| 00:19.19 | narnia | brlcad, ping |
| 00:19.39 | danfalck | hey there |
| 00:20.25 | narnia | danfalck, hello. |
| 00:20.55 | narnia | been trying to update the express lib but i have hit a roadblock. :-( |
| 00:22.13 | narnia | somehow i need to obtain a copy of ISO 10303-11:2004. www.iso.ch wants 260.00 swiss francs for the pdf version. |
| 03:06.16 | *** join/#brlcad narnia (~terrylr@johann.blauedonau.com) |
| 05:16.36 | CIA-3 | BRL-CAD: 03brlcad * 10brlcad/configure.ac: take care of the bsd sys/mount.h present but cannot be compiled problem by making the test account for the export symbol name (addresses sf bug 1109930) |
| 06:09.49 | CIA-3 | BRL-CAD: 03brlcad * 10brlcad/configure.ac: just need to include sys/param.h |
| 09:08.51 | CIA-3 | BRL-CAD: 03brlcad * 10brlcad/src/mged/doevent.c: |
| 09:08.51 | CIA-3 | BRL-CAD: illumination selection should not take the aspect ratio into account or you end |
| 09:08.52 | CIA-3 | BRL-CAD: up unable to select geometry when the window is more wide than tall. Selection |
| 09:08.52 | CIA-3 | BRL-CAD: bands need to include a band for all objects displayed on the vertical. (sf bug |
| 09:08.52 | CIA-3 | BRL-CAD: 1114269) |
| 09:15.07 | CIA-3 | BRL-CAD: 03brlcad * 10brlcad/NEWS: fixed non-square window mged illumination selection (sf bug 1114269) |
| 10:06.44 | *** join/#brlcad Pimpi (~frank@p50821C9F.dip0.t-ipconnect.de) |
| 10:31.44 | *** join/#brlcad brainZzZ (stirps@c-67-171-62-252.client.comcast.net) |
| 11:55.21 | *** part/#brlcad brainZzZ (stirps@c-67-171-62-252.client.comcast.net) |
| 15:16.45 | *** join/#brlcad brainZzZ (stirps@c-67-171-62-252.client.comcast.net) |
| 15:39.28 | brlcad | hello brainZzZ |
| 15:39.30 | brainZzZ | free brlcad. |
| 15:39.46 | brlcad | hmm? |
| 15:40.24 | brlcad | brl-cad is free :) |
| 15:40.38 | brlcad | I'm not so free usually, though |
| 17:10.40 | *** part/#brlcad brainZzZ (stirps@c-67-171-62-252.client.comcast.net) |
| 19:15.19 | narnia | brlcad, ping |
| 19:25.32 | brlcad | narnia: pong |
| 19:28.06 | brlcad | saw your note about 10303-11 |
| 19:29.31 | brlcad | that one I do not have, but am curious why you think you need it.. from what I saw, libexpress covered the IS already |
| 19:49.05 | narnia | brlcad, libexpress has not been updated to the 10303-11:2004 standard. it blows it brains out on new express files. |
| 19:52.00 | narnia | brlcad, i have looked at several express/step projects i have found on the net. most have been abandoned or they are written in languages that require commericial packages. i.e. http://exp-engine.sourceforge.net/ is a good example. it is written in lisp. not that bad however the interface requires lispworks professional edition. the lispworks-personal edition will not work. |
| 19:53.41 | narnia | brlcad, numerous urls concerning express/step are broken. some are just totally useless. |
| 19:56.18 | narnia | brlcad, concerning express/step i have several interesting urls which basically say "express/step is much hype about nothing". if express/step are so popular why are the projects appear to be abandoned. |
| 20:00.31 | brlcad | i've seen exp-engine, it's interesting and could possibly be used in it's command-line processing mode -- but i think we should prefer a C library like libexpress of course |
| 20:01.25 | brlcad | the modifications to revision 2 of -11 were posted on the website iirc -- just not the whole thing |
| 20:01.56 | brlcad | there's also another C++ implementation that hooked to python that looked similarly promising, but libexpress still looks like it's the best bet |
| 20:11.01 | narnia | brlcad, the modifications to revision 2 of -11 have not been posted. there is an addendum to revision 1 of -11 is available for download from either iso.org/ansi.org. |
| 20:13.08 | narnia | brlcad, the problem with the express-engine project is that it is not maintainable. how many people want to learn common lisp? i learned lisp when symbolics computers were the current fad. expert-systems and knowledge engineering were the latest buzz words. |
| 20:16.12 | narnia | brlcad, the latest-pre-release of the express-engine is only available for windows. i find no unix vesion |
| 20:16.49 | brlcad | oh i agree that it's the lisp impl is the least desireable -- but it is a possibility to simply call it as an external process |
| 20:17.35 | brlcad | i'd rather fix libexpress myself, pick it up as a maintained library |
| 20:18.07 | brlcad | talked to the original author a little bit, he's not really interested in continuing it's development any more |
| 20:18.27 | narnia | brlcad, i agree. maintaining lex,yacc, and c is a whole easier than yet another language. |
| 20:19.02 | narnia | brlcad, unless the yet another language was perl. |
| 20:20.44 | narnia | brlcad, there is another project on sourceforge named osexpress. it use 'ant' ( a java make ) and an 'antlr' (www.antlr.org) recursive-descent-parser. it was last updated in 2001. |
| 20:20.45 | brlcad | to date, there's not really been any involvement with perl in brl-cad beyond scripting some interfaces internally to get some job done |
| 20:21.37 | brlcad | so i'd be a "little" hesitant to toss it in unless there's some really active devs that want to use it and maintain it -- we don't quite have that situation yet |
| 20:22.08 | brlcad | did you see the one in c++ that had hooks through python? |
| 20:22.16 | narnia | brlcad, the only reason for needing the -11:2004 express language reference manual is to modify the lex and yacc parser. |
| 20:22.57 | narnia | let me google for it a minute. |
| 20:23.59 | brlcad | did you actually get the yacc file to convert with bison/yacc? |
| 20:24.33 | brlcad | did you fix the problem, or did it just work on whatever platform you were testing? |
| 20:34.05 | narnia | brlcad, yes. the yacc file compiles with bison. |
| 20:34.18 | narnia | brlcad, i fixed it. |
| 20:34.45 | brlcad | what was it? |
| 20:35.05 | narnia | i found a powerpoint document concerning pyexpress but i find no links to any source code or home pages. |
| 20:36.19 | narnia | there were several errors in the yacc file. and there were errors in the lex file. they still need to be cleaned up. |
| 20:37.42 | narnia | for some reason yyresult was declared as an 'int' and also as a struct pointer. after reading the yacc file in detail it was clear that they were not the same thing. bison did not grok that. |
| 20:38.30 | narnia | i also converted any references to array sys_errlist to strerror(). |
| 20:38.33 | brlcad | hmmm |
| 20:39.20 | brlcad | yeah, the latter wasn't a problem on the system I was on just by chance |
| 20:39.23 | narnia | fedex does work and i even got express pretty print to work. |
| 20:40.17 | narnia | gnu-3.4 did not like the sys_errlist and kept complaining about it. to silence gnu i just fixed it. |
| 20:41.05 | narnia | there were also some probelms with the yaccflags and flexflags in the original makefile and mkrules files. |
| 20:41.39 | narnia | the makefile and mkrules still need to be rewritten. |
| 20:42.24 | brlcad | oh, they are would be ditched if we pull it in |
| 20:42.33 | narnia | the code in general was in need of cleanup which i have been doing. |
| 20:43.26 | brlcad | much easier to work up a Makefile.am for the library than to fix/mimic their build |
| 20:43.39 | narnia | agreed |
| 20:44.09 | brlcad | are you "narnia" on sf.net? |
| 20:45.18 | narnia | not sure what you are asking. on sourceforge i am terrylr or aslan. for some reason freenode will not let me use the nickname 'aslan'. |
| 20:46.20 | brlcad | aslan must be one of the restricted bot names from past abuse |
| 20:46.41 | brlcad | I could probably get it unblocked if you really want it |
| 20:47.38 | narnia | on the emc project on sourceforge.net i am terrylr. https://sourceforge.net/project/memberlist.php?group_id=6744 |
| 20:49.34 | narnia | once libexpress is squared away we still need to parse part21 files. i have been looking at the nist scl c++ code and have started changeing it over to straight c. |
| 20:50.59 | narnia | having iso-10303-21 would be nice to have but not necessary at this point. |
| 20:53.27 | narnia | iso-10303-21:2002 is industrial automation systems and integration -- Product data representation and exchange -- Part 21: Implementation methods: Clear text encoding of the exchange structure |
| 20:54.13 | brlcad | I'm not familiar with that one |
| 20:54.30 | brlcad | what does it cover that is not covered by an AP itself? |
| 20:55.07 | narnia | that is not clear to me. |
| 20:55.56 | brlcad | what's hinting you towards needing to parse part21 files then? |
| 20:56.21 | narnia | the step files that i have. |
| 20:56.40 | narnia | they all start with ISO-10303-21; |
| 20:56.40 | narnia | HEADER; |
| 20:56.40 | narnia | FILE_DESCRIPTION((''),'2;1'); |
| 20:56.54 | brlcad | from my understanding, we need an express parser to read the ap201/ap214 express schema files .. we feed it the ap and then the parser should be able to parse a conformant .step file for either of those ap's |
| 20:57.38 | brlcad | ah, hrm.. maybe feed it the express schema for 21 too .. i'll look more closely at my .step files here |
| 20:59.41 | narnia | here is an extract from a step file i have. |
| 20:59.44 | narnia | #13584=DIRECTION('',(0.E0,-1.E0,0.E0)); |
| 20:59.44 | narnia | #13585=VECTOR('',#13584,1.6E1); |
| 20:59.44 | narnia | #13586=CARTESIAN_POINT('',(4.670394736842E1,1.725E2,-1.535688475947E2)); |
| 20:59.44 | narnia | #13587=LINE('',#13586,#13585); |
| 21:00.39 | narnia | the <#num> may be viewed as either a variable or a macro. |
| 21:01.53 | narnia | then there is the need for the glue to create a brl-cad database. |
| 21:04.58 | narnia | on ansi.org ISO 10303-21:2002 is 72 pages at $144.00 usd. so basically $2.00 usd per page. the borders on obscene pricing. |
| 21:19.56 | brlcad | yeah, most of the pricing is a bit rediculous for a spec |
| 21:22.20 | brlcad | money grubbing .. |
| 21:23.08 | brlcad | if they really wanted wide standards adoption, there's no real reason to not make it freely downloadable and charge for printed versions or special electronic (tapes, etc) |
| 21:31.15 | narnia | agreed |
| 23:53.12 | *** join/#brlcad ibot (ibot@apt.bot.TimRiker.active.supporter.pdpc) |
| 23:53.13 | *** topic/#brlcad is http://brlcad.org/ || BRL-CAD is now Open Source! || Screenshots: http://sourceforge.net/project/screenshots.php?group_id=105292 || http://brlcad.org/images/mged.jpg || Release 7.2.0 is under way today (20050211)... |
