| 00:02.37 | starseeker | brlcad: One of the gentoo devs has downloaded the BRL-CAD tarball and will be looking at the autoconf magic :-) |
| 00:03.07 | starseeker | brlcad: Dunno what that means for ebuild inclusion, but at least the interest level is there |
| 00:03.43 | louipc | :D |
| 00:04.46 | louipc | starseeker: http://sourceforge.net/projects/buildconf/ |
| 00:05.59 | yukonbob | starseeker: re: tcl/tk fan -- I guess I am :) ... |
| 00:12.23 | brlcad | ~starseeker++ |
| 00:12.35 | brlcad | awesome |
| 00:13.34 | brlcad | yukonbob: you?! never... |
| 00:13.38 | brlcad | *grin* |
| 00:13.45 | starseeker | brlcad: :-) :-). |
| 00:14.03 | starseeker | brlcad: We should have a betting pool now - how long it takes the dev to cry uncle ;-) |
| 00:14.20 | brlcad | the autoconf magic really just boils down to running autoreconf and a few other verfication items |
| 00:14.38 | brlcad | more details on the buildconf website |
| 00:14.59 | starseeker | I think he's interested in how you're checking if tcl support et. al. are advanced enough |
| 00:15.11 | brlcad | in your particular instance, you were just missing the -Im4 |
| 00:15.16 | brlcad | ahh |
| 00:15.20 | brlcad | that magic |
| 00:15.36 | starseeker | brlcad: Hmm. Maybe there's an option for eautoreconf |
| 00:15.41 | brlcad | probably |
| 00:15.52 | brlcad | it's probably just a wrapper like autogen.sh |
| 00:16.07 | starseeker | Well, running autogen works fine :-) |
| 00:16.10 | brlcad | yep |
| 00:16.16 | brlcad | tis why it exists |
| 00:16.19 | brlcad | it just makes it work |
| 00:16.36 | starseeker | Now, on to the documentation :-) |
| 00:16.43 | brlcad | hence, http://buildconf.brlcad.org/ ;) |
| 00:16.46 | starseeker | where do I find the word versions? |
| 00:18.34 | brlcad | if you need to explain the tcl logic (which you shouldn't probably need to), the tests are as such .. tests for various libs that we're compatible with (up through 8.6), then tests for header availability, then it performs a compilation functionality test where it actually builds a tcl interpreter app and tries to run it |
| 00:19.32 | louipc | haha sweet |
| 00:19.55 | brlcad | if that all works, and it should as it's exactly what we need it to do for mged, then the test passes and system tcl is used. if it fails, the build falls back to using internal (and this of course all presume that the default enable-tcl-build is in effect, i.e. =auto) |
| 00:20.04 | brlcad | if you force it on or off, it overrides the tests |
| 00:21.04 | brlcad | you find the word versions from me, I've been e-mailing them to folks interested just because I don't really want those versions to propagate around the net (given they are readily editable, but with no revision history/controls or easy merge features) |
| 00:21.18 | brlcad | pm me your e-mail and I'll send you the big one |
| 00:23.01 | starseeker | ok :-) |
| 00:27.08 | yukonbob | brlcad: :) |
| 00:27.25 | yukonbob | starseeker: are you working on docbook-ing? |
| 00:27.46 | starseeker | I"m going to try ;-) |
| 00:28.07 | yukonbob | nice -- I've been threatening to do the same, but haven't done anything yet... |
| 00:28.43 | yukonbob | brlcad: I've never heard to ms-word master copies of the docs that generated the pdfs, though -- am I understanding correctly that there are such docs? |
| 00:29.25 | louipc | how would you docbook it? |
| 00:29.38 | louipc | I was thinking of working with the poppler project because they can already convert to html |
| 00:30.09 | louipc | and they do convert to some weird custom xml format so why not make it docbook instead? |
| 00:30.59 | brlcad | yukonbob: yes.. hrm, I'm pretty sure I mentioned that the last time we were talking about this a couple months ago too when that other guy was working on it |
| 00:30.59 | starseeker | I was planning on going more primitive at first - just get the text and images out of word into something sane and go over it by hand |
| 00:31.07 | brlcad | i'd sent him one of the .doc's at the time |
| 00:31.12 | starseeker | good way to learn about the system while I'm at it |
| 00:32.40 | yukonbob | brlcad: if it was mentioned, I missed it or forgot it... no matter, though; now I'm up to speed. |
| 01:15.48 | brlcad | okay, uploaded |
| 03:55.51 | starseeker | For extracting content from the word docs, the wv tools seem to have done OK so far. wvDocBook generates a rather odd looking xml file and (more importantly) the image files, and wvText generates a readable text file (trying Vol I) |
| 03:56.16 | starseeker | Now the hard part - what the toolchain is for writing and processing a docbook file. |
| 03:58.04 | starseeker | brlcad: I'm seeing things like saxon, fop, xalan docbook-xml-dtd, docbook-xsl-stylesheets, etc. - how did you want to approach this? |
| 03:58.16 | starseeker | a lot of the tools I'm seeing recommended need java |
| 03:59.03 | starseeker | I'm also hearing sgml is no longer the direction to take with docbook, xml is the "new" way to go. |
| 03:59.06 | brlcad | ideally the one with the least dependencies, easiest to maintain, easiest to install, and most portable |
| 03:59.16 | starseeker | hrm |
| 03:59.21 | brlcad | yeah, definitely xml docbook or xml docbook lite |
| 03:59.25 | louipc | cool |
| 03:59.38 | brlcad | doesn't matter if it's 5 or 4.5 |
| 03:59.42 | starseeker | Apache FOP seems to be a "big player" in this... |
| 04:00.09 | brlcad | FOP is, but it can also be a pain to set up |
| 04:00.18 | starseeker | Can we make do without it? |
| 04:00.42 | brlcad | ideally something that doesn't require the java vm is easier to manage |
| 04:00.57 | brlcad | but if something else java-based is that much better, then so be it |
| 04:01.05 | starseeker | If we want an XSL-FO processor... let's see what else is out there... |
| 04:01.09 | louipc | starseeker: what are those wv tools? |
| 04:01.26 | brlcad | i've used a handful of the docbook processors, and none of them seemed "great" .. so it's really just finding the one that's the least of a headache |
| 04:01.28 | starseeker | They're used by Abiword, IIRC. |
| 04:02.11 | starseeker | "Tool for conversion of MSWord doc and rtf files to something readable" |
| 04:02.14 | brlcad | ideally there will be a make target like 'make doc' or ./configure --enable-docs that will generate the documentation by running the docbook processor(s) |
| 04:02.20 | starseeker | http://wvware.sourceforge.net/ |
| 04:02.54 | louipc | thanks |
| 04:02.54 | starseeker | There's a wv2 which I think is an update - may not be needed for this case, but we'll see |
| 04:03.07 | brlcad | so something that can be readily set up under freebsd would be nice so it can be tied to the website automatically via automated builds |
| 04:04.21 | starseeker | Does xmlroff ring any bells? |
| 04:05.00 | brlcad | not particularly |
| 04:05.54 | brlcad | output formats of importance are pdf and html, other useful outputs would be txt, rtf/doc, ps, and odf |
| 04:06.29 | starseeker | I'm going to have to take some time and dig - the tool chain for docbook seems both complex and... er... confusing |
| 04:08.19 | brlcad | it is pretty complex |
| 04:10.45 | louipc | oh wow wv is sweet |
| 04:10.58 | louipc | that does it |
| 04:11.05 | starseeker | Does what? |
| 04:13.15 | louipc | extracts the content and images and references the images in the document :D |
| 04:13.24 | starseeker | Yep :-) |
| 04:13.55 | starseeker | the docbook it spits out I think is not what we want though - it preserves all the literal blank lines, at least at a quick glance |
| 04:14.03 | starseeker | great way to get images though |
| 04:14.12 | starseeker | We want human editable XML |
| 04:14.23 | louipc | yeah empty para |
| 04:14.34 | starseeker | And as I'm digging I'm not seeing too many alternatives to Apache FOP :-( |
| 04:15.29 | starseeker | PassiveTeX is apparently pretty well outdated |
| 04:15.38 | louipc | :/ |
| 04:16.32 | starseeker | Lotta commercial ones... |
| 04:16.49 | starseeker | That's a surprise, I would have thought there would be wider tool support for docbook |
| 04:17.48 | starseeker | I guess we can do it, but there'll have to be some rather extensive documentation on the documentation system ;-) |
| 04:19.09 | starseeker | Oh, well. At least extraction was easy! |
| 04:21.08 | starseeker | brlcad: Is LaTeX out due to it's lack of good output options? |
| 04:21.10 | louipc | :D |
| 04:28.12 | brlcad | starseeker: no, not out of the question, nothing really is |
| 04:28.30 | brlcad | just whatever is easiest to set up and automate that gives the output :) |
| 04:30.28 | brlcad | it doesn't even have to "match" the output, it should just look good |
| 07:25.38 | *** join/#brlcad Z80-Boy (n=clock@zux221-122-143.adsl.green.ch) |
| 07:26.38 | *** join/#brlcad elite01 (n=elite01@dslb-088-070-006-080.pools.arcor-ip.net) |
| 07:30.28 | *** join/#brlcad elite01_ (n=elite01@dslb-088-070-021-167.pools.arcor-ip.net) |
| 11:04.40 | *** join/#brlcad elite01_ (n=elite01@dslb-088-070-021-167.pools.arcor-ip.net) |
| 11:06.34 | *** join/#brlcad cad78 (n=438e820f@bz.bzflag.bz) |
| 11:57.55 | Z80-Boy | brlcad: I tried to edit an arbn in the graphics menu editor and it said it's not implemented for this kind of primitive |
| 11:58.27 | Z80-Boy | brlcad: I want to ask the coefficients for the planes, which actually define a normal vector, is it a vector that points out of the body or into the body? |
| 12:31.11 | *** join/#brlcad elite01_ (n=elite01@dslb-088-070-021-167.pools.arcor-ip.net) |
| 13:19.19 | ``Erik | could also use 'jade' like the fbsd project... :D |
| 13:31.17 | brlcad | I liked jade when I was working that area |
| 13:31.32 | Z80-Boy | brlcad: did you figure out yesterday anything on my machine? |
| 13:31.34 | brlcad | it's a bit of a trick to get docbook/xml to work with it, but it works |
| 13:31.50 | brlcad | Z80-Boy: yeah, I compiled and ran the latest with no problems :) |
| 13:31.59 | Z80-Boy | brlcad: remotely through X? |
| 13:32.03 | brlcad | yep |
| 13:32.08 | Z80-Boy | brlcad: latest == which version? |
| 13:32.18 | brlcad | latest == latest sources in cvs |
| 13:32.49 | brlcad | could have made 7.10.0 work, but that tarball had a few files missing and some other minor annoyances not worth even disabling |
| 13:32.53 | Z80-Boy | what did you type to run mged? |
| 13:32.57 | brlcad | s/few files/a file/ |
| 13:33.20 | Z80-Boy | did I supply a corrupted tarball? |
| 13:33.48 | brlcad | no, there's just a file missing from the dist that was fixed that same day, but missing nonetheless |
| 13:33.59 | brlcad | .2 fixed it shortly thereafter |
| 13:34.05 | brlcad | and the fix is simple, just turn off jove |
| 13:34.12 | brlcad | or touch the missing file |
| 13:34.12 | Z80-Boy | what did you type in which dir to run the CVS mged? I want to try too. |
| 13:34.51 | brlcad | well, ideally you should install, otherwise the problem you were probably running into is an LD_LIBRARY_PATH issue since Tcl/Tk only build as shared libraries by default |
| 13:35.05 | brlcad | so you have to install to get the tcl/tk libs, or you have to set your ld_lib path |
| 13:35.08 | Z80-Boy | but you couldn't install you weren't root |
| 13:35.18 | Z80-Boy | so what did you type to verify that mged works? |
| 13:35.21 | brlcad | sure, I just set the ld_lib path |
| 13:35.43 | brlcad | LD_LIBRARY_PATH=src/other/tcl/unix:src/other/tk/unix src/mged/mged |
| 13:36.14 | Z80-Boy | Really, seems to work. |
| 13:36.23 | brlcad | you doubted? |
| 13:36.30 | Z80-Boy | no |
| 13:36.34 | brlcad | :) |
| 13:36.40 | Z80-Boy | how can I do the CVS checkout and compile myself? |
| 13:36.51 | brlcad | ~cadcvs |
| 13:36.51 | ibot | To obtain BRL-CAD from anonymous CVS: cvs -d :pserver:anonymous@brlcad.cvs.sourceforge.net:/cvsroot/brlcad login && cvs -d :pserver:anonymous@brlcad.cvs.sourceforge.net:/cvsroot/brlcad checkout -P brlcad |
| 13:36.57 | Z80-Boy | wow :) |
| 13:37.59 | Z80-Boy | missing information: "and press ENTER". |
| 13:38.11 | Z80-Boy | (to get through the password) |
| 13:38.13 | ``Erik | you need to make sure that you have reasonably recent vrsions of automake, autoconf and libtool and you need to run the 'autogen.sh' script |
| 13:38.29 | brlcad | yeah, he had those |
| 13:38.31 | Z80-Boy | which brl-cad must have done? |
| 13:38.41 | brlcad | those obsd really annoying wraps them so that you have to set env vars |
| 13:38.52 | brlcad | yeah, you have to run: |
| 13:39.03 | brlcad | export AUTOCONF_VERSION=2.60 |
| 13:39.15 | brlcad | export AUTOMAKE_VERSION=1.10 |
| 13:39.25 | Z80-Boy | http://store.grainsurfboards.com/products/the-root |
| 13:39.25 | brlcad | or something like that, might have your versions wrong |
| 13:39.43 | brlcad | just look for: locate bin/automake |
| 13:40.32 | brlcad | put that version into the vars |
| 13:40.52 | ``Erik | the number of 'menstraul calendars' on freshmeat is just disturbing O.o |
| 13:42.42 | brlcad | er, one? |
| 13:43.06 | brlcad | i see one app for a mobile phone, but nothing else |
| 13:43.20 | ``Erik | I've seen a fistful listed over the years |
| 13:43.29 | Z80-Boy | .oO(those women) |
| 13:43.59 | brlcad | nothing coming up on search, this must be something you look for a lot :P |
| 13:44.24 | ``Erik | well, probably a combination of geeks who've never had to worry about that and married geeks trying to breed |
| 13:44.34 | ``Erik | heh, yes, I must control my ovum. O.o |
| 13:44.51 | brlcad | there's three more if you search my ation |
| 13:45.00 | brlcad | s/my/by/ |
| 13:46.44 | ``Erik | google pulls up a slew of 'em if you just dump keywords in |
| 13:47.00 | ``Erik | but a lot are multiple hit dealies... project page, activity stats, screenshot, etc all for one project |
| 13:48.52 | *** join/#brlcad Elperion (n=Bary@p54877BE3.dip.t-dialin.net) |
| 14:02.12 | Z80-Boy | so *_VERSION set, ./autogen.sh ran |
| 14:04.54 | Z80-Boy | 7.10.3, woohoo :) |
| 14:05.48 | Z80-Boy | brlcad: question: the normal vectors for boundary planes specified when building arbn on commandline, are they pointing outwards or inwards relative to the resulting solid? |
| 15:08.40 | brlcad | outwards |
| 15:09.08 | Z80-Boy | and it asks for distance between the plane and the origin |
| 15:09.26 | Z80-Boy | Is this still the distance itself when the vector doesn't have a unit length? |
| 15:10.17 | brlcad | the vector will be unitized |
| 15:10.34 | brlcad | so if you give something > 1, then length will be scaled |
| 15:10.49 | brlcad | if the vector is the size you want, then the distance/scaling factor should be 1 |
| 15:10.59 | Z80-Boy | I don't get it |
| 15:11.35 | Z80-Boy | "so if you give something > 1" -> does "something" refer to vector length or the distance? |
| 15:12.02 | Z80-Boy | "then length will be scaled" -> does this mean I enter 800 and the plane may possibly end up being 1600 instead of 800? |
| 15:12.03 | brlcad | example: in arbn arbn 6 1 0 0 10 0 1 0 10 0 0 1 10 -1 0 0 10 0 -1 0 10 0 0 -1 10 .. creates a 20x20x20 box because of the unit vectors |
| 15:12.25 | brlcad | if you give a vector length > 1 |
| 15:14.13 | brlcad | example2: in arbn2 arbn 6 5 0 0 10 0 1 0 10 0 0 1 10 -1 0 0 10 0 -1 0 10 0 0 -1 1 .. creates a smaller box because of the 5 0 0 vector (so length is 10 / 5 = 2) .. making the box 12x20x20 |
| 15:14.48 | Z80-Boy | oh so it's a misnomer in brl-cad |
| 15:15.01 | brlcad | oop, missing the last 10 |
| 15:15.04 | Z80-Boy | it asks for distance, but actually it's not distance, but the fourth equation coefficient |
| 15:15.48 | brlcad | er, it's still a distance, but still also scaled by the vector |
| 15:15.55 | brlcad | it could probably say "scaled distance" |
| 15:16.25 | Z80-Boy | this is really frustrating - I am already trying to determine how to make a triangle with chopped corners for the 3rd day |
| 15:16.32 | Z80-Boy | First I read the Volume II blah blah |
| 15:16.55 | Z80-Boy | which gave me a lot of irrelevant information, but the really important one - which way the vector points and what the fourth number means - was absent |
| 15:17.14 | Z80-Boy | Then I tried to use the graphical clickable editor just to figure out it isn't supported for arbn |
| 15:17.23 | Z80-Boy | ted is also not supported for arbn... |
| 15:17.36 | brlcad | well, you could have figured out "which way the vector points" in a minute by just creating one |
| 15:18.05 | Z80-Boy | At least now, after several hours fiddling with calculator, I can put in the number and with a bit of luck not make a mistake a get the right shape |
| 15:18.16 | Z80-Boy | brlcad: how? |
| 15:18.40 | brlcad | in ? |
| 15:18.47 | Z80-Boy | how to try it out? |
| 15:19.20 | brlcad | by creating a simple shape, like in the example I gave |
| 15:19.28 | brlcad | simple box, boom done |
| 15:19.56 | brlcad | had the vectors been reversed, it would have been pretty apparent |
| 15:19.58 | Z80-Boy | hmm but if the shape doesn't display then I don't know if it's because my theory about how it works inside is wrong or I forgot some hidden aspect of mged which is possibly not mentioned in the manual... |
| 15:20.06 | Z80-Boy | A method of trial and error is not acceptable |
| 15:20.15 | Z80-Boy | It should be documented |
| 15:20.39 | brlcad | are we really going to have this same discussion again? |
| 15:20.41 | brlcad | of course it should |
| 15:20.45 | brlcad | are you going to document it? |
| 15:20.45 | Z80-Boy | also those things like "what it really does depends on mged's internal states" -> great. How do I know now how? |
| 15:21.01 | Z80-Boy | If I get really pissed, maybe even yes :) |
| 15:21.18 | brlcad | technically it's not even that it's inward or outward |
| 15:21.24 | brlcad | because it's plane coefficients |
| 15:21.41 | brlcad | you never define a point and a vector, you just define the vector |
| 15:21.47 | Z80-Boy | it's some kind of equation like ax+bx+cz<=/=>/</>d |
| 15:21.47 | brlcad | ergo, it has to be from the origin |
| 15:22.20 | brlcad | where everything is an outward direction effectively |
| 15:22.43 | Z80-Boy | I guess somehow in the U. S. Army learned it 30 years ago from the source code and now it doesn't have to be documented until he dies |
| 15:23.09 | Z80-Boy | and someone else has to take over his chair and start typing the numbers instead of him, am I right? ;-) |
| 15:23.23 | brlcad | he/his? who? |
| 15:23.45 | Z80-Boy | What is actually the official documentation of brl-cad? |
| 15:24.19 | brlcad | the pdf's on the website, the manual pages, and everthing in doc |
| 15:24.37 | Z80-Boy | Like a lot of mysteries around BRL-CAD |
| 15:24.55 | Z80-Boy | For example - there are Volumes II, III, IV - is there also a mysterious Volume I? If yes, what's inside? |
| 15:25.05 | brlcad | a lot of "expert-friendly" knowledge that you learn the more you use it |
| 15:25.17 | brlcad | and a lot of work needed on making it more user-friendly and easy to use of course |
| 15:25.22 | Z80-Boy | the problem is I use BRL-CAD only ocassionally on Ronja |
| 15:25.49 | Z80-Boy | sometimes I go soldering, sometimes I go welding, sometimes debug electronics, sometimes I need to add a new 3d model or modify existing one |
| 15:25.57 | Z80-Boy | For that I need a usable reference guide |
| 15:26.04 | brlcad | dude, I completely agree |
| 15:26.11 | brlcad | the issue is just a matter of time and priorities |
| 15:26.18 | Z80-Boy | And not information where the crucial detail is always missing |
| 15:26.32 | brlcad | i could be working on docs, and that would help for a short while .. but then there are bigger problems that would go unaddressed |
| 15:26.39 | Z80-Boy | Is it possible to change the source of Volume II and add notes about what it really does at the latest detail level? |
| 15:26.49 | brlcad | other problems that *only* I can take care of at the moment |
| 15:27.02 | CIA-4 | BRL-CAD: 03erikgreenwald * 10brlcad/src/mged/qray.c: change (char)NULL to CNULL ( (char)0x0 ) to quiet the warnings about casting to different size |
| 15:27.05 | brlcad | it is possible, that's part of the docbook work we were talking about earlier |
| 15:27.06 | Z80-Boy | I always remember that arbn does a n-plane demarked body but never remember if the inequality is this or that way |
| 15:27.20 | Z80-Boy | if a rotation goes counter or clockwise and that's exactly what's missing in these references |
| 15:27.25 | brlcad | vol I is on the website, it's just exceptionally brief |
| 15:27.48 | Z80-Boy | Where is Vol I ? I can't find it |
| 15:27.55 | brlcad | frankly, I didn't remember what the inputs to an arbn were either, I just made one and found out |
| 15:27.58 | brlcad | it's the first link |
| 15:28.03 | brlcad | under docs |
| 15:28.27 | Z80-Boy | Overview of BRL-CAD? |
| 15:28.30 | brlcad | yep |
| 15:28.51 | Z80-Boy | But there's no indication it's Volume I! |
| 15:29.23 | brlcad | the docs predate the open sourcing and the new build system, so vol I actually included all the details about the license agreement, the old build system installation instructions per platform, and other details that are simply no longer relevant |
| 15:29.53 | brlcad | uhm, who cares if it doesn't say it's vol I ... wtf does that matter? |
| 15:30.14 | Z80-Boy | Now you saw it |
| 15:30.17 | brlcad | that's really getting critical on something that really doesn't matter |
| 15:30.21 | ``Erik | it has a '1' by it, does that count? |
| 15:30.43 | ``Erik | just like volume 2 has a '2' by it... |
| 15:30.43 | ``Erik | :D |
| 15:30.46 | brlcad | there are plenty of real things to complain about, a label on that doc isn't one of them |
| 15:30.50 | Z80-Boy | Then the doxygen documentation would be volume 7 |
| 15:31.01 | Z80-Boy | And ARL technical reports volume 8 |
| 15:31.09 | ``Erik | ok? |
| 15:31.16 | brlcad | does it matter? |
| 15:31.19 | brlcad | seriously? |
| 15:31.24 | Z80-Boy | I didn't complain I just asked where is the volume I |
| 15:31.57 | Z80-Boy | Becuase I thought the information that's missing in Vol II is maybe there |
| 15:32.29 | Z80-Boy | What is supposed to contain the most detailed reference on arbn? Volume II? |
| 15:32.33 | brlcad | you asked, I answered :) |
| 15:33.13 | brlcad | it's just not even worth the edit, or this discussion to futz with it since the effort that is needed is the conversion to docbook |
| 15:34.26 | Z80-Boy | What needs to be converted to docbook? Only the Volume-* PDF's? Or also the manual pages? Online helps? doxygen doc? |
| 15:34.59 | brlcad | the vol pdfs for starters, ideally everything and just have the various formats auto-generated |
| 15:37.25 | Z80-Boy | OMG is it rasterized at billion DPI? |
| 15:37.27 | brlcad | the "volume" formats aren't necessarily going to be preserved .. some info is verbose/redundant, some info is outright missing |
| 15:37.59 | ``Erik | heh |
| 15:38.24 | Z80-Boy | Printing into ps and running through ps2pdf didn't help... |
| 15:42.47 | brlcad | http://ftp.brlcad.org/MGED_Reference_Card_front.png http://ftp.brlcad.org/MGED_Reference_Card_back.png |
| 15:48.31 | Z80-Boy | no |
| 15:48.34 | Z80-Boy | but I have one at home |
| 15:49.23 | Z80-Boy | 2 pages PNG, 700kB each, that's not very practical |
| 15:50.30 | CIA-4 | BRL-CAD: 03erikgreenwald * 10brlcad/src/util/pl-X.c: fix sign difference |
| 15:50.34 | Z80-Boy | I made this: http://ronja.twibright.com/3d/refcard.png |
| 15:51.42 | ``Erik | hah, doh, didn't even know z80-boy was karel :D |
| 15:52.45 | Z80-Boy | L0CALZ 0NLY |
| 15:54.13 | Z80-Boy | Now I can open it in Links on one desktop and just grab the screen with mouse and move it around :) |
| 15:55.01 | Maloeran | Ahah http://news.bbc.co.uk/2/hi/americas/6998619.stm |
| 15:56.11 | ``Erik | heh |
| 16:00.18 | Z80-Boy | I put it together cut out the white strip in the middle, around the image, scaled as much down as it would be still legible and converted to 16 colors without dithering (to not get jagged text) to not make the PNG load long |
| 16:51.58 | CIA-4 | BRL-CAD: 03erikgreenwald * 10brlcad/src/mged/points/process.c: re-organize and clean up (fixing compiler warnings) |
| 16:52.28 | minute | brlcad: How is LDAP going? |
| 16:52.29 | CIA-4 | BRL-CAD: 03erikgreenwald * 10brlcad/src/mged/points/count.c: explicitely set id's type as int |
| 17:27.01 | ``Erik | heh, I hasn't thought about that before... 'e' is for draw and 'd' is for erase |
| 17:27.01 | yukonbob | ``Erik: ;) -- what does 'e' stand for? 'd' could easily be 'delete', but 'e'? |
| 17:32.09 | brlcad | woot, finally got rounded corner working |
| 17:33.37 | brlcad | Z80-Boy: uploaded your image, thanks |
| 17:34.00 | Z80-Boy | brlcad: you're welcome |
| 17:34.28 | brlcad | http://ftp.brlcad.org/MGED_Quick_Reference_Card.png |
| 17:34.47 | brlcad | I'll link it on the new site as data is migrated |
| 17:35.22 | Z80-Boy | cool |
| 17:35.48 | Z80-Boy | the card looks quite helpful actually I didn't use it because it was so slow to turn the pages |
| 17:36.15 | Z80-Boy | (xpdf viewer) |
| 17:37.05 | Z80-Boy | is it allowed to use the BRL-CAD logo in a free licensed movie end titles? |
| 17:37.41 | Z80-Boy | something like 3D modelling software (BRL-CAD logo) U. S. Army Aberdeen Proving Ground or however it is? |
| 17:38.36 | Z80-Boy | cause it looks quite cool |
| 17:39.40 | brlcad | welcome to use any of these: http://brlcad.org/images/logo/ |
| 17:41.27 | ``Erik | bob: I dunno, "edit"? *shrug* |
| 17:42.26 | Z80-Boy | They have "Army Research Laboratory" replaced with Open Source |
| 17:42.37 | Z80-Boy | I like more the ARL ones it makes an impression that a serious shit is going on |
| 17:43.24 | brlcad | and that same reason that it gives that impression makes it a bit harder to redistribute/use :) |
| 17:43.51 | Z80-Boy | cause of the ARL trademark or whatever? |
| 17:43.55 | Z80-Boy | So I should use the Open Source one? |
| 17:44.03 | brlcad | probably |
| 17:44.37 | Z80-Boy | but I can put ARL into the credits, right? |
| 17:44.59 | brlcad | you're making it, you can credit whomever you like |
| 17:46.22 | brlcad | the intention, however, is to move more and more towards a more familiar "global" open source project with developers and participants all over the world involved |
| 17:46.33 | Z80-Boy | OK |
| 17:46.49 | brlcad | ARL is still a major participant and funder of some of the developers, but in the end they're just a participant like everyone else |
| 17:47.00 | Z80-Boy | cause then I would have to start writing ARL et al.? |
| 17:47.19 | Z80-Boy | are there any non-ARL developers? |
| 17:47.26 | brlcad | sure, plenty |
| 17:47.34 | brlcad | how active they are varies heavily from time to time |
| 17:47.49 | Z80-Boy | does this opening mean that ARL is loding interest in the project and it's having some better software instead? |
| 17:47.54 | brlcad | and the ones paid certainly are the most active, but that won't necessarily always be true |
| 17:48.06 | brlcad | not in the least |
| 17:48.09 | Z80-Boy | loding -> losing |
| 17:48.18 | brlcad | the point is to encourage others to participate and get involved |
| 17:48.36 | brlcad | instead of just having people sitting on the other side of a fence waiting for stuff to get thrown over |
| 17:48.52 | Z80-Boy | I just don't understand armies are usually secretive and now they open a 3D modeling tool with 30 years of devel |
| 17:49.11 | ``Erik | hum, lots of stuff has been released through the years |
| 17:49.40 | ``Erik | 'ping', the improvements to the bsd4.2 network stack that found its way into winderz and pretty much every other current ipv4 capable os... |
| 17:49.51 | Z80-Boy | from ARL? |
| 17:49.54 | ``Erik | yeah |
| 17:49.57 | Z80-Boy | lol :) |
| 17:50.24 | ``Erik | from the same team, even... the guy who architected and did most of BRL-CAD is the one who did ping and the stack stuff |
| 17:50.49 | Z80-Boy | and now is dead because of a car crash? |
| 17:50.52 | ``Erik | plus loads of publically available documents come outta here |
| 17:50.54 | ``Erik | yeah |
| 17:51.05 | Z80-Boy | was it his mistake or the other's one? |
| 17:51.12 | ``Erik | <-- wasn't there |
| 17:51.45 | Maloeran | <-- wasn't born |
| 17:51.49 | Z80-Boy | I take it as an indication that road travel is dangerous |
| 17:51.51 | ``Erik | um |
| 17:51.54 | ``Erik | in 2000? |
| 17:52.00 | Maloeran | Oh, the guy died in 2000? |
| 17:52.02 | ``Erik | yeah |
| 17:52.11 | Maloeran | I thought it was a very old story, from the beginning of ping and brlcad |
| 17:52.27 | ``Erik | ping was '83, BRL-CAD kinda sorta started in '79 |
| 19:10.27 | CIA-4 | BRL-CAD: 03erikgreenwald * 10brlcad/src/mged/clone.c: |
| 19:10.27 | CIA-4 | BRL-CAD: Fixed a bug with funky matrix manipulation. Fixed a bug with the old object |
| 19:10.27 | CIA-4 | BRL-CAD: being written with the new matrix applied. |
| 19:14.43 | minute | brlcad: How is progress on LDAP? |
| 19:18.31 | CIA-4 | BRL-CAD: 03erikgreenwald * 10brlcad/src/mged/clone.c: rotation on v5 primivites |
| 19:28.17 | *** join/#brlcad Z80-Boy (i=clock@77-56-74-3.dclient.hispeed.ch) |
| 19:28.35 | Z80-Boy | brlcad: now my mged works, thanks |
| 19:37.40 | CIA-4 | BRL-CAD: 03erikgreenwald * 10brlcad/src/mged/tedit.c: search the tcl environment for an editor variable |
| 19:38.43 | CIA-4 | BRL-CAD: 03erikgreenwald * 10brlcad/NEWS: editor tcl variable |
| 19:39.35 | ``Erik | the comb cloning stuff seems fundamentally wrong to me O.o |
| 19:40.54 | Z80-Boy | It even seems to calculate with rt, another sign that the CVS works on my OpenBSD machine |
| 19:41.08 | Z80-Boy | Didn't have any problem did just the autogensh ./configure make make install |
| 19:42.01 | ``Erik | <-- did a fistful of fixes a bit back for obsd41 in a vm |
| 19:42.15 | ``Erik | didja up to 41, or are you still on 40? |
| 19:44.25 | Z80-Boy | yes I'm still 4.0 |
| 19:44.31 | ``Erik | hum, nifty |
| 19:46.10 | *** join/#brlcad elite01 (n=elite01@dslb-088-070-021-167.pools.arcor-ip.net) |
| 20:05.17 | Z80-Boy | that's good that it works even on the older version |
| 20:05.28 | Z80-Boy | it's actually not so old. |
| 20:06.55 | ``Erik | it worked fine on 38 *shrug* |
| 20:07.46 | Z80-Boy | hehe |
| 20:55.35 | brlcad | minute: I've been working on the website style all day today, tweaking style some more |
| 20:56.33 | brlcad | fixed the block page, modified the headers, changed the menu style and titles, and a few other details throughout |
| 21:01.24 | minute | brlcad: cool |
| 21:03.34 | brlcad | i'm not sure how to apply the theme to the wiki? |
| 21:03.55 | brlcad | i modified the block template to get the rounded corners |
| 21:04.12 | brlcad | I'd presume the same can be applied to the wiki, but can you point me at the template files? |
| 21:04.17 | minute | brlcad: They automatically apply to the wiki I think. |
| 21:04.24 | minute | it is a symlinked template |
| 21:04.42 | minute | try purging the parser cache via adding ?action=purge to the url |
| 21:07.14 | brlcad | still not there |
| 21:07.37 | brlcad | do you see it on, http://my.brlcad.org/wiki/Documentation |
| 21:07.56 | brlcad | it's picked up the other css changes |
| 21:08.00 | brlcad | but not the template change |
| 21:10.39 | CIA-4 | BRL-CAD: 03erikgreenwald * 10brlcad/src/mged/clone.c: Display all newly generated objects, not just the first. Autoview after display. |
| 21:10.55 | minute | hmm |
| 21:10.56 | brlcad | block.tpl.php is the tempalte page |
| 21:11.08 | minute | yeah |
| 21:11.13 | minute | that is included in the mediawiki thing |
| 21:11.19 | brlcad | hrmp |
| 21:12.35 | Z80-Boy | sounds like brlcad is going to use mediawiki? |
| 21:14.20 | brlcad | minute: it's not in the source, it's using the block class, but not the template |
| 21:14.32 | brlcad | Z80-Boy: yes |
| 21:14.45 | brlcad | it's in their docs, iirc |
| 21:14.47 | minute | the mediawiki sidebar is missing some elemts of the blocks in drupal |
| 21:15.29 | minute | wow the iPod nano is nice |
| 21:16.19 | Z80-Boy | brlcad: I had mediawiki on Ronja shortly and it got filled with viagra-cialis link spam |
| 21:16.29 | brlcad | Z80-Boy: heh |
| 21:16.38 | Z80-Boy | I hope brl-cad will have it condfigured in a way that this is not possible |
| 21:16.42 | brlcad | yep |
| 21:16.49 | brlcad | already is |
| 21:17.00 | minute | fixing as we speak |
| 21:17.11 | Z80-Boy | good |
| 21:17.12 | brlcad | following some stuff that we use on bzflag's mediawiki install |
| 21:17.27 | brlcad | which has been spam-free for a long time now (with massive exposure) |
| 21:17.49 | Z80-Boy | I abandoned the wiki after the incident since it would take so much work to clean all the pages - and people tended to submit incorrect information that was causing trouble to other people |
| 21:18.08 | minute | install captcha and poof spam gone |
| 21:18.11 | Z80-Boy | Like a population plan where a couple of parts had the wrong value -> very nasty |
| 21:18.23 | brlcad | has to be a good captcha, though |
| 21:18.44 | Z80-Boy | captcha can be outsourced to horny people who fill it in and get a porn video for it |
| 21:18.47 | brlcad | we went through two other captchas that were pretty useless .. still got loads of hits |
| 21:18.59 | brlcad | especially since we wanted to retain the ability allow anonymous edits |
| 21:19.20 | brlcad | recaptcha's pretty darn good though |
| 21:22.37 | minute | brlcad: xbox XD |
| 21:23.05 | brlcad | :) |
| 21:23.24 | brlcad | i can't claim that, though, took example from another site |
| 21:23.33 | minute | heh, I understand |
| 21:23.37 | minute | corners are dificult |
| 21:23.45 | ``Erik | heh, is there a captcha that makes ya answer stupid questions? like it shows a munged up "2+2" and you have to type "4" ? :) |
| 21:24.55 | brlcad | that was another thing we added to bz's form -- first we tried just "type the name of this game here" .. and if it had 'bz' anywhere in the answer, it let them past |
| 21:25.12 | brlcad | surprisingly still got too much spam (few times a week) |
| 21:25.37 | brlcad | same when we added a simple equation in the php that you had to answer to register |
| 21:26.00 | brlcad | definitely decreased the spam massively, but still too much for our hit-rate |
| 21:26.23 | ``Erik | hm, if it's just on register, or it's always the same, it'd be trivial for a spammer to circumvent |
| 21:30.32 | brlcad | nah, it was random |
| 21:31.21 | Z80-Boy | Hehe reCaptcha interesting concept using captcha to digitize books |
| 21:31.55 | Z80-Boy | someone could make fake wikis which spammers would try to break through so they would distribute the recaptchas to porn sites, people would solve them and the wiki would discard the info after 5 minutes or so |
| 21:32.10 | brlcad | :) |
| 21:32.18 | Z80-Boy | Interesting how it pushes AI forward |
| 21:32.34 | brlcad | yeah, it's pretty cool |
| 21:32.44 | brlcad | even if it gets fully cracked, it'll help :) |
| 21:33.28 | minute | brlcad: fixed |
| 21:33.37 | starseeker | Z80-Boy: Ah, yes :-) I saw that digitizing books from captcha idea - I like it |
| 21:34.41 | brlcad | awesome! |
| 21:34.45 | brlcad | how'd you do it? |
| 21:34.59 | minute | I had to add the corners to the sidebar divs in mediawiki |
| 21:35.10 | brlcad | where are those files? |
| 21:35.16 | brlcad | (in case I need to tweak) |
| 21:35.17 | minute | page.tpl.php |
| 21:35.23 | starseeker | brlcad: I'm not quite sure how to proceed here. If neither precise formatting nor precise content duplication is needed, I can probably have something reasonable looking for Vol1 (as a test case) pretty quickly. That would have a very "default" LaTeX look though. Would it be worth a test case? |
| 21:35.24 | brlcad | really, hrmph |
| 21:35.42 | minute | brlcad: there is an if statementn if it is drupal then drupal deals with it, otherwise we deal with it. |
| 21:35.56 | brlcad | starseeker: it really depends.. i'm also not sure I'd focus on vol I regardless |
| 21:36.08 | starseeker | brlcad: it's short ;-) |
| 21:36.22 | brlcad | it's also 90% there in html form :) |
| 21:36.31 | starseeker | before I wade into the big ones, I'd like to have a rough idea of what's needed for formatting |
| 21:36.43 | brlcad | well, what approach are you thinking of taking? |
| 21:36.59 | brlcad | vol III and IV are short(er) |
| 21:37.03 | starseeker | wv output a pretty good txt representation, and the docbook converstion extracted the images |
| 21:37.14 | brlcad | so then it's just docbook cleanup? |
| 21:37.29 | starseeker | Well, that's one option. I prefer LaTeX personally |
| 21:37.29 | Z80-Boy | Can they also make AS - Artificial Stupidity? |
| 21:37.30 | brlcad | because messy docbook is better than no docbook for a starter :) |
| 21:37.59 | starseeker | I have a fair bit of familiarity with LaTeX, and can probably get something basic put together pretty quickly |
| 21:38.09 | brlcad | starseeker: latex is great stuff, but just about every major doc project has migrated or is in the progress of migrating to docbook |
| 21:38.24 | starseeker | brlcad: Ah. |
| 21:38.49 | brlcad | latex doesn't fully decouple the presentation either, so it's harder to apply it to "non-traditional" output formats |
| 21:39.06 | brlcad | they're generally a bear to set up, but once set up, they just work |
| 21:39.24 | brlcad | most of the work is in the stylesheet, like web work |
| 21:39.33 | starseeker | If we go that route, I'd recommend adding the full toolchain to the src/other directory :-/ |
| 21:39.38 | brlcad | minute: how does it look to you? |
| 21:40.15 | brlcad | picking / evaluating the toolchain is part of the work :) |
| 21:40.15 | minute | in IE it has crappy antialiasing and the bottom set of corners are broken. otherwise good |
| 21:40.30 | starseeker | brlcad: OK, I'll work on docbook. Do you have any good examples that would make good "guides"? |
| 21:41.07 | brlcad | some platforms make the tools really easy, because they do everything in the package management system |
| 21:41.27 | brlcad | I know there are RH rpms for docbook-tools, there's probably something similar in ports |
| 21:41.54 | brlcad | the docbook manual is out of date with respect to the tools now |
| 21:42.23 | starseeker | Right, but I'll need to familiarize myself with the markup |
| 21:42.57 | brlcad | it always has been |
| 21:43.19 | brlcad | and likely will be for a long time to come -- it's still one of the best at dealing with equations |
| 21:43.47 | brlcad | might be useful to familiarize and model after what the guys are doing for the linuxdoc project |
| 21:44.24 | brlcad | http://tldp.org/ |
| 21:45.49 | brlcad | nifty, http://tldp.org/LDP/LDP-Author-Guide/html/x2docbook.html |
| 21:46.42 | brlcad | that's for the import side, the bigger problem is docbook -> output |
| 21:47.46 | starseeker | http://www.ibm.com/developerworks/xml/library/xml-matters3.html |
| 21:47.55 | starseeker | True, but I need something to try to convert first ;-) |
| 21:48.01 | brlcad | ah, cool -- looks like they use jade |
| 21:48.15 | brlcad | at least, they have tools that they've wrapped over jade in their docbook-utils package |
| 21:48.39 | brlcad | based off of redhat's docbook-tools package |
| 21:51.33 | brlcad | found it in ports, installing |
| 21:51.45 | brlcad | you have an account on the server? |
| 21:52.33 | starseeker | me? not to my knowledge |
| 21:57.22 | minute | goodnight |
| 22:01.16 | starseeker | night |
| 22:01.23 | starseeker | ah-ha - found it. |
| 22:01.28 | starseeker | Hrm. Docbook 3.1 |
| 22:01.34 | starseeker | oh, well - enough for a start |
| 22:02.17 | starseeker | brlcad: I think I've gotten the toolchain installed on my Gentoo box, at least most of it - there doesn't seem to be a convenient "docbook-tools" meta-ebuild unless I'm missing something |
| 22:03.06 | brlcad | there probably is :) |
| 22:03.13 | brlcad | maybe docbook-utils |
| 22:03.34 | starseeker | nope. |
| 22:03.54 | *** join/#brlcad b0ef (n=b0ef@062016141081.customer.alfanett.no) |
| 22:04.22 | starseeker | docbook-sgml-utils... but we want XML |
| 22:05.24 | brlcad | that probably works |
| 22:06.08 | brlcad | xml is a subset of sgml, you just provide an xml descriptor instead of docbook, and apply the stylesheet as needed |
| 22:06.46 | starseeker | OK |
| 22:06.50 | brlcad | yeah |
| 22:06.51 | brlcad | http://www.ibiblio.org/pub/Linux/docs/HOWTO/other-formats/html_single/DocBook-OpenJade-SGML-XML-HOWTO.html#AEN482 |
| 22:07.04 | starseeker | :-) |
| 22:09.31 | starseeker | brlcad: I'll take some time and go over Docbook - I've got some terminology to adsorb. |
| 22:19.21 | brlcad | minute: hrmph, broken you say .. that's no good .. and goodnight! thanks for fixing the wiki template |
| 22:19.35 | brlcad | i'll try not to break it again too much :) |
| 22:26.01 | ``Erik | *yawn* |
| 22:26.22 | Z80-Boy | Hmm the research in AI may lead to |
| 22:26.26 | ``Erik | <-- digs latex a lot, but would probably go back to sucking fbsd's project wang and imitate their doc sgml setup O:-) |
| 22:26.38 | Z80-Boy | instead of building something intelligent we realize we all are stupid |
| 22:28.50 | Z80-Boy | "task difficult for computers are easy for humans and vice versa. For example motoric skills are easy for humans" -> didn't feel like that when I was trying to learn to surf |
| 22:29.10 | Z80-Boy | reCaptcha is a good thought though |
| 22:29.30 | brlcad | more importantly, it's actually effective (so far) |
| 22:30.12 | Z80-Boy | Actually do you know that burrows-wheeler algorithm for compression? |
| 22:31.53 | Z80-Boy | maybe concepts - like a table - are not like numbers in out minds, but like pointers. |
| 22:32.06 | Z80-Boy | A table would be like a place in our experience. |
| 22:32.16 | Z80-Boy | Or more exactly a set of closely related places |
| 22:32.38 | Z80-Boy | Set of all moments in our life when a table played some role |
| 22:33.31 | Z80-Boy | burrows wheeler can associate between positions in a text without actually knowing into what units it should break the text first to figure out what to match with what |
| 22:34.05 | Z80-Boy | it's a damn simple algoritm - a qsort called on a pointer array where strcmp() is the compare function. |
| 23:48.18 | *** join/#brlcad qazwsx (n=qazwsx@unaffiliated/adgj) |
| 23:48.25 | qazwsx | I'm working on a procedural modeling project on LInux, so I'm looking for something that allows me to script in extrusions + boolearn operatons on solid objects / water tight meshes. Blender is not very stable for this ... anyone know how brlcad is for these operations? |
| 23:48.58 | brlcad | heh, it's kinda built for that sort of thing |
| 23:49.25 | qazwsx | really? |
| 23:49.27 | brlcad | there's a C api for doing this, with loads of examples in the src/proc-db directory of the source distribution |
| 23:49.34 | qazwsx | victory is mine |
| 23:49.36 | qazwsx | muhahahahaha |
| 23:49.54 | brlcad | there's also a fully-scriptable mged interface that you can script for in Tcl or build up commands from another scripting language |
| 23:50.01 | qazwsx | so this supports things like "here's a 2D curve, make it into a 2D surface", here's a 2D surface, make it into a 3D solid |
| 23:50.12 | qazwsx | and "here are two water tight meshes, give me the intersection / union / difference of them/ |
| 23:50.27 | qazwsx | ? |
| 23:50.28 | brlcad | you could work that way, but brl-cad prefers to work predominantly with 3D shapes |
| 23:50.36 | qazwsx | how do I generate 3D shapes? |
| 23:50.46 | qazwsx | (I'm just starting to compile brlcad right now, in ./configure) |
| 23:50.49 | brlcad | we have a sketch primitive that you can use for curves and extrusions, but the primitives are better |
| 23:50.56 | qazwsx | cool |
| 23:51.07 | brlcad | well, do you care about the C api more or the scripting interface? |
| 23:51.12 | qazwsx | since this project is old, I'm guessing a crappy modern day laptop will be fine for it? |
| 23:51.15 | brlcad | they're completely different from each other |
| 23:51.19 | qazwsx | I'm familiar from C to scheme, hmmm |
| 23:51.26 | qazwsx | I really care about procedural modeling |
| 23:51.37 | qazwsx | um, what does the C interface give me? |
| 23:51.42 | brlcad | you can do procedural modeling with either |
| 23:51.50 | qazwsx | (over the scripting langauge) |
| 23:51.52 | brlcad | it gives you speed, and you can hook into libraries |
| 23:52.06 | brlcad | like if you wanted to tie to simulation codes or something else |
| 23:52.19 | qazwsx | ah, no simuation, just generation |
| 23:52.27 | qazwsx | woot, "make" now :-) |
| 23:52.39 | brlcad | another example might be using some l-grammar tree library |
| 23:52.57 | qazwsx | I guess the useful documenation will be "3: Principles of Effective Modeling" and "5: Developing Applications" |
| 23:53.00 | qazwsx | as in l-systems? |
| 23:53.02 | brlcad | or NN's or GA's or whatever.. if you need to hook to a library or need performance, that's the way to go |
| 23:53.06 | brlcad | yeah |
| 23:53.13 | qazwsx | what's NN? neural network? GA? genetic algorithm? |
| 23:53.16 | brlcad | yep |
| 23:53.26 | qazwsx | dumb question: why would I use either NN or GA in modeling? |
| 23:53.58 | brlcad | oh, I could think of a few interesting projects :) |
| 23:54.33 | qazwsx | I can't think of a single reason to use them, so now I'm curious |
| 23:54.37 | brlcad | particularly a GA simulation driving geometry generation, or propagating agent behavior via geometry |
| 23:54.58 | brlcad | shape recognition, bin packing problems, knapsack problems |
| 23:55.17 | brlcad | lots of possibilities |
| 23:55.27 | brlcad | anyways, pretty aside from the point if that's not what you need :) |
| 23:58.47 | brlcad | so there's lots of scripting possibilities, you can script directly in mged (which is a full tcl shell) or outside mged and generate geometry transcripts |
| 23:58.52 | ``Erik | c/scheme? which impl? |
| 23:58.54 | ``Erik | <-- scheme geek |
| 23:59.32 | qazwsx | brl cad supports scheme? |
| 23:59.46 | ``Erik | no :( sadly |
| 23:59.53 | brlcad | here's one example that I recently whipped up in shell script that generates the SGI logo: http://my.brlcad.org/~sean/sgi.sh |
