| 00:13.48 | *** join/#brlcad dreeves (n=dreeves@67.130.253.14) | |
| 00:22.05 | *** join/#brlcad typ0_ (n=coder@um-sd06-125-2.uni-mb.si) | |
| 00:22.21 | AlexandreGuedes | brlcad: where i can get .g file format specification ? |
| 00:59.34 | yukonbob | hello, cadheads |
| 01:04.25 | madant | AlexandreGuedes, best bet is the code + http://ftp.arl.army.mil/~mike/papers/brlcad5.0/newdb.html |
| 01:16.59 | AlexandreGuedes | madant: this link is not working, but I found: http://brlcad.org/OLD/newdb/newdb.html |
| 01:17.18 | AlexandreGuedes | is the same thing ? |
| 01:21.30 | madant | AlexandreGuedes, yep :) |
| 01:22.07 | *** join/#brlcad typ0 (n=coder@um-sd06-125-2.uni-mb.si) | |
| 01:22.28 | madant | hmm.. er.. the army.mil one says draft 19 july |
| 01:26.22 | AlexandreGuedes | WARNING: this document is out of date and incomplete, but a useful start for those writing to the .g format. See the librt and libwdb interfaces for more up-to-date information. |
| 01:27.08 | madant | exactly :) |
| 01:27.25 | AlexandreGuedes | it iis ok |
| 01:27.49 | AlexandreGuedes | i should develop a sample patch... |
| 01:28.04 | AlexandreGuedes | Someone could tell as should be a patch ? |
| 01:28.14 | madant | have u succeeded in checking out the repo and building ? |
| 01:28.27 | AlexandreGuedes | yes... |
| 01:28.35 | AlexandreGuedes | and make install.. |
| 01:28.51 | AlexandreGuedes | is all runing .. |
| 01:29.24 | madant | well you could take up any of the manageable things on the BUGS list or any small feature request at the sourceforge tracker or the things in the main TODO.. small things ofcourse :) |
| 01:29.55 | AlexandreGuedes | hmm |
| 01:29.55 | madant | AlexandreGuedes, I am not a mentor :) just a contributor |
| 01:30.11 | AlexandreGuedes | ok |
| 01:30.33 | AlexandreGuedes | tha pache can be a file .. ? |
| 01:30.43 | AlexandreGuedes | the patch* |
| 01:31.59 | madant | AlexandreGuedes, how to create a patch is mentioned here http://sourceforge.net/tracker/?func=add&group_id=105292&atid=640804 |
| 01:32.52 | madant | It is just an easy method for transmitting whatever changes you have made to the code for removing the bug, or implementing a new feature etc. so that a developer can easily apply it , check it etc. |
| 01:34.34 | madant | the diff tool as well as svn diff basically concatenates all the changes you have made to a single patch file which you can submit at the sourceforge page i mentioned above |
| 01:35.29 | madant | then someone from brl-cad can check the patch by applying it using the patch tool etc. |
| 01:35.45 | madant | thinks maybe this will explain things better than moi :) http://ariejan.net/2007/07/03/how-to-create-and-apply-a-patch-with-subversion/ |
| 01:36.01 | *** join/#brlcad jdoliner (n=jdoliner@c-68-51-76-57.hsd1.il.comcast.net) | |
| 01:40.13 | AlexandreGuedes | hhmm now I comprehended |
| 01:41.58 | madant | no problem :) go have some awesome time reading brl-cad code ;) some of it is decades old and still awesome |
| 01:44.31 | AlexandreGuedes | is a lot of code lines |
| 01:48.53 | madant | heh.. indeed ;) http://www.ohloh.net/p/brlcad/analyses/latest |
| 01:51.19 | AlexandreGuedes | ohhh |
| 03:02.04 | brlcad | AlexandreGuedes: you don't really need the .g file specification to make a converter |
| 03:02.18 | brlcad | there is an API for accessing/creating/reading geometry |
| 03:02.42 | brlcad | src/librt and include/raytrace.h include/rtgeom.h and include/wdb.h |
| 03:30.44 | AlexandreGuedes | i'm back |
| 03:39.05 | AlexandreGuedes | brlcad: ok i'm analising it |
| 03:40.02 | *** part/#brlcad jdoliner (n=jdoliner@c-68-51-76-57.hsd1.il.comcast.net) | |
| 03:57.54 | brlcad | AlexandreGuedes: great |
| 03:57.57 | *** join/#brlcad kanzure (i=bryan@66.112.232.233) [NETSPLIT VICTIM] | |
| 03:57.57 | *** join/#brlcad poolio (n=poolio@bz.bzflag.bz) [NETSPLIT VICTIM] | |
| 03:57.57 | *** join/#brlcad piksi (i=piksi@pi-xi.net) [NETSPLIT VICTIM] | |
| 03:58.31 | brlcad | AlexandreGuedes: librt is where most of the exporters work, libwdb is used by most of the importers |
| 03:58.31 | brlcad | librt is read/write -- libwdb is write-only |
| 03:59.38 | *** join/#brlcad dreeves (n=dreeves@67.130.253.14) | |
| 03:59.54 | AlexandreGuedes | There is some specific documentation ? |
| 04:00.13 | AlexandreGuedes | i'm lookin it now.. Converting_Geometry.pdf |
| 04:00.13 | brlcad | there is quite a bit of documentation, dpends what you're trying to do |
| 04:00.24 | brlcad | yeah, that's a good starting point |
| 04:00.35 | brlcad | for an exporter, this simple example is relevant: http://brlcad.org/wiki/Example_db_walk_tree |
| 04:01.37 | brlcad | looking at the existing converters is also good |
| 04:01.41 | brlcad | src/conv |
| 04:05.32 | AlexandreGuedes | there are already converters from .g to .vrml, .obj, .x3d |
| 04:07.35 | AlexandreGuedes | brlcad: some formats can represent many properties.. |
| 04:08.27 | AlexandreGuedes | which ones of them are more important for brlcad ? |
| 04:09.18 | brlcad | AlexandreGuedes: the ones that deal with the geometry itself of course |
| 04:09.33 | brlcad | other metadata specifically related to solid modeling and solidity |
| 04:09.38 | brlcad | the hierarchy, if any |
| 04:10.23 | brlcad | the rest is mostly attribute, parametrics, or 2D data that isn't easily converted |
| 04:11.39 | AlexandreGuedes | mostly : points, vectors, vertexes ... |
| 04:13.48 | brlcad | ah, yeah -- the 1D/2D/drafting entities can either be imported as sketch data, just stashed as attributes, or ignored |
| 04:13.58 | brlcad | they're not directly related to solid modeling generally speaking |
| 04:14.41 | brlcad | and solid modeling does take precedence over other purposes like design, drafting, and machining .. until there are devs to better support those domains |
| 04:46.05 | *** join/#brlcad dreeves (n=dreeves@67.130.253.14) | |
| 04:51.49 | *** join/#brlcad cad07 (n=3cf9db5d@bz.bzflag.bz) | |
| 05:05.36 | *** join/#brlcad andre|away (n=Andre_Ca@189.71.13.123) | |
| 05:08.19 | AlexandreGuedes | brlcad: are you here ? |
| 05:09.05 | brlcad | ~ask |
| 05:09.05 | ibot | Questions in the channel should be specific, informative, complete, concise, and on-topic. Don't ask if you can ask a question first. Don't ask if a person is there; just ask what you intended to ask them. Better questions more frequently yield better answers. We are all here voluntarily or against our will. |
| 05:09.34 | AlexandreGuedes | brlcad: i'm writing my project |
| 05:09.51 | brlcad | highly recommends submitting two applications if you have the time |
| 05:10.01 | brlcad | AlexandreGuedes: great |
| 05:10.35 | AlexandreGuedes | i'll offer importers to vrml, obj and x3d |
| 05:10.55 | AlexandreGuedes | and a interactive vizualization... |
| 05:11.19 | AlexandreGuedes | using opengl.. |
| 05:11.45 | brlcad | the converters can all go together on the same applications |
| 05:11.48 | AlexandreGuedes | brlcad: You think that is very or little ? |
| 05:12.15 | brlcad | it's not clear until you start coding |
| 05:12.27 | brlcad | depends how good you are at reading exising code |
| 05:14.24 | AlexandreGuedes | And the interactive visualization can be together ? |
| 05:15.10 | brlcad | interactive visualization using opengl will be very difficult without finishing the impementation of BREP support |
| 05:16.09 | brlcad | at least implementing BREP tessellation support and CSG to BREP conversion support |
| 05:17.08 | brlcad | still, please do submit both even stubbed out and we can discuss them in further detail. |
| 05:17.18 | brlcad | wanders |
| 05:20.52 | AlexandreGuedes | Unfortunately gsoc deadline coincided with other commitments... |
| 05:21.17 | AlexandreGuedes | this week i have others deadlines |
| 05:31.31 | deeeffache | Hey, if you wanted to create something that would generate.. say, a gear, that worked like, 'creategear numberteeth radius,' would it be best to write a shell script or is there another way of handling this? |
| 05:41.35 | *** join/#brlcad Ralith|trip (i=40f692fb@gateway/web/ajax/mibbit.com/x-68f140c27b9f8f2d) | |
| 05:41.44 | Ralith|trip | checks in |
| 05:57.47 | *** part/#brlcad Ralith|trip (i=40f692fb@gateway/web/ajax/mibbit.com/x-68f140c27b9f8f2d) | |
| 05:59.03 | yukonbob | deeeffache: could be done in Tcl, using mged for example. |
| 06:14.58 | AlexandreGuedes | brlcad: hey, thanks |
| 06:15.32 | AlexandreGuedes | now i'm understandig more about brlcad models |
| 06:17.55 | AlexandreGuedes | i'll be away, see you later |
| 06:33.51 | *** join/#brlcad AlexandreGuedes_ (n=chatzill@189-92-134-105.3g.claro.net.br) | |
| 06:35.00 | *** join/#brlcad jdoliner (n=jdoliner@c-68-51-76-57.hsd1.il.comcast.net) | |
| 08:29.44 | *** join/#brlcad d_rossberg (n=rossberg@bz.bzflag.bz) | |
| 08:35.29 | d_rossberg | brlcad: your comment on the new importer proposal is private |
| 08:35.52 | d_rossberg | i.e. the student should not be able to see it |
| 08:47.34 | *** join/#brlcad madant (n=madant@117.196.149.244) | |
| 10:15.59 | *** join/#brlcad mafm (n=mafm@223.Red-83-49-86.dynamicIP.rima-tde.net) | |
| 10:19.08 | *** join/#brlcad madant (n=madant@117.196.145.122) | |
| 10:27.44 | CIA-40 | BRL-CAD: 0385.186.3.59 07http://brlcad.org * r1327 10/wiki/Talk:Animation: New page: Hello |
| 10:54.12 | *** join/#brlcad madant (n=d@117.196.145.122) | |
| 10:56.28 | CIA-40 | BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/delete: deleted "[[Talk:Animation]]": content was: 'Hello' (and the only contributor was '[[Special:Contributions/85.186.3.59|85.186.3.59]]') |
| 10:56.51 | brlcad | d_rossberg: really?? oops, thanks! |
| 11:07.31 | mafm | hi |
| 11:08.33 | *** join/#brlcad madant (n=d@117.196.145.122) | |
| 11:20.00 | *** join/#brlcad andre|away_ (n=Andre_Ca@189.71.13.123) | |
| 11:25.50 | mafm | brlcad: mail FAIL: "It was never expected that the project woluThere is also a separate MGED+Archer" |
| 12:06.19 | *** join/#brlcad elite01 (n=omg@unaffiliated/elite01) | |
| 12:09.37 | *** join/#brlcad BigAToo (n=BigAToo@pool-96-230-124-198.sbndin.btas.verizon.net) | |
| 12:31.19 | brlcad | mafm: ah, oops, thanks -- "never expected that the project would 'finish' as it's a multi-year effort." |
| 12:43.06 | mafm | brlcad: I figured it out, just noted it in the case that you want to fix it :) |
| 12:43.06 | mafm | bbiab |
| 12:52.21 | brlcad | fixed and extended |
| 13:00.23 | ``Erik | cocks an eyebrow at the 'tops' program on his mac |
| 13:09.07 | *** join/#brlcad samrose (n=samrose@adsl-99-147-180-206.dsl.lgtpmi.sbcglobal.net) | |
| 13:17.06 | *** join/#brlcad BigAToo (n=BigAToo@pool-96-230-124-198.sbndin.btas.verizon.net) | |
| 13:36.42 | *** join/#brlcad madant (n=d@117.196.141.148) | |
| 14:26.46 | *** join/#brlcad _sushi_ (n=_sushi_@84-72-93-63.dclient.hispeed.ch) | |
| 14:31.51 | starseeker | wills his nosebleed to stop so he can get on the friggin road |
| 14:33.50 | brlcad | gives starseeker some cocaine to congeal the bleeding |
| 14:34.11 | starseeker | heh - best example ever of cure being worse than disease |
| 14:37.26 | _sushi_ | Do you know how to feed a long content of a Makefile variable into xargs? |
| 14:37.37 | brlcad | you probably wouldn't think that after trying it ;) |
| 14:39.14 | brlcad | _sushi_: grep variable Makefile | xargs ... |
| 14:40.25 | brlcad | or just the content: grep variable Makefile | sed 's/[^=][^=]*=[ ]*//g' | xargs |
| 14:41.57 | ``Erik | that assumes that the variable isn't split with escaped newlines |
| 14:41.59 | _sushi_ | I have variable $(GENERATED) which is a dynamically computed list of all files generated during compilation of Ronja |
| 14:42.09 | _sushi_ | I cannot grep it out of the Makefile. It's not written there. |
| 14:42.26 | ``Erik | if you have a bsd make handy, do "make -V VARIABLE" |
| 14:42.26 | _sushi_ | It's so many kilobytes it cannot be passed on a commandline. |
| 14:42.36 | ``Erik | gnumake is r-tarded |
| 14:43.05 | _sushi_ | I need it for gnu make |
| 14:43.13 | _sushi_ | it apparently doesn't have -V |
| 14:43.43 | _sushi_ | is there a limit on size of environment variable? |
| 14:43.50 | brlcad | true dat about escaped newlines, though easy to unescape all lines before the grep too -- one of our scripts does that |
| 14:44.10 | brlcad | _sushi_: most have some upper limit |
| 14:44.21 | brlcad | 65k on many nowdays |
| 14:45.06 | ``Erik | woke up too early, is already hungry :/ |
| 14:45.57 | ``Erik | wow, sgi for 25m? O.o |
| 14:48.01 | CIA-40 | BRL-CAD: 03Sean 07http://brlcad.org * r1328 10/wiki/Developer_Documents: stub in a development tips section for emacs and studio |
| 14:48.26 | _sushi_ | wow, it works |
| 14:48.30 | brlcad | ``Erik: lee is up for lunch |
| 14:48.38 | ``Erik | yeah, talked to him about 20m ago |
| 14:48.52 | brlcad | _sushi_: what does? |
| 14:49.00 | _sushi_ | export GENERATED |
| 14:49.03 | _sushi_ | calling an external script |
| 14:49.09 | ``Erik | jim isn't interested in a real lunch, just a bowling alley trip or something |
| 14:49.14 | _sushi_ | and there for filename in $GENERATED; do rm $filename; done |
| 14:49.25 | ``Erik | I was kinda thinking japan house, sake might help this headache out :) |
| 14:58.09 | *** join/#brlcad AlexandreGuedes (n=chatzill@187-24-1-98.3g.claro.net.br) | |
| 15:08.59 | _sushi_ | now it's already working with GENERATED :) |
| 15:09.38 | *** join/#brlcad madant (n=d@117.196.134.27) | |
| 15:25.22 | *** join/#brlcad typ0 (n=coder@um-sd06-125-2.uni-mb.si) | |
| 15:25.40 | *** join/#brlcad BigAToo (n=BigAToo@pool-96-230-124-198.sbndin.btas.verizon.net) | |
| 15:27.36 | *** join/#brlcad boerrebj_ (n=boerrebj@ti0036a380-1610.bb.online.no) | |
| 15:28.14 | CIA-40 | BRL-CAD: 03Sean 07http://brlcad.org * r1329 10/wiki/Emacs: add some functions, style, and behavior setups |
| 15:30.43 | CIA-40 | BRL-CAD: 03Sean 07http://brlcad.org * r1330 10/wiki/Emacs: pre markers so [] display |
| 15:53.07 | *** join/#brlcad gaurav_hiiii (n=chatzill@64.251.25.150) | |
| 16:22.57 | CIA-40 | BRL-CAD: 03Suryajith 07http://brlcad.org * r1331 10/wiki/User:Hippieindamakin87: New page: ==Suryajith== I am a final year undergraduate at Indian Institute of Technology,Kanpur, majoring in Mechanical engineering. ===Abstract=== I wish to implement a library/routine which comp... |
| 16:27.20 | *** join/#brlcad madant_ (n=d@117.196.143.218) | |
| 16:53.33 | *** join/#brlcad jdoliner (n=jdoliner@c-68-51-76-57.hsd1.il.comcast.net) | |
| 17:14.37 | *** join/#brlcad samrose (n=samrose@c-24-11-214-181.hsd1.mi.comcast.net) | |
| 17:24.06 | *** join/#brlcad _sushi_ (n=_sushi_@77-58-247-216.dclient.hispeed.ch) | |
| 18:08.07 | CIA-40 | BRL-CAD: 03brlcad * r34141 10/brlcad/trunk/misc/win32-msvc8/ (4 files in 4 dirs): update uuids |
| 18:09.20 | CIA-40 | BRL-CAD: 03brlcad * r34142 10/brlcad/trunk/src/gtools/g_diff.c: include sysv.h for bcmp() |
| 18:13.41 | CIA-40 | BRL-CAD: 03Suryajith 07http://brlcad.org * r1332 10/wiki/User:Hippieindamakin87: |
| 18:35.44 | CIA-40 | BRL-CAD: 03Sean 07http://brlcad.org * r1333 10/wiki/Google_Summer_of_Code/Project_Ideas: remove the priority distinction |
| 19:30.59 | CIA-40 | BRL-CAD: 03brlcad * r34143 10/brlcad/trunk/include/sysv.h: hah, missed committing this fracking declaration |
| 19:48.52 | CIA-40 | BRL-CAD: 03brlcad * r34144 10/brlcad/trunk/bench/run.sh: typo |
| 19:54.15 | CIA-40 | BRL-CAD: 03Sean 07http://brlcad.org * r1334 10/wiki/Google_Summer_of_Code/Project_Ideas: restore a benchmark database website idea |
| 20:01.02 | *** join/#brlcad andax (n=andax__@d213-102-41-51.cust.tele2.ch) | |
| 20:48.03 | *** join/#brlcad dreeves (n=IceChat7@67.130.253.14) | |
| 21:19.10 | jdoliner | brlcad could you provide a bit of guidance on a patch? |
| 21:25.03 | brlcad | jdoliner: sure, fix something ;) |
| 21:26.40 | brlcad | i.e., look at the BUGS file -- most of those items are a lot less work than the TODO file items (though todo also has a few tasks that are pretty simple) |
| 21:27.14 | brlcad | another thing is to start trying out a variety of the command-line tools until you find some unexpected behavior or a crash, then fix that |
| 21:27.20 | brlcad | usually doesn't take too long ;) |
| 21:29.19 | jdoliner | yes I'm working on bug: 2561346 |
| 21:36.07 | brlcad | looks |
| 21:36.22 | jdoliner | it has to do with tesselation failure |
| 21:39.58 | jdoliner | so I need a bit of help on the flow of the program |
| 21:40.30 | jdoliner | nmg_booltree_lead_trees is failing |
| 21:45.57 | jdoliner | tess* that is |
| 21:46.41 | jdoliner | and it's gotta be failing since ip-idb_meth->ft_tessellate is failing |
| 21:48.00 | jdoliner | but I don't really know what that references at all |
| 21:48.53 | mafm | brlcad: requirements are the usual: proposal in the wiki, mailing list, and sending patches? |
| 21:49.14 | brlcad | mafm: yeah |
| 21:49.37 | mafm | patches being in the interim period, right? |
| 21:49.37 | brlcad | jdoliner: so did you get john's response? |
| 21:49.41 | brlcad | what the problem was |
| 21:49.58 | jdoliner | no I didn't get john's response |
| 21:50.04 | jdoliner | did he send one? |
| 21:50.18 | brlcad | jdoliner: read the comments on the bug ;) |
| 21:51.06 | jdoliner | oh yes I did see that |
| 21:51.08 | brlcad | also might as well mention that ip->idb_meth is a callback table to specific primitives -- ft_tessellate calls a primitive's tess() function |
| 21:51.34 | brlcad | so if it's an ellipsoid, ft_tessellate() is a function pointer to ell_tess() |
| 21:51.49 | jdoliner | in this case it seems to be rt_extrude_tess() |
| 21:51.56 | brlcad | right |
| 21:52.18 | jdoliner | and that's the function I think I should modify, am I right about that? |
| 21:52.33 | brlcad | probably one of several really given what the problem is |
| 21:52.38 | *** join/#brlcad AlexandreGuedes (n=chatzill@187-24-53-56.3g.claro.net.br) | |
| 21:52.45 | brlcad | it's a matter of garbage-in == garbage-out |
| 21:53.01 | AlexandreGuedes | hi |
| 21:53.06 | brlcad | it's bad geometry, so we can either detect it and abort or fix it automatically if it's fixable |
| 21:53.11 | mafm | brlcad: patches being in the interim period, right? |
| 21:53.11 | brlcad | hello AlexandreGuedes |
| 21:53.17 | brlcad | mafm: what does that mean? |
| 21:53.18 | mafm | hi AlexandreGuedes |
| 21:53.42 | mafm | that patches doesn't have to be submitted before 3rd April deadline |
| 21:54.09 | mafm | just with time enough to evaluate it properly |
| 21:54.27 | brlcad | oh heck no -- patches can be submitted all the way up until we finalize selections (though the earlier the better obviously) |
| 21:54.30 | jdoliner | well it certainly is fixable |
| 21:54.39 | brlcad | which is like april 15th maybe? forget the timeline |
| 21:54.51 | jdoliner | yes april 15th is correct |
| 21:55.23 | brlcad | jdoliner: in this case, definitely -- you can detect that various segments need to be reversed if they don't form a loop |
| 21:55.30 | jdoliner | 1 question, is there any worry with fixing it, that we're maybe changing some feature of the geometry that the user wanted? |
| 21:55.59 | brlcad | and if that's the approach taken, reversing the segments, it probably belongs in import5() so that it's fixed once read off of the disk |
| 21:57.44 | mafm | goody |
| 21:57.47 | jdoliner | k so we want to make sure, anytime we import that we don't have these reversed segments |
| 22:00.00 | jdoliner | where can I find the import5() function? |
| 22:17.35 | brlcad | jdoliner: same file as tess() |
| 22:18.02 | brlcad | src/librt/primitives/sketch has the sketch primitive that is used with the extrude primitive |
| 22:18.43 | brlcad | jdoliner: alternative, you could make sure you only write out / export "correct" sketches .. but that will still allow them to be initially created and cause bad behavior |
| 22:19.25 | brlcad | alternative is to just detect the problem during import and print a message that there are incorrectly oriented segments |
| 22:19.46 | *** join/#brlcad BigATo1 (n=BigAToo@pool-96-230-124-198.sbndin.btas.verizon.net) | |
| 22:19.53 | brlcad | put that detection in a function, then it could be added to import/export/prep/tess/whatever |
| 22:20.38 | jdoliner | sorry that's nmg_booltree_lead_tess you're talking about |
| 22:20.53 | jdoliner | because Im not finding anything called import when I search |
| 22:23.36 | *** join/#brlcad louipc (n=louipc@archlinux/trusteduser/louipc) | |
| 22:26.24 | jdoliner | k nm found it |
| 22:30.40 | CIA-40 | BRL-CAD: 03Suryajith 07http://brlcad.org * r1335 10/wiki/User:Hippieindamakin87: |
| 22:32.07 | CIA-40 | BRL-CAD: 03Suryajith 07http://brlcad.org * r1336 10/wiki/User:Hippieindamakin87: /* Proposed approach */ |
| 22:33.05 | hippieindamakin8 | brlcad, comments please |
| 22:33.08 | hippieindamakin8 | http://brlcad.org/wiki/User:Hippieindamakin87 |
| 22:33.51 | hippieindamakin8 | madant_, mafm ^ |
| 22:38.58 | mafm | nice, hippieindamakin8 |
| 22:39.16 | hippieindamakin8 | mafm, comments :) |
| 22:39.37 | mafm | aesthetically? |
| 22:39.42 | mafm | technically, no idea :) |
| 22:40.08 | hippieindamakin8 | mafm, :D |
| 22:40.22 | hippieindamakin8 | mafm, meant technically |
| 22:41.00 | hippieindamakin8 | should be getting ready with a bug fix .. now i can do my course projects for the next 2-3 days |
| 22:41.28 | mafm | hippieindamakin8: I don't know much about b-reps or CSG or anything :P |
| 22:41.47 | mafm | as a proposal, if you want to submit that as application in the google app |
| 22:42.00 | mafm | it should have maybe less detail (or maybe not) |
| 22:42.16 | hippieindamakin8 | mafm, are u familiar with the computational geometry algos using approximation ? |
| 22:42.24 | mafm | but it must contain milestones, plans for it (in weeks or something) and this kind of thing |
| 22:43.10 | hippieindamakin8 | mafm, rt exactly |
| 22:43.17 | hippieindamakin8 | i forgot to mention the timeline :| |
| 22:43.25 | mafm | http://brlcad.org/wiki/OpenGL_GUI_Framework#Initial_Project_from_GSoC_2008 |
| 22:43.58 | hippieindamakin8 | mafm, thanks i shall add it to both of them |
| 22:43.59 | mafm | I think that mine from last year works pretty well (it's incomplete, lacking personal info and so on, but gives you an idea about milestones, deliverables and other sections) |
| 22:44.49 | mafm | (my full proposal was more detailed, now that I see it in detail :) -- I can provide you one of my proposals where you can see the overall structure ) |
| 22:45.26 | mafm | and I'm not familiar with the details of CAD or rendering in any sense, I only take part in the infrastructure of things -- GUI, interfaces, etc :) |
| 22:45.44 | hippieindamakin8 | mafm, i ll try to add it in :) i just have gone through madant_ 's proposal |
| 22:46.27 | hippieindamakin8 | mafm, havent u put in any app this year? |
| 22:46.48 | mafm | I'm trying, right now |
| 22:48.01 | mafm | madant_'s one is also a good think to inspire you, specially with the schedule |
| 22:48.41 | mafm | but hey, brlcad is the Big Boss, so take advantage of him (erm... with tender and care :) ) |
| 22:48.43 | hippieindamakin8 | yeah that is elaborate |
| 22:48.53 | hippieindamakin8 | mafm, yeah ;) |
| 23:07.44 | *** join/#brlcad Malyce (n=iamtanma@wlanaccess-ext.jacobs-university.de) | |
| 23:08.22 | Malyce | Hello |
| 23:08.33 | Malyce | I was reading through the Core C++ interface |
| 23:09.01 | Malyce | where can I find the class diagram for it ? |
| 23:09.47 | *** join/#brlcad dreeves (n=IceChat7@67.130.253.14) | |
| 23:10.18 | brlcad | hippieindamakin8: even at a glance, already better than last year :) |
| 23:10.59 | brlcad | you need to include a timeline with some basic goals/deliverables/objectives spelled out -- something that can help track progress |
| 23:11.03 | *** join/#brlcad Lezard (n=lezardfl@189.58.208.16) | |
| 23:11.10 | brlcad | that goes for others as well, if you don't have a timeline yet |
| 23:11.22 | brlcad | it'll be one of my first comments |
| 23:11.24 | hippieindamakin8 | brlcad, working on it. thanks |
| 23:11.29 | hippieindamakin8 | :) |
| 23:11.59 | mafm | hippieindamakin8: also, make sure to read all the required stuff in the website |
| 23:12.02 | brlcad | also, feel free to link to a personal site or portfolio or paper or whatever if you have it to share |
| 23:12.29 | brlcad | doesn't need to be a resume, frankly don't care much where you've worked -- but if you can show things you've actually done, that can be useful |
| 23:12.38 | hippieindamakin8 | brlcad, sure. i ll put the links. |
| 23:13.06 | brlcad | hippieindamakin8: yeah, what mafm said too -- there's a list of things you should be sure to include/cover |
| 23:13.11 | hippieindamakin8 | brlcad, rt now the ftp site of uNC is down so i shall put up the acm links |
| 23:13.12 | mafm | hippieindamakin8: such as: http://brlcad.org/wiki/Google_Summer_of_Code/Acceptance#List_your_milestones |
| 23:13.19 | Malyce | hiya |
| 23:13.22 | mafm | (and the rest of things in that page) |
| 23:13.29 | mafm | hi Malyce |
| 23:13.39 | Malyce | I managed to find the UML diagrams for Geometry engine and services |
| 23:13.49 | brlcad | hippieindamakin8: don't want links to other people's stuff, I mean *your* stuff if you have anything |
| 23:13.51 | Malyce | but couldn't find the one for Core C++ interface |
| 23:13.51 | hippieindamakin8 | brlcad, with sleeplessness i forgot abt the requirements |
| 23:14.06 | hippieindamakin8 | brlcad, as in anything related to that ? |
| 23:14.11 | Malyce | Could you pass me a link ? |
| 23:14.59 | mafm | sorry Malyce, that's classified information |
| 23:15.12 | hippieindamakin8 | i shall set a link to my resume/site. my paper isnt anywhere close to this :P |
| 23:15.46 | mafm | (malyce: I mean... I don't know, maybe other people does ;) ) |
| 23:16.06 | hippieindamakin8 | mafm, i just got a hosting from drupal(the free one) a couple of weeks ago :P that is pretty decent as a matter of fact |
| 23:16.35 | mafm | :) |
| 23:16.47 | Malyce | eh ? ok |
| 23:17.00 | Malyce | as long as the discussion is about passing links to previous works |
| 23:17.13 | Malyce | I don't really have links to what I have done |
| 23:17.37 | Malyce | But, for the past year, I have been working as a research assistant to formalize Solidworks 2008 |
| 23:18.21 | Malyce | I wrote VBA code for the Solidworks API, to extract CAD data to OmDoc(XML) and HASCASL(Haskell) |
| 23:18.48 | Malyce | And in the end, basically creating pointclouds from the CAD data |
| 23:19.04 | Malyce | Would such an explanation be any good ? |
| 23:20.11 | mafm | Malyce: if you're planning an application, you should read http://brlcad.org/wiki/Google_Summer_of_Code/Acceptance and do something like http://brlcad.org/wiki/User:Homovulgaris |
| 23:21.00 | mafm | milestones, schedule and other stuff depends heavily on what task do you plan to tackle |
| 23:21.05 | hippieindamakin8 | madant_'s app has become an example aint it :) |
| 23:21.10 | mafm | but in general, it should give you an idea |
| 23:21.24 | brlcad | Malyce: there aren't uml diagrams for the coreInterface -- but there is writeup for it on the wiki |
| 23:21.32 | Malyce | Yes, I read that. |
| 23:21.35 | brlcad | and the classes are pretty neatly organized -- it's not that complex |
| 23:21.49 | Malyce | oh, you mean the one from the Developer docs ? |
| 23:22.00 | brlcad | Malyce: also note that the uml diagrams that you read are likely out of date - but they do still have useful information |
| 23:22.05 | mafm | hippieindamakin8: I had the link around |
| 23:22.16 | brlcad | mostly on what those uml diagrams call the GeometryEngine is not entirely/still true |
| 23:22.27 | brlcad | and yes, the dev docs |
| 23:22.27 | hippieindamakin8 | mafm, madant_ would be elated if he hears this |
| 23:23.02 | hippieindamakin8 | neways i better complete editing the proposal. its already 4:55 am and got a class from 10. |
| 23:23.18 | Malyce | I read a todo list in the repository for Geometry Engine |
| 23:23.28 | mafm | Malyce: also mine http://brlcad.org/wiki/OpenGL_GUI_Framework (personal info, background and so on not included), was accepted last year |
| 23:23.45 | brlcad | hippieindamakin8: for references, I meant personal-portfolio information, like if you had a personal website where you show other projects you work on |
| 23:23.48 | brlcad | if not, no big deal |
| 23:23.49 | Malyce | It tells about what should be the end product, but not about the current issues |
| 23:23.51 | mafm | hippieindamakin8: get to work! :P :) |
| 23:23.52 | *** part/#brlcad jdoliner (n=jdoliner@c-68-51-76-57.hsd1.il.comcast.net) | |
| 23:24.14 | hippieindamakin8 | brlcad, i do have one. |
| 23:24.18 | hippieindamakin8 | i ll link it up |
| 23:24.20 | Malyce | Is there such a todo list, where I can find what part people are currently working ? |
| 23:25.20 | brlcad | Malyce: mostly by hanging out in irc and talking -- we don't keep anything documented like that because it is *constantly* changing for the most part |
| 23:26.36 | Malyce | I see. So to include a concrete list of what I can achieve, I should go through the current state of the repository ? |
| 23:26.57 | Malyce | I don't want to sound too vague, when I write a list of objectives. |
| 23:27.44 | brlcad | yeah, the more specific you can be, the better |
| 23:27.57 | brlcad | it is expected that you will have to learn much as you get started |
| 23:28.06 | brlcad | so pad for some of that time in your proposal |
| 23:28.19 | brlcad | there will have to be a lot of time for communication/discussion too throughout |
| 23:28.31 | brlcad | no student should be working in a vacuum |
| 23:29.08 | brlcad | as for the state of the repository, that's best understood by looking through what you can and asking questions -- I can probably answer a lot more quickly than you can find |
| 23:29.17 | brlcad | or at least point you in the right direction |
| 23:29.35 | brlcad | there's not much time left so there will be more time to tweak after the deadline |
| 23:29.48 | mafm | brlcad: the global illumination rendered was not done by andrecastelo? |
| 23:29.53 | mafm | renderer* even |
| 23:30.02 | Malyce | You said yesterday about the future merging of Core interface with Geometry Engine |
| 23:30.18 | Malyce | I have taken a look at both. I am still formulating some good objectives |
| 23:30.24 | Malyce | Any hints ? |
| 23:31.45 | brlcad | mafm: what do you mean? he worked on it, but he didn't finish |
| 23:32.20 | *** join/#brlcad AlexandreGuedes_ (n=chatzill@187-24-72-109.3g.claro.net.br) | |
| 23:32.32 | brlcad | in fact, he didn't get to the global illumination aspect, he got to basic flat shade ray-tracing |
| 23:32.47 | brlcad | Malyce: are you the one familiar with SW? |
| 23:32.56 | Malyce | yes |
| 23:33.01 | brlcad | okay |
| 23:33.11 | Malyce | I have been working on it for quite a while |
| 23:33.16 | Malyce | not as a modeller though |
| 23:33.23 | Malyce | only with VB and the API |
| 23:33.48 | brlcad | well to relate to their api a little bit -- if I remember their api correctly, their approach does things a little like if we merged our geometry service together with the geometry engine |
| 23:34.08 | brlcad | basically that you attach to their geometry engine and send it actions/queries to perform |
| 23:34.16 | mafm | ah ok, it'll be a WIP too |
| 23:35.01 | Malyce | Hmm, I thought core services' syntax resembled more like their API syntax |
| 23:35.20 | brlcad | the goal of the geometry engine is to separate service actions, access control, and revisioning from the core representation and capabilities of geometry |
| 23:35.21 | Malyce | I haven't seen any examples for either Geometry Engine or services |
| 23:35.39 | brlcad | there are some coreinterface examples on the wiki |
| 23:35.43 | Malyce | Yes |
| 23:35.46 | Malyce | I have seen |
| 23:35.49 | brlcad | k |
| 23:35.54 | Malyce | but not for Geometry engine/services |
| 23:36.06 | Malyce | Are there any ? |
| 23:36.33 | brlcad | for all intensive purposes GeometryEngine==coreInterface |
| 23:36.51 | brlcad | geometry service is not involved |
| 23:37.01 | Malyce | I think I am beginning to understand. |
| 23:37.25 | Malyce | So, when you said you wanted to merge core interfaces with geometry engine, they both complement each other |
| 23:37.25 | mafm | brlcad, from your ideas list: Create a 3D geometry browser graphical interface using a common graphics engine (e.g. OGRE, Open Scene Graph, Crystal Space) -- isn't this the GUI that we've been making? |
| 23:37.33 | brlcad | the GeometryService *project* required the dev that was working on it to define some aspects of a GeometryEngine before one existed, before coreInterface was even to the point it is at now |
| 23:37.35 | Malyce | instead of substituting each other |
| 23:38.08 | Malyce | ok |
| 23:38.13 | brlcad | so the GS dev ended up duplicating some of the same design that was implemented by the other dev that was working on the coreInterface (d_rossberg) |
| 23:38.27 | brlcad | so they should be merged |
| 23:38.46 | brlcad | I think coreInterface has a lot more design to it already, but still has a LONG way to go |
| 23:38.51 | Malyce | So, is the plan to extend core interfaces, till it converges with Geometry engine ? |
| 23:38.58 | brlcad | yeah, sorta |
| 23:39.06 | Malyce | Sound cool |
| 23:39.17 | brlcad | in my mind, coreInterface is the GE but with a bad name :) |
| 23:39.27 | Malyce | Is the wiki out of date ? |
| 23:39.39 | brlcad | but then there are some aspects in the currently GS/GE-named GE that need to merge with it |
| 23:39.45 | brlcad | not horribly no |
| 23:39.58 | brlcad | that's where things are at -- a lot of work remains |
| 23:40.36 | brlcad | so probably your first step would be to work with d_rossberg and d-lo to merge the two or find the best way to consolidate them |
| 23:41.14 | Malyce | So core interfaces needs to cover all the standard libraries that BRL devs work with ? |
| 23:41.24 | Malyce | Should I put that down as an objective ? |
| 23:41.36 | brlcad | then itemize a few specific goals, like maybe adding support for all of the primitive types, or maybe defining brep entities, or maybe adding hierarchy traversal support, etc |
| 23:42.04 | brlcad | brl-cad's core libraries are presently LIBBU, LIBBN, LIBRT, LIBWDB, and LIBGED |
| 23:42.20 | brlcad | the Geometry Engine is an OO layer that sits on top of that functionality |
| 23:42.45 | brlcad | and uses those libraries to provide functionality |
| 23:43.05 | Malyce | Pardon me. I understand some of your lingo, but others I don't get: Defining brep entities ? |
| 23:43.28 | brlcad | depending on what portion of the API you work on, the task may even involve some refactoring -- pushing functionality that is in libged into librt, for example, so that the GE can use it more effectively |
| 23:43.40 | brlcad | brep is "boundary representation" |
| 23:43.52 | brlcad | http://en.wikipedia.org/wiki/Boundary_representation |
| 23:44.21 | Malyce | Hierarchy transversal support has to do with inheritance ? Or does it have something to do with model parts ? |
| 23:44.23 | brlcad | http://en.wiki.mcneel.com/content/upload/images/Brep_structure.gif <-- pretty good image |
| 23:44.29 | brlcad | albeit detailed |
| 23:44.41 | brlcad | model parts .. traversing down a model hierarchy |
| 23:44.48 | brlcad | doing things while you traverse |
| 23:44.52 | brlcad | it's a common operation in librt |
| 23:45.00 | Malyce | That seems like something I am much more familiar with |
| 23:45.01 | brlcad | it becomes a set of iterators in OO |
| 23:45.16 | Malyce | Sounds simple |
| 23:45.30 | brlcad | that's why one of daniel's examples is traversal iirc -- you can see the difference |
| 23:45.34 | brlcad | looks for the link |
| 23:46.27 | brlcad | http://brlcad.org/wiki/Example_db_walk_tree |
| 23:46.40 | brlcad | that uses librt to traverse a geometry file and print all regions |
| 23:47.02 | Malyce | Thanks a ton brlcad |
| 23:47.11 | Malyce | this was exactly the sort of thing I was looking for |
| 23:47.16 | brlcad | this is similar but with the C++ interface as it currently stands: http://brlcad.org/wiki/CoreInterface_Tree_Walker_Example |
| 23:47.59 | brlcad | notes that needs lots of improvement :) |
| 23:48.43 | brlcad | no reason the C++ version should be that much more text than the C version :) |
| 23:48.57 | brlcad | though in all fairness, they do slightly different things |
| 23:49.01 | Malyce | ok, dumb question maybe, but is it not possible to save the file, and just run the db_walk to get the same result ? |
| 23:49.13 | brlcad | hm? |
| 23:49.26 | Malyce | It seems you have hierarchy transversal in the geometry files, but not in core interfaces, is it not ? |
| 23:49.28 | brlcad | save what file? |
| 23:49.46 | Malyce | I assumed the first example was transversal inside a stored database file |
| 23:50.11 | brlcad | our geometry database files (our .g file format) contain geometry with hierarchical information |
| 23:50.27 | Malyce | yes |
| 23:50.46 | brlcad | say you open a file and want to list all objects, that's easy enough, but say you want to print the hierarchy itself, then you need to traverse over the objects in their hierarchy order |
| 23:50.49 | Malyce | so, why can't a dirty shortcut like the one I mentioned above be used ? |
| 23:51.01 | Malyce | oh |
| 23:51.22 | Malyce | So the hierarchy isn't preserved ? |
| 23:51.38 | brlcad | of course it's preserved, but it's just bits on a disk |
| 23:51.53 | Malyce | so ? |
| 23:52.07 | brlcad | so I think you're missing something :) |
| 23:52.35 | brlcad | the API in this case is a routine to walk down a tree |
| 23:52.42 | Malyce | Is this the difference between doing it in realtime, and retrieving the info from storage ? |
| 23:52.48 | brlcad | on the C side in librt, one of those functions is db_walk_tree() |
| 23:53.12 | brlcad | you provide various callbacks to db_walk_tree() to tell it what to do when it encounters various object types |
| 23:53.26 | brlcad | your callbacks are invoked, and your application does what it needs to |
| 23:54.09 | brlcad | that's very much a procedural method of iteration, what one would generally expect with a C api |
| 23:54.25 | brlcad | for the C++ api, I would generally expect something else |
| 23:55.00 | Malyce | like ? |
| 23:55.21 | Malyce | Just something higher level ? |
| 23:55.36 | brlcad | which is what is shown in that example for the coreInterface -- there it defines a TopObjectIterator along with a callback, and as it iterates over 'good' objects (wtf), it calls that callback |
| 23:56.31 | brlcad | expectation is what it does -- iterator pattern instead of procedural callbacks |
| 23:58.35 | Malyce | And we can't have the same thing in C ? |
| 23:58.43 | Malyce | Is the difference in Classes ? |
| 23:58.54 | Malyce | iterators are not possible in C ? |
| 23:59.27 | mafm | languages with OO capabilities are cleaner in some ways, as this one :) |
| 23:59.46 | brlcad | mafm: right, there are no iterators |
