irclog2html for #brlcad on 20050616

00:35.35 *** join/#brlcad cad430 (~db9f524d@bz.bzflag.bz)
03:50.58 *** join/#brlcad brlcad (~brlcad@brlcad.bronze.supporter.pdpc) [NETSPLIT VICTIM]
03:50.58 *** join/#brlcad archivist (~archivist@host217-35-103-47.in-addr.btopenworld.com) [NETSPLIT VICTIM]
03:50.58 *** mode/#brlcad [+o brlcad] by irc.freenode.net
03:54.09 *** join/#brlcad TheLastSpartan (guu@myth.gibbscam.com)
05:58.45 *** join/#brlcad CIA-3 (~CIA@flapjack.navi.cx)
06:00.57 *** join/#brlcad d_rossberg (~c28bf505@bz.bzflag.bz)
09:35.57 *** join/#brlcad clock- (~clock@183.45.203.62.cust.bluewin.ch)
10:17.01 clock- hello
10:22.43 d_rossberg Hi
10:33.46 clock- Does rtweight work in 7.2.6 for you?
10:33.57 clock- For me it stops in infinite loop spewing out this:
10:34.15 clock- error parsing line 21742 of density file.
10:34.15 clock- 0 args recognized instead of 3
10:34.18 clock- error parsing line 21742 of density file.
10:34.18 clock- 0 args recognized instead of 3
10:34.29 clock- Where the number constantly increased (sorry accidentally pasted twice)
10:34.35 clock- Although 7.2.2 worked OK on this.
11:20.45 d_rossberg i've never used rtweight
11:22.51 d_rossberg but there were made changes in reading .density between 7.2.2 and 7.2.4
11:24.27 d_rossberg also the error message you mentioned was introduced in version 7.2.4
11:29.28 d_rossberg e.g., every line has to contain 3 entries (fscanf has to return 3, in your case it's 2)
11:31.37 d_rossberg sorry, it's 0, of course
11:35.24 CIA-3 BRL-CAD: 03brlcad * 10brlcad/sh/make_dmg.sh: reorder the applescript and add an update and secondary method to attempt to force the window bounds and position
11:39.11 CIA-3 BRL-CAD: 03brlcad * 10brlcad/misc/macosx/Resources/ (ReadMe.rtfd/TXT.rtf Welcome.rtfd/TXT.rtf): update version to 7.3.0
13:07.48 clock- The content of the file:
13:07.50 clock- 1 7.7641 Mild Steel
13:07.50 clock- 2 1.522 Manufactured Rubber
13:07.50 clock- 3 2.579 Window Glass
13:07.50 clock- 4 8.920 Copper
13:07.52 clock- 5 2.700 Aluminium
13:07.55 clock- I see 3 entries.
13:08.27 clock- d_rossberg: and my file doesn't contain line number 10000 :)
14:00.40 d_rossberg clock: what's the first line number it complains about? (sorry, i was out of my office for a while)
14:05.54 d_rossberg clock-: if it's number 1 try to concatenate the second and third entry, i.e. "1 7.7641Mild Steel" etc.
14:07.45 d_rossberg the format string for fscanf is "%d %f%s"
14:08.15 d_rossberg maybe there should be a space between %f and %s
14:12.14 clock- d_rossberg: It's number 1
14:13.04 clock- d_rossberg: concatenated, and does the same.
14:13.12 clock- 1 7.7641Mild Steel
14:22.16 d_rossberg good ... er ... bad
14:23.50 d_rossberg i'll test it on an example ...
14:30.03 d_rossberg OK, i've tested it and i could reproduce your problem
14:30.38 d_rossberg try Mild_Steel, Manufactired_Rubber, etc.
14:47.37 brlcad clock-: there is very little error checking on the .density file
14:51.26 brlcad clock-: can you paste the file somewhere? it could very well be something simple like a trailing empty newline
14:51.51 brlcad there shouldn't be any empty lines or it will not parse correctly
14:53.34 brlcad spaces in the material name should be fine, the %s catches everything after the number with any whitespace separating
14:54.08 brlcad that might be a tab required between the %d and the %f .. I'd have to check
14:54.55 brlcad hmm.. not a tab
14:55.28 d_rossberg the problem is: fscanf("%s") != fgets
14:57.26 d_rossberg %s: String, up to first white-space character (according to MSDN)
14:57.46 brlcad hmm
14:57.49 archivist it might be more sensible for the string to be quoted
14:57.51 brlcad didn't used to be fscanf..
14:58.09 clock- brlcad: btw - the day before yesterday we have been talking about Mike Muus
14:58.22 clock- brlcad: yesterday I was riding a tram and it crashed with a truck :)
14:58.36 brlcad eek
14:58.59 brlcad ahh, looks like lbutler recently changed the .density file parsing..
14:59.00 clock- brlcad: one has to be careful here. Swiss are precise people, especially precise when it comes to targetting a tram with a truck.
14:59.20 brlcad heh
14:59.36 clock- brlcad: it happened in the very historic centre of Zurich, and the truck was Swiss (Thurgau)
14:59.46 brlcad well, to be precise, Mike's last name is Muuss ;)
14:59.54 clock- Aha, Muuss
15:01.20 clock- brlcad: I have been living in Prague for 25 years and nothing like that happened to me - and I have been going to university every day by a tram through the centre, for several years.
15:01.35 clock- brlcad: and after 6 months in Switzerland, bang- :)
15:01.37 brlcad d_rossberg: sure enough.. it used to be (void)fgets( buf, BUFSIZ, densityfp );
15:01.47 brlcad looks like an inadvertent bug injected
15:02.29 clock- Is it allowed to have whitespace contained in "Mild Steel"?
15:02.45 d_rossberg 7.2.2: yes, 7.2.4: no
15:02.48 brlcad clock-: it's good that you're at least well enough to be able to walk and talk about it
15:03.11 clock- brlcad: actually, it looked like nothing happened to anyone. The people just all walked out
15:03.21 brlcad lucky
15:03.32 clock- (after I opened the door with emergency opening. The tram driver obviously even didn
15:03.59 clock- t have the basic training that if something like that happens, he has to open the door because the truck's tanks can be leaking and could ignite, and fry the people inside).
15:04.08 clock- Only a while after I walked out, the driver opened the doors.
15:04.59 clock- brlcad: however 3 consecutive windows of the tram were de-glassed where the truck hit.
15:05.32 clock- brlcad: hmm, after replacing Mild Steel with Mild_Steel, it seems to work :)
15:06.19 brlcad i'll see if I can make a fix for that later today (unless d_rossberg beats me to it) ;)
15:06.25 clock- brlcad: does BRL-CAD support also modelling tram and truck crashes?
15:06.47 d_rossberg i'll try to fix it today
15:06.48 brlcad you mean automatic collision deformation?
15:06.51 clock- Material #15 Mildly_Hardened_Passenger_That_Regularly_Attends_Fitness_Room
15:07.24 clock- brlcad: what is automatic collision deformation?
15:07.49 brlcad the ability to impose a physical deformation on an object and utilize it's material properties
15:08.01 brlcad so, for example .. I have a model of a building and a truck
15:08.15 clock- Material #16 Old_Lady_That_Dies_From_Cardiac_Arrest_Just_From_Seeing_The_Truck_Approaching
15:08.26 brlcad I push the truck into the building and have it automatically deform both or either the truck/building
15:08.46 clock- brlcad: is this possible in software at all?
15:08.58 brlcad yes, it's possible
15:09.04 brlcad it's not easy, but it's possible to implement
15:09.15 clock- brlcad: btw have you seen the Sandia National Laboratory tests with propelling rail engine and airplane motor against 5m thick concrete wall?
15:09.17 brlcad we don't currently have that ability
15:09.19 clock- That was wonderful :)
15:09.30 clock- Rail engine -> didn't notice (just fell out of the track)
15:09.37 clock- airplane motor -> vaporized
15:10.06 brlcad no, I haven't but it sounds like other tests I might have seen :)
15:10.54 clock- brlcad: which ones? (unless this is an information kept secret to keep Axil Of Evil invading the U. S. A.)?
15:11.26 clock- (everyone who's not with us is against us! (Especially when it comes to breaking the Geneve conventions))
15:12.12 brlcad oh, all sorts
15:12.45 archivist they ran a train with a nuclear flask into something in the uk
15:14.40 brlcad ooh, interesting talk
15:14.46 clock- archivist: nuclear flask?
15:14.50 brlcad implicit surfaces for complex shapes
15:15.36 archivist yes its used for moving nuclear rod/wast by rail
15:15.47 clock- aha.
15:15.54 clock- And did it stay together?
15:16.34 clock- I have seen the pictures of hardened nuclear lava flowing from steam pipes of Chernobyl reactor.
15:16.58 archivist yes it was a pr success
15:17.01 clock- The guy at gamma spectroscope workplace in Prague said it was 500 times normal background that day - and Prague is waaay away from Chernobyl.
15:19.17 clock- Actualy about the distance as Italy
15:20.35 clock- brlcad: why did you strive for brlcad being opensource?
15:29.24 brlcad why?
15:29.37 brlcad why not?
15:29.46 brlcad there are many reasons actually
15:30.47 brlcad there are the traditional selfish reasons of getting improvements to BRL-CAD from a larger community of users/developers, but the main reasons are for BRL-CAD's own sake
15:31.16 brlcad I believe that it's a great CAD package in many ways with tremendous capability and possibilities for improvement
15:32.18 brlcad those improvements are very hard to realize or justify in a research or even commercial environment often, being open source lets the tool improve and adjust more rapidly to the community at large
15:33.09 brlcad we've always distributed source, but it was a real pain to actually get and use .. now it's very simple to get it and slowly becoming even more easier to use
15:50.18 CIA-3 BRL-CAD: 03d_rossberg * 10brlcad/src/rt/viewweight.c: read the names from .density until new-line
15:50.32 brlcad heh, yep, beat me to it :)
15:51.14 d_rossberg with the same solution?
15:51.33 brlcad nope
15:51.45 brlcad I was reverting it back to fgets with extra error checking
15:52.22 brlcad how portable is that %[^]?
15:52.51 d_rossberg i found it in MSDN and tested it on Debian 3.1 (sarge)
15:54.20 d_rossberg the problem with fgets is: a return of 0 doesn't mean neccessary an error
15:54.27 brlcad I think the common name can probably be made completely optional .. i'll have to give it a test here
15:54.43 brlcad yeah, that's fine
15:54.55 brlcad the name can be auto-generated for 0 returns
15:55.13 brlcad I believe that was part of the original idea too
15:56.23 d_rossberg i think, if a line contains no name, the fgets in the old solution has read the next line
15:56.57 brlcad fgets is supposed to stop at a newline and include the newline
15:57.24 brlcad so it 'should' be "\n" though that's what I'm testing now
15:58.28 d_rossberg that's right, but if fscanf reads the whole line (because of the missing name) fgets will start on the next line (and makes it a name)
16:00.12 brlcad hmm, okay
16:03.22 brlcad then I shall leave it alone ;)
16:03.52 d_rossberg at least until the next problem ;-)
16:03.58 brlcad yep
16:04.44 brlcad i might add a check for i == 2
16:40.54 CIA-3 BRL-CAD: 03d_rossberg * 10brlcad/NEWS: fixed .density file parser bug in rtweight
16:40.58 clock- what happens when I set material to 0 instead of default 1?
16:41.21 clock- Will it be taken as weightless?
16:41.37 brlcad no
16:41.44 brlcad it's just a table lookup
16:42.39 brlcad so a zero material just correlates to the material of type 0 in the density file
16:43.16 brlcad theoretically, you can use any non-negative number up to 32k, after 7.2.4 or up to 100 pre 7.2.4
18:25.40 brlcad ~translate de en Feierabend
18:25.56 brlcad cool
18:35.02 clock- ~translate de en unschuldig
18:35.17 clock- ~translate de en Schaffhausen
18:35.25 clock- ;-)
18:35.31 clock- ~translate de en Zurich
18:35.54 clock- ~translate de en Feuerverzinkerei
18:36.07 clock- ~translate de en Aluminiuminimumimunitaet
18:36.17 clock- ~translate de en verzinken
18:36.30 clock- ~translate de en verzinnen
18:36.43 clock- ~translate de en verzinntes
18:36.53 clock- ~translate de en Stahlblech
18:37.10 clock- ~translate en de tin
18:37.16 clock- ~translate en de can
18:37.22 clock- ~translate en de box
18:37.39 clock- ~translate de en verzinntes Stahlblechkasten
18:38.02 clock- wow, finally I can replace the annoying "tinned steel tin tin" in Ronja :)
18:39.42 *** join/#brlcad AllenHarvey (~8cb91c2a@66.111.56.50)
18:40.17 AllenHarvey Hello, I am having a major problem with the database format of BRLCAD 7.2 and would like some assistance
18:40.52 AllenHarvey I am using an application that was written with BRLCAD 4.4 in mind and can only read the older database format.
18:41.04 AllenHarvey is there a tool to convert from the new format to the old format?
18:41.19 AllenHarvey or is there a way to obtain a key for an brlcad 4.4?
18:49.10 brlcad hello AllenHarvey
18:50.21 brlcad AllenHarvey: there is a tool, though I'd highly recommend against going backwards
18:50.53 brlcad you're very likely to break geometry or otherwise loose information going from the v5 database spec to the v4 database spec
18:51.20 brlcad well, maybe not break geometry, but definitely might loose geometry
18:52.20 AllenHarvey what tool is that?
18:53.00 brlcad what's the application, if I may ask?
18:53.06 AllenHarvey MEVA 6.1.2
18:54.07 brlcad ahh
18:54.20 brlcad I've been working on upgrading MEVA to use the v5 format
18:54.34 AllenHarvey You told me that once before actually
18:55.00 AllenHarvey but with them converting their program to windows, i don't think they are in a hurry to upgrade their brlcad
18:55.04 brlcad heh, ok .. sorry, senility ;)
18:55.19 AllenHarvey actually, i have no clue how they are handling that issue with their next upgrade
18:55.31 brlcad actually, they are very interested -- talked to them a couple times since I last talked to you
18:55.40 AllenHarvey good good
18:56.03 AllenHarvey I wonder if they never knew brl-cad had advanced quite a bit
18:56.04 brlcad the larger issue is that they're merging MEVA into another larger framework
18:56.12 AllenHarvey what framework is that?
18:56.43 brlcad the name escapes me right this second
18:57.02 AllenHarvey does this mean that meva will become part of a larger application and not be it's own fully stand alone application
18:57.43 brlcad I can't speak with any authority, but that was my personal take on it
18:57.57 AllenHarvey by the way, I have only been able to get ahold of David Watts when it comes to MEVA, the 2 main developers have ignored everyone of my emails. I even made contact with an analyst who works with them
18:58.01 brlcad part of the .. *argl* .. something Framework
18:58.05 AllenHarvey and she told them to contact me but still haven't
18:58.51 AllenHarvey in any case, is there a way you can provide with me a key to brlcad 4.4 or the utility to downgrade the database?
19:00.14 brlcad I can do the latter, though I'll actually have to read the source to remember
19:00.18 brlcad gimme a sec
19:00.27 AllenHarvey no problem, thanks
19:03.11 brlcad got it?
19:03.20 AllenHarvey let me check
19:03.30 AllenHarvey you emailed it, correc?
19:03.37 AllenHarvey if so, i haven't received it yet
19:03.48 brlcad no no
19:03.51 brlcad other irc window
19:07.38 brlcad clock-: I'm actually not so positive about setting the material id to 0 now -- did you try it?
19:07.54 clock- brlcad: yes.
19:08.02 clock- 2.80319e+272 kg is not right, I guess :)
19:08.14 clock- We don't have enough matter in the universe for that :)
19:09.11 clock- brlcad: is it possible to make some program generate ASCII programs for mged which would put them into .g database and this way automatically generate e.g. a 3D model of printed circuit board with all the holes and tracks?
19:09.24 clock- brlcad: what should I do to make it not count?
19:10.29 clock- brlcad: what -> how
19:11.48 clock- brlcad: how -> what. Multiple brain errors detected in a row. Press RESET to continue or any other key to reset.
19:15.12 clock- hm the e+267 doesn't seem to be caused by material id 0
19:17.48 archivist hmm pcb dexign progs output drill table has all the hole onfo for a pcb,and the gerber output has track info, height and shape was crap when i last used a pcb prog in anger
19:18.31 clock- Problem: http://ronja.twibright.com/3d/
19:18.47 clock- The two bottommost consoles report both the same mass 4.26886e+267 kg
19:19.00 clock- I tried BRL-CAD 7.2.2 and 7.2.6 and both have the same problem.
19:19.07 clock- .density file is the link at the bottom.
19:21.06 clock- Is this my fault or is BRL-CAD broken?
20:08.03 *** join/#brlcad ibot (ibot@apt.bot.TimRiker.active.supporter.pdpc)
20:08.03 *** topic/#brlcad is http://brlcad.org/ || BRL-CAD is now Open Source! || Release 7.2.6 is now posted (20050612) || Several will be at Siggraph 2005, BoF meeting on Monday 11am-12noon || 'brlcad' is attending the 2005 International Conference on Shapes and Solids (Jun 13-17)
20:13.05 *** join/#brlcad clock-_ (~clock@39.46.77.83.cust.bluewin.ch)
20:26.49 brlcad clock-_: i've seen the e+272 problem, I was debugging it earlier
20:26.57 brlcad it's basically using a density of -1
20:28.10 brlcad it's a bug that's observed with an empty .density too
20:28.57 brlcad it's rarely ever encountered as those material IDs are actually relatively "standaradized" across historical use
20:29.20 brlcad e.g. id 1 is always steel, there is no id 0
20:37.05 *** join/#brlcad clock- (~clock@39.46.77.83.cust.bluewin.ch)
21:44.59 CIA-3 BRL-CAD: 03brlcad * 10brlcad/regress/Makefile.am: add weight.sh to the EXTRA_DIST list so that it's added to source distributions and so users will be able to run the test suite.. I swear I commited this already

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.