded female hackers.) |
13:51.54 |
d-lo |
I just threw up a little |
13:52.00 |
``Erik |
your welcome :D |
13:52.20 |
starseeker |
ah, but if BRL-CAD has solved the CAD part of
the equation, the lack of Pro/E import will be a glaring
weakness |
13:52.54 |
starseeker |
although I suppose SoidWorks might actually be
of more immediate interest - what is the most common online CAD
format, anyway? |
13:54.17 |
``Erik |
obj *cough* |
13:54.25 |
starseeker |
heh |
13:54.36 |
starseeker |
I suppose these guys may have some idea:
http://www.3dcontentcentral.com/default.aspx |
13:54.48 |
``Erik |
.max might be big, too |
13:55.11 |
starseeker |
pity it's almost certain none of those models
are licensed so we can use 'em... |
13:55.23 |
starseeker |
.max isn't a cad format per-say though, is
it? |
13:55.25 |
``Erik |
I have a parser for the old 3ds format, but it
works with the dos version, not the 'new' version when they went to
winderz |
13:55.39 |
``Erik |
it's visually oriented, just like
obj |
13:56.17 |
``Erik |
I think I showed ya my importer, the
half-C/half-scheme thing? |
13:56.38 |
starseeker |
yeah, scary :-) |
13:57.01 |
``Erik |
what? I used C to do the things C is really
good at and scheme to do the things scheme is really good at
:D |
13:57.05 |
``Erik |
<-- thought it was keen |
13:57.39 |
starseeker |
oh, quite keen - but still scary :-) |
13:57.51 |
brlcad |
starseeker: got a reply back from that 2D cad
dev -- he's going to remove his gpl dep and relicense as lgpl, but
doesn't have much time to help integrate |
13:57.57 |
starseeker |
arrgh - yeah, thought so: You may not i)
distribute Data as part of any service or ii) copy or post any Data
on any Internet site or iii) broadcast Data in any media or iv) use
the Data in a manner that is competitive with this 3D
ContentCentral service. |
13:58.04 |
starseeker |
brlcad: sweeet! |
13:58.05 |
*** join/#brlcad Yoshi47
(~jan@64.235.102.210) |
13:58.14 |
starseeker |
brlcad: nice work! |
13:58.24 |
starseeker |
what was that link again? |
13:59.12 |
``Erik |
hm, 'bitrock installbuilder' |
13:59.47 |
``Erik |
whoa, clanlib is still alive |
14:00.49 |
starseeker |
brlcad: nevermind, found it |
14:02.59 |
brlcad |
``Erik: alive and well .. their guys work on
it pretty consistently |
14:03.10 |
brlcad |
hearts
clanlib |
14:03.19 |
starseeker |
oh yeah - Dime is GPL, from Coin3d |
14:05.42 |
d-lo |
looks @ clanlib...... purty
neat! |
14:07.07 |
starseeker |
brlcad: how does his constraint solver tie in
with our libpc stuff? |
14:08.24 |
brlcad |
I'd treat it as an implementation
detail |
14:08.35 |
brlcad |
no telling how much of his solver is 2D
specific |
14:08.50 |
starseeker |
nods |
14:10.14 |
starseeker |
fair enough - at some point I need to study
both our libpc stuff and varkon's approach to the problem - it's
not a topic I really feel comfortable with |
14:11.16 |
starseeker |
hah, coo |
14:11.17 |
starseeker |
l |
14:11.25 |
starseeker |
brlcad: check this out: http://code.google.com/p/osifont/ |
14:11.44 |
starseeker |
unfortunately it's currently GPL... |
14:13.25 |
``Erik |
a long time ago, a friend and I set out to
make a worms2 clone for linux using clanlib, couldn't get the
performance up to snuff on our 120 and 166 mhz machines :/ probably
naivete in composition techniques |
14:14.44 |
brlcad |
starseeker: constraint solving is a tricky bit
-- kind of like implementing get_closest_point() or
rt_poly_roots() |
14:15.02 |
brlcad |
but similarly, there are some basic well-known
methods that will do the job "good enough" |
14:15.26 |
starseeker |
nods |
14:52.04 |
brlcad |
https://www.ohloh.net/languages/compare?measure=projects&percent=true&l0=autoconf&l1=automake&l2=cmake&l3=-1&l4=make&l5=-1&commit=Update |
14:52.59 |
``Erik |
wonders what systems are
missing O.o |
14:53.06 |
``Erik |
wonders who'll join his lunch
posse |
14:54.27 |
brlcad |
interesting that autotools have been on the
rise for the past 6 months |
14:57.25 |
d-lo |
well, they're all on the rise :) |
14:58.28 |
brlcad |
yeah, but cmake's still pretty
linear |
15:02.05 |
brlcad |
my guess is that there are other unlisted
build systems starting to bottom out, and the increase is just a
usual influx of new projects getting added |
15:02.14 |
brlcad |
systems like imake and .. cake |
15:02.42 |
brlcad |
ant would have been an interesting comparison,
but I guess it gets lost in the xml aggregate |
16:36.16 |
d-lo |
bah, if I have a bunch of files that have ;1
appended to the end of the file names, how can I bulk get rid of
them? |
16:37.50 |
d-lo |
I have tried: find ./ -name "*;1" | sed
's/\;1//g' |
16:38.07 |
d-lo |
but that only SHOWED the changes and didn't
apply them to the file names. |
16:38.24 |
d-lo |
I cannot see in the sed man page what switch I
use to apply the changes :/ |
16:53.43 |
brlcad |
you need a lil more scripting foo
magic |
16:54.32 |
d-lo |
I got it to work by piping it thru rename
instead of sed. |
16:54.40 |
brlcad |
something like: for i in `find . -name "*;1"`
; do cp $i `echo $i | sed 's/\;1//g'` ; done |
16:55.15 |
d-lo |
forgot that ubuntu ships with rename |
16:55.23 |
brlcad |
yeah, linuxy |
16:55.27 |
brlcad |
common pattern at least |
16:55.38 |
brlcad |
goes to the hardware store to
play |
16:55.45 |
d-lo |
whatcha buyin? |
17:01.19 |
``Erik |
sed only alters the text, it has no ability to
'apply' anything |
17:28.04 |
kanzure |
brlcad: do you know about
vehicleforge.mil? |
17:28.30 |
kanzure |
and if not, i'd like you to be involved in a
proposal i'm crafting |
17:35.15 |
``Erik |
403 O.o |
17:47.46 |
CIA-42 |
BRL-CAD: 03erikgreenwald * r40169
10/isst/trunk/sdl/event.c: scale float/sink to scene size |
18:01.19 |
kanzure |
it doesn't exist yet |
18:01.23 |
kanzure |
https://www.fbo.gov/utils/view?id=11a895334e76707406e3b78c918357cd |
18:01.34 |
kanzure |
this is pretty much what my 'skdb' project is
(apt-get for hardware, and such) |
18:37.52 |
*** join/#brlcad _psilva
(~psilva@static-96-255-52-7.washdc.fios.verizon.net) |
20:03.43 |
starseeker |
turns mildly red as he sees
how configure.ac did its summary printout - I think perhaps I
overengineered the CMake one a bit |
20:04.11 |
starseeker |
it should be robust though... |
20:39.33 |
*** join/#brlcad R0b0t1
(~Enigma@unaffiliated/r0b0t1) |
22:32.56 |
starseeker |
``Erik: yeah, it was my CFLAGS line in the
quick, hacky src/other/libterm CMakeLists.txt that did it |
22:35.03 |
starseeker |
getting some errors during
compile... |
22:35.06 |
starseeker |
bah |
22:47.38 |
CIA-42 |
BRL-CAD: 03starseeker * r40170
10/brlcad/branches/cmake/src/other/libtermlib/CMakeLists.txt: Try
this approach to the compile flags for libterm - Xcode doesn't like
the old way. |
22:48.37 |
starseeker |
hmm - ``Erik, looks like I need more
dependency info for Xcode - the first pass has failures to build,
but the second succeeds |
22:56.26 |
starseeker |
ah, I might have known - conf/COUNT and
conf/DATE |
23:06.55 |
CIA-42 |
BRL-CAD: 03starseeker * r40171
10/brlcad/branches/cmake/CMakeLists.txt: Xcode exposed the need for
explicit dependency specifications for include/conf/COUNT and
include/conf/DATE - Xcode build now succeeds. |
23:09.37 |
kanzure |
brlcad: are you around? |
23:13.42 |
``Erik |
thinks he'll have to get
something like
http://craftside.typepad.com/craftside/images/2008/09/17/devil_horn_headband_2.jpg
so'z starseeker knows when he's playing devils advocate
:D |
23:53.06 |
starseeker |
O.o |
23:53.36 |
starseeker |
if you wear that I'm gonna start calling you a
"manager in training :-P" |
23:54.20 |
starseeker |
was worth it though - Xcode building "for
free" is pretty darn cool |
23:57.14 |
``Erik |
don't worry, I probably won't start dressing
like http://travis.kroh.net/archives/000549.jpg |
23:57.17 |
``Erik |
probably |
23:57.59 |
``Erik |
generated a vcproj to throw over on the
winderz machine yet? |
23:58.18 |
starseeker |
is that link safe for work? |
23:58.23 |
starseeker |
no, not yet |
23:58.35 |
starseeker |
trying to cook up some minimal C code for time
deltas |
23:58.37 |
``Erik |
yeah, it's a pic from an open source
convention |
23:58.53 |
``Erik |
um, gettimeofday, cook the millisecs, then
expand it out using ctime |
23:59.12 |
``Erik |
though gettimeofday may not be on
winderz |
23:59.22 |
starseeker |
that's probably finer than we need
anyhow |
23:59.42 |
``Erik |
true |
23:59.44 |
starseeker |
as far as configure and build timing goes,
1sec ~= 0 |
23:59.45 |
``Erik |
time(3) ? |
23:59.55 |
starseeker |
yeah, that's what I'm targeting |
23:59.58 |
``Erik |
it's c99 |