irclog2html for #brlcad on 20060111

00:32.47 *** join/#brlcad AchiestDragon (n=dave@whipy.demon.co.uk)
00:48.24 *** join/#brlcad PrezKennedy (n=Apathy@pcp010175pcs.dover01.de.comcast.net)
01:21.57 *** join/#brlcad justin_ (n=justin@pcp0011649600pcs.aberdn01.md.comcast.net)
01:29.57 ``Erik o.O
01:30.17 ``Erik <-- munches his frozen food
01:30.19 ``Erik where is it?
02:04.10 *** join/#brlcad DTRemenak (n=DTRemena@DHCP-170-143.caltech.edu)
03:53.14 *** join/#brlcad AchiestDragon (n=dave@whipy.demon.co.uk)
03:53.43 brlcad ``Erik: across from the Walmart on 40
04:09.22 pra5ad is there a good oss quicken/m$ money variant?
04:09.53 brlcad emacs
04:36.50 justin_ bc
04:46.21 ``Erik gnucash
05:02.14 phcoder is that what you use justin_?
05:07.42 *** join/#brlcad phcoder_ (n=jlowens@pcp0011650294pcs.aberdn01.md.comcast.net)
05:10.56 Maloeran "gnucash" sure has an interesting ring to it
05:11.30 phcoder_ g'night all
05:11.36 *** part/#brlcad phcoder_ (n=jlowens@pcp0011650294pcs.aberdn01.md.comcast.net)
06:06.03 *** join/#brlcad archivist (n=archivis@host217-35-76-52.in-addr.btopenworld.com)
06:44.56 *** join/#brlcad CIA-4 (i=cia@flapjack.navi.cx)
09:03.36 *** join/#brlcad clock_ (n=clock@zux221-122-143.adsl.green.ch)
09:04.04 *** join/#brlcad brlcad (n=sean@bz.bzflag.bz) [NETSPLIT VICTIM]
09:06.01 *** join/#brlcad kaol (n=kari@sammakko.yok.utu.fi) [NETSPLIT VICTIM]
09:58.12 *** join/#brlcad giuSerpe (n=giuSerpe@host31-48.pool82107.interbusiness.it)
09:58.15 giuSerpe hi all
11:30.45 *** join/#brlcad phcoder (n=phcoder@pcp0011650294pcs.aberdn01.md.comcast.net)
11:37.14 *** join/#brlcad CIA-4 (i=cia@flapjack.navi.cx)
13:12.03 *** mode/#brlcad [+o brlcad] by ChanServ
15:34.32 CIA-4 BRL-CAD: 03bob1961 * 10brlcad/src/tclscripts/archer/Archer.tcl: The edit panels are no longer using Scrolledframes. Modify _select_node method to always update the edit panel. _reset_edit no longer calls _refresh_tree
16:35.01 CIA-4 BRL-CAD: 03bob1961 * 10brlcad/src/tclscripts/archer/ShaderEdit.tcl: Arrange to call the shader changed callback when editing the shader form.
16:37.03 CIA-4 BRL-CAD: 03bob1961 * 10brlcad/src/tclscripts/archer/Archer.tcl: Modify the _apply_edit method to not call _refresh_tree.
17:19.22 *** join/#brlcad pier (n=pier@151.56.239.49)
17:31.40 pier Hi everybody
17:34.52 pier I wondered if anyone could explain, or provide the place where to find some more detailed information about instanceing rather then copying a comb group pf objects in brlcad. Thanks :)
17:37.44 pier sorry! I meant referencing objects
18:13.21 brlcad pier: sure
18:15.11 pier hi brlcad
18:15.24 clock_ brlcad: I would like to know as well
18:15.24 pier is the doms over?
18:15.38 clock_ brlcad: I always do it in a way from which brlcad segfaults
18:16.14 brlcad clock_: your method is specifically a bug ;)
18:16.28 brlcad multiply referencing the same object in one combination
18:17.00 brlcad should work, but isn't common practice and apparently mged isn't behaving right when you do that (though raytracing has no problem with it)
18:17.04 clock_ brlcad: is my method a bug of itself or is my method a bug of brlcad?
18:17.16 clock_ brlcad: but how should I do it?
18:17.49 clock_ brlcad: can you propose another method that is as simple as the one I use, and doesn't segfault mged?
18:17.56 clock_ brlcad: btw is this segfault already fixed?
18:19.11 brlcad clock_: i don't think your method is a bug in itself, seems reasonable to union an object N times and apply a different matrix to each
18:19.23 brlcad the bug is just an assumption that mged is making
18:19.29 brlcad yet the libraries are "doing the right thing"
18:19.40 brlcad hence why rt is right
18:19.41 clock_ brlcad: so it isn't fixed yet?
18:19.50 clock_ brlcad: do you have a bugreport from me about the bug?
18:19.56 brlcad yes, I do
18:20.03 brlcad it's not fixed yet
18:20.13 brlcad haven't tracked it down in mged
18:20.16 clock_ brlcad: do you plan to fix it?
18:20.21 brlcad yes
18:20.22 clock_ brlcad: can you reproduce it?
18:20.26 brlcad yes
18:20.50 clock_ hmm I could fix it myself
18:20.53 clock_ if I had time
18:22.04 clock_ if it's not a delayed memory corruption then it's probably trivial to fix
18:23.08 brlcad the problem itself is pretty easy to understand, not so sure about the easiest way to fix
18:23.49 brlcad when you go to apply a matrix, mged doesn't/didn't keep track of which /path/to/geometry you selected when you had multiple 'geometry' in your 'to' object
18:23.58 brlcad so it always applies to the first
18:24.10 brlcad eventually, that causes something to go awry in the display manager iirc
18:25.15 brlcad should just be a matter of adding an index or reference ID so you know which instance you're referring to
18:26.19 brlcad pier: to answer your question, place the objects you want to reference in a combination, and simply refer to that combination by name in other places
18:26.53 brlcad every time you use an object in a combination/region, it's by named reference
18:27.15 pier say I made a gusset with several holes and made a region of it
18:27.16 brlcad it's only when you create a shallow or deep copy that something becomes duplicated
18:27.26 brlcad aha.. heh
18:27.32 brlcad same case as what clock was trying ;)
18:28.04 pier and I need 16 of them
18:28.12 brlcad bizzare that we never ran into that problem here
18:28.49 brlcad pier, basically you make a template combination for a single hole
18:29.01 pier cp-ing all the holes and plates looks the hard way to get the whole thing done
18:30.08 pier ... how do I do that?
18:30.17 pier I mean a template
18:30.24 brlcad you shouldn't need to copy the whole thing, you just need to copy the one combination so you have unique names
18:30.49 pier really??? Is it as simple as that?
18:30.50 brlcad to make a "template", it's just a regular combination
18:31.09 pier I'll try it right away
18:31.10 brlcad even if you have to wrap one combination into another to abstract it
18:31.26 brlcad then copy that wrapper, move it wherever
18:31.33 pier will a region do?
18:31.38 pier as well
18:31.47 brlcad it's SUPPOSED to work without needing unique names, but clock_ did find a bug in that regard
18:31.55 brlcad a region will work just as well
18:32.03 brlcad though you will have to be careful of overlaps
18:32.29 brlcad remember that a region is meant to distinguish something that occupies physical space
18:32.44 pier at the moment I drew a plate and extruded it
18:32.45 brlcad whereas a combination is just a grouping
18:32.54 pier ok I got it
18:33.17 brlcad in pro/e or unigraphics lingo -- a brl-cad region is a part
18:33.35 brlcad combinations below a region are just csg operations
18:33.43 pier ... used to use solid edge / works
18:33.44 brlcad combinations above a region would be an assembly
18:34.02 brlcad same thing in solidworks
18:34.14 pier ok
18:34.48 pier so making cp piastra1.r piastra2.r is supposed to work
18:34.53 brlcad i only mention it as most people don't correctly understand combinations for a while ;)
18:35.00 brlcad sure
18:35.21 brlcad that will only copy the references in piastra1.r
18:35.33 brlcad so piastra2.r is referencing the same objects
18:36.21 brlcad there's a different cp to make a deep copy that actually creates copies of everything underneath piastra1.r
18:36.46 pier ok done.. now the point is ..how do I move the second?
18:36.59 brlcad apply a matrix to it
18:37.10 pier lord I'm so slow..
18:37.19 brlcad gui method, matrix edit on edit menu
18:37.36 brlcad select any of the primitives in the list that are in the copy
18:37.37 pier ok I'll go back to the man page to see what a matrix is...
18:38.13 brlcad no need
18:38.33 brlcad it's a mathematical matrix
18:38.45 brlcad that basically describes how to correctly orient and position the object
18:38.49 brlcad among other things
18:39.10 pier in the menu appears the list of all the sigle shapes
18:39.12 brlcad so every time you rotate something, or move it, or scale it, etc .. that's a change to some matrix
18:39.20 brlcad yes
18:39.28 brlcad pick any of them that say piastra2.r
18:39.36 brlcad then you'll get a second menu
18:39.42 brlcad pick piastra2.r
18:40.07 brlcad then any edit operation will apply directly to piastra2.r
18:40.08 pier that's why I got to a halt .... I was afraid of clicking on a single voice and doing a mess of it
18:40.25 brlcad no danger yet :)
18:40.36 brlcad you can always 'reject' when you're in edit mode
18:41.18 pier there's no region... in the list
18:41.19 brlcad so if you think you screwed up, try typeing the 'reject' command into the cmd window, or select Reject on the edit menu (only necessary when you're in edit mode)
18:41.50 brlcad er, what's your list?
18:42.23 pier clicking menu->matrix sel
18:42.30 brlcad it should be a list like /piastra2.r/someobject/blah /piastra2.r/someobject2/blah /piastra1.r/someobject/blah /piastra1.r/someobject2.blah
18:42.42 *** join/#brlcad Guu (i=guu@myth.gibbscam.com)
18:42.49 pier yep
18:43.00 pier shapes
18:43.16 brlcad yes, pick _any_ that say piastra2.r
18:43.19 pier and extruded object
18:43.21 pier ok
18:43.37 brlcad i.e. your copy that you want to move
18:43.44 pier ok I double clicked and blam a second menu
18:44.05 brlcad then on the next list, pick piastra2.r if that's what you want to move
18:44.31 pier ok ok
18:44.33 brlcad note that the others in the list would let you apply a transformation to something inside piastra2.r without affecting piastra1.r too
18:44.42 pier now it's smooth
18:45.12 brlcad yes, the interface is a little counter-intuitive since you select a primitive first
18:45.24 pier Lord I made hundreds of damned holes in this machine
18:45.26 brlcad there was a reason for it a long time ago.. ;)
18:45.46 brlcad there's also a pattern tool for replicating objects ;)
18:46.02 pier an hard copy of them
18:46.23 pier that not hinerit the properties of the father
18:46.52 brlcad ahh ;)
18:47.30 pier I mean the pencil that is not sharpened :)
18:48.18 pier page 19 VOL III
18:48.57 pier what is the way?
18:49.12 pier to replicate a comb?
19:19.28 CIA-4 BRL-CAD: 03brlcad * 10brlcad/src/librt/db_open.c: prevent a segfault if caller provides a null mode to db_open()
19:20.06 CIA-4 BRL-CAD: 03brlcad * 10brlcad/src/librt/dir.c: init mat to NULL before checking if it's null.
19:21.50 brlcad there is no pencil on page 19 of vol III
19:22.01 brlcad it's a helicopter
19:22.12 brlcad maybe you mean vol IV
19:22.25 brlcad er, II
19:23.39 pier last but fourth line
19:23.53 brlcad ahh
19:24.06 pier :)
19:25.03 pier brlcad: does p 0 180 0 work with matrix rotation?
19:25.23 pier for a 180 deg rot round Y axes
19:25.58 brlcad p is relative adjustments for whatever edit operation is selected
19:26.27 brlcad so that would rotate 180 degrees about the Y axis
19:26.48 pier it works with shapes but returns an error with matrix rotation
19:26.57 brlcad if you wanted to rotate, translate, or scale absolute instead of relatively, you'd use the rot, tra, and sca commands instead
19:27.31 pier without checking the proper box in the mged graph win
19:28.45 pier gas! it worked
19:29.58 ``Erik O.o
19:30.20 brlcad heh
19:30.28 pier ? breaks the winds?
19:30.41 brlcad pier: rot and orot commands for relative/absolute for objects
19:31.00 pier thanks a lot for the help
19:31.02 brlcad p is for editing parameters to pritives, my bad ;)
19:32.02 brlcad pier: do you understand the difference between figure 10 and figure 11?
19:32.23 pier mmmh oral test..
19:32.24 brlcad usually you just use the method in figure 11
19:34.02 pier referencing allows to make any altering to the prototype to be immediately transferred to the referred objects
19:34.26 brlcad it's not that it's "transferred", it's that they are actually all referring to the same thing
19:35.00 brlcad just how you respond to both "pier" and "Pierpaolo" .. they are both references to the "real you"
19:35.15 pier a matrix would represent just the coordinates of the referred new objects along with the rotations
19:35.27 brlcad so if the "real you" changes, so do "pier" and "Pierpaolo"
19:35.33 pier ok ok
19:36.06 brlcad right, the matrix just orients the object in space
19:37.36 pier that was what I meant about sharpening just one pencil (the original) and getting the other imediately sharpened... they are just images of the same real object (with material properties?)
19:38.10 brlcad right
19:38.17 brlcad cept not necessarily with material properties
19:38.23 pier ok
19:38.29 brlcad that depends where you've applied material properties
19:38.41 pier can be changed?
19:38.42 brlcad if your template is a region with material propertis, then yes, it'll inherit them
19:38.48 pier ok
19:38.59 brlcad if it's just a generic object with no properties, then there's nothing to inherint
19:39.07 pier whereas a comb has no mat prop
19:39.44 brlcad intrinsically, it's just a grouping
19:39.57 brlcad though the edit will let you apply shader parameters
19:40.05 brlcad s/edit/editor/
19:40.16 brlcad grr
19:40.18 brlcad ibot shut up
19:40.19 ibot yes, Master brlcad
19:41.06 pier brlcad: you are a very good teacher if you are able to teach me these things (rather complicated things)
19:41.31 clock_ brlcad: is it AI or what that it can understand English language and even correct it?
19:42.24 brlcad clock_: neither
19:42.51 brlcad s/blah1/blah2/ means "substitute the blah1 in what I just said with blah2"
19:43.38 brlcad ibot is just recognizing the s///, then finding a match in my previous line, then blabbing
19:43.46 brlcad relatively new feature, not fond of it really
19:44.10 pier is there a project for translating the manual in other languages?
19:44.30 brlcad there is not, though there have been one or two very interested
19:44.54 brlcad i'd like to get them converted from their current form to docbook sometime
19:45.03 brlcad then get translations of the docbook for various languages
19:45.36 brlcad right now it's pdf and msword (yes, the original is actually msword, bleh)
19:45.49 brlcad (wasn't me!) ;)
19:46.06 pier ok so if any help is needed I would like to volunteer
19:46.16 brlcad excellent
19:46.20 pier tex would be fine?
19:46.22 brlcad do you know docbook? :)
19:46.34 brlcad or willing to learn it
19:46.42 pier I'll have to have a look round the web
19:46.43 brlcad tex would be a distant second preference
19:47.05 pier heard about it quite a lot recently
19:47.06 brlcad docbook has a much better separation of content and style
19:47.38 brlcad it's easier to deal with docbook when you see examples
19:48.04 brlcad e.g. if you check out the gforge sources on gforge.org, their documentation is in docbook form
19:48.45 brlcad the linuxdoc is/was primarily docbook as their main format as well for all linux documentation (manpages, howtos, faqs, etc)
19:49.14 pier ok
19:49.41 brlcad could maybe even start with something more simple like converting one of the manpages to docbook (like rt's)
19:49.52 pier does Norman Walsh happen to be the one of the tcl-tk book?
19:50.50 pier none it was Welsh
19:51.02 brlcad ah, yeah
19:51.03 pier ok I'll have a go with it
19:51.04 brlcad Dan Welsh
19:51.36 pier rt man page?
19:53.36 brlcad there are manpages for most of the commands
19:53.41 brlcad man rt
19:53.49 brlcad if you've set up your system MANPATH
19:53.55 brlcad if not, try "brlman rt"
19:54.43 pier man rt work fine
19:54.47 pier works
19:56.02 brlcad pier: here's a starter
19:56.03 brlcad http://opensource.bureau-cornavin.com/crash-course/
19:57.31 pier great
19:59.58 brlcad here's a decent example of a rather extensive xml document for the gforge project: http://gforge.org/plugins/scmcvs/cvsweb.php/gforge/docs/docbook/docbook/?cvsroot=gforge
20:00.10 brlcad xml docbook document, that is
20:00.34 brlcad (select the rev number, e.g. 1.1, not the file name)
20:03.32 pier damit... last week I missed a refresher course at school about XML... I didn't think it could come in handy
20:04.29 joevalleyfield they rarely teach docbook (document) style XML, they usually teach it for data exchange
20:04.56 joevalleyfield so chances are what you missed isn't too applicable
20:05.20 pier ok
20:05.48 pier looks like html
20:05.52 pier a bit
20:06.18 joevalleyfield it's really similar to xhtml2, if that's out yet
20:07.13 joevalleyfield you tend to mark up nested groupings if logically related material
20:07.39 joevalleyfield s/if/of/
20:13.56 brlcad joevalleyfield: is that what you really meant? :)
20:17.14 ``Erik windows sucks and blows, both in most unpleaseant ways
20:19.32 pier amazing: Even if you have never used either DocBook or other markup languages (like LinuxDoc) before, you should be able to become proficient in it just by reading through this guide
20:19.53 joevalleyfield you told it to shutup ;)
20:35.02 pier so the path would be: 1)getting aquainted with docbook 2) pouring the translated text in the new form namefile.docbook
20:36.08 *** join/#brlcad CIA-4 (i=cia@flapjack.navi.cx)
20:43.50 pier Is docbook-sgml the packege needed for this purpose?
20:45.40 pier ver 4.4
20:50.59 CIA-4 BRL-CAD: 03brlcad * 10brlcad/src/librt/dir.c: heh, wrong line and bad syntax.. never mind
20:53.58 brlcad pier: yes and sure
20:54.12 brlcad though i'd give the docbook-xml a stab first
20:54.55 pier ok then I'll download it too
20:55.02 brlcad that way the file will actually render and work with existing xml tools too, even most browsers support some rudimentary xml markup
20:55.48 pier ok then
21:05.06 pier downloaded dsssl version too.... haven't the slightest idea what that is but what the hell...
21:05.59 pier night to everybody and thanks for the help... hope to be able to return sooner or later the kind help
21:07.24 joevalleyfield dsssl is predecessor to xsl
21:07.32 joevalleyfield its scheme based
21:07.41 joevalleyfield if you're into (((( instead of <<<
21:08.33 pier thank you very much joe (yet it's all greek to me :) )
21:09.15 pier got to get down with the tutorial first
21:09.21 joevalleyfield glad you're looking at it pier
21:10.22 brlcad working on docs is generally pretty high visibility stuff, and very appreciated..
21:11.15 brlcad I forget the exact numbers, but the last time I processed the web logs it was in the high thousands, like 10 or 20k downloads of the tutorials, etc
21:11.29 brlcad and that was many many months ago now
21:12.13 pier Yes I agree... it's the best way to spread out a good job
21:14.08 *** join/#brlcad clock_ (n=clock@84-72-61-225.dclient.hispeed.ch)
21:14.10 pier work
21:15.51 *** part/#brlcad pier (n=pier@151.56.239.49)
21:19.10 ``Erik maybe I should look into dsssl o.O
21:19.27 joevalleyfield are you a lisp lover, eric???
21:19.30 joevalleyfield don't do it
21:19.36 ``Erik yeah, actually, of the scheme variety
21:19.37 ``Erik heh
21:20.04 ``Erik *woops* :)
21:20.20 joevalleyfield :)
21:20.36 ``Erik (also, if you look at the freebsd ports, my name is attached to like half of the scheme related ones)
21:20.51 joevalleyfield doesn't scheme natively have continuations?
21:21.00 ``Erik yes
21:21.13 ``Erik first order
21:21.27 ``Erik call-with-current-continuation is the spec name, but most impl's have call/cc as a shortcut
21:22.27 ``Erik (guile is one of the very few that don't have the shortcut)
21:24.42 CIA-4 BRL-CAD: 03brlcad * 10brlcad/src/librt/db_open.c: remove dead init line, dbi_eof is set to something else two lines after.
21:24.44 *** topic/#brlcad by brlcad -> http://brlcad.org/ || BRL-CAD is now Free Software! || BRL-CAD Windows port has been merged to HEAD, not yet stable yet || December/January is a unified iteration to allow time for holidays || Archer is now on
21:24.48 brlcad <PROTECTED>
21:24.49 brlcad oop
21:26.19 *** topic/#brlcad by brlcad -> http://brlcad.org/ || BRL-CAD is now Free Software! || BRL-CAD Windows port has been merged to HEAD, not yet stable yet || Archer is now on HEAD as well, kick the tires, report any bugs || there WILL be a 7.6.8 released ASAP for January after all
21:26.48 brlcad ``Erik: if you need mods for 7.6.6 building in ports, lemme know and they can get into 7.6.8
21:27.01 brlcad e.g. the iwidgets test
21:27.02 ``Erik I've already commited them
21:27.05 brlcad it won't be the stuff on head
21:27.14 brlcad ah, okay, i'll try to find them
21:27.42 brlcad or you can commit them to the rel-7-6-branch too
21:28.00 ``Erik just a second...
21:29.17 ``Erik configure.ac has some tests and src/liboptical/Makefile.am has that hack we discussed that breaks aix supposedly
21:30.28 ``Erik I believe that's it
21:30.35 ``Erik I can mfc them if you want
21:31.23 ``Erik I'm not convinced that the C will allow an outisde itcl/itck/iwidget impl yet
21:31.51 ``Erik (that's what I was looking into when justin rebooted the minicluster)
21:31.59 ``Erik and fuck my arm hurts
21:37.20 ``Erik grrrrrr, I think justin justin just rebooted it again
21:38.05 ``Erik mv reboot reboot.orig ; echo "#!/bin/sh\necho Don't do that" > reboot
21:38.07 ``Erik o.O
21:44.46 *** join/#brlcad AchiestDragon (n=dave@whipy.demon.co.uk)
21:54.25 ``Erik or mebbe it crashed, heh
22:11.53 CIA-4 BRL-CAD: 03brlcad * 10brlcad/src/librt/db5_io.c: correctly free our allocated resources in db5_fwrite_ident() even on fatal error by stashing the result.
22:19.45 joevalleyfield diff -I is my new best friend
22:20.28 *** join/#brlcad DTRemenak (n=DTRemena@DHCP-170-143.caltech.edu)
22:20.51 brlcad ``Erik: nah, he keeps causing the crash
22:21.50 *** join/#brlcad PrezKennedy (n=Apathy@pcp010175pcs.dover01.de.comcast.net)
22:23.13 brlcad joevalleyfield: heh
22:26.08 PrezKennedy digg.com is turning into a vigilante website
22:26.22 PrezKennedy "omg letz haxxor or DDNS anyone who does something wrong online"
22:26.29 PrezKennedy *DDOS
22:37.54 *** join/#brlcad cad964 (n=55e2c6af@bz.bzflag.bz)
22:39.58 *** join/#brlcad cad885 (n=55e2c6af@bz.bzflag.bz)
22:43.35 CIA-4 BRL-CAD: 03brlcad * 10brlcad/src/librt/db_open.c: clean up db_create()'s logic considerably. annoyingly, the version check for writing out the ident can't go away since db_fwrite_ident() has no means to determine the version (need some wrapper interface)
22:45.59 CIA-4 BRL-CAD: 03brlcad * 10brlcad/src/librt/db_io.c: ws
22:52.36 joevalleyfield whoa
23:02.29 ``Erik get a private room, geez
23:54.08 ``Erik heh
23:59.56 *** join/#brlcad cad079 (n=55e2c6af@bz.bzflag.bz)

Generated by irclog2html.pl by Jeff Waugh - find it at freshmeat.net! Modified by Tim Riker to work with blootbot logs, split per channel, etc.