| 00:30.55 | ``Erik | interesting |
| 00:31.22 | ``Erik | "and sometimes osx" |
| 02:57.05 | brlcad | downloaded the osx binary, no go (must be intel me thinks, at a ppc atm) |
| 06:08.41 | *** join/#brlcad clock_ (n=clock@77-56-92-186.dclient.hispeed.ch) |
| 07:12.23 | *** join/#brlcad clock_ (n=clock@zux221-122-143.adsl.green.ch) |
| 07:45.03 | *** join/#brlcad d_rossberg (n=rossberg@bz.bzflag.bz) |
| 08:16.18 | *** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6) |
| 09:48.58 | *** join/#brlcad mafm (n=mafm@elnet-111.lip.pt) |
| 09:49.23 | mafm | hallo |
| 09:56.47 | *** join/#brlcad Elperion (n=Bary@p5B14CA61.dip.t-dialin.net) |
| 10:01.14 | *** join/#brlcad dtidrow (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net) |
| 10:42.11 | *** join/#brlcad elite01 (n=elite01@dslb-088-071-040-234.pools.arcor-ip.net) |
| 11:15.42 | *** join/#brlcad Elperion (n=Bary@p5B14CA61.dip.t-dialin.net) [NETSPLIT VICTIM] |
| 11:15.42 | *** join/#brlcad archivist (n=archivis@host81-149-119-172.in-addr.btopenworld.com) [NETSPLIT VICTIM] |
| 11:15.42 | *** join/#brlcad poolio (n=poolio@bz.bzflag.bz) [NETSPLIT VICTIM] |
| 11:15.49 | *** join/#brlcad b0ef (n=b0ef@062016141231.customer.alfanett.no) [NETSPLIT VICTIM] |
| 11:17.05 | *** join/#brlcad brlcad (n=sean@pdpc/supporter/silver/brlcad) |
| 11:17.05 | *** join/#brlcad yukonbob (i=1000@s142-179-54-198.bc.hsia.telus.net) [NETSPLIT VICTIM] |
| 11:17.05 | *** join/#brlcad starseeker (n=starseek@bz.bzflag.bz) [NETSPLIT VICTIM] |
| 11:17.05 | *** join/#brlcad cosurgi (i=janek@irc.cool.waw.pl) [NETSPLIT VICTIM] |
| 11:17.05 | *** mode/#brlcad [+o brlcad] by irc.freenode.net |
| 11:17.06 | *** join/#brlcad elite01 (n=elite01@dslb-088-071-040-234.pools.arcor-ip.net) [NETSPLIT VICTIM] |
| 11:17.06 | *** join/#brlcad dtidrow (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net) [NETSPLIT VICTIM] |
| 11:17.18 | *** join/#brlcad ChanServ (ChanServ@services.) |
| 11:17.18 | *** join/#brlcad ``Erik (i=erik@c-68-54-174-162.hsd1.md.comcast.net) [NETSPLIT VICTIM] |
| 11:17.18 | *** join/#brlcad clock_ (n=clock@zux221-122-143.adsl.green.ch) [NETSPLIT VICTIM] |
| 11:17.18 | *** join/#brlcad mafm (n=mafm@elnet-111.lip.pt) [NETSPLIT VICTIM] |
| 11:17.18 | *** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6) [NETSPLIT VICTIM] |
| 11:17.18 | *** join/#brlcad d_rossberg (n=rossberg@bz.bzflag.bz) [NETSPLIT VICTIM] |
| 11:17.19 | *** join/#brlcad PrezKennedy (i=Matthew@whitecalf.net) [NETSPLIT VICTIM] |
| 11:17.19 | *** join/#brlcad CIA-22 (n=CIA@208.69.182.149) [NETSPLIT VICTIM] |
| 11:17.19 | *** join/#brlcad alex_joni (n=juve@emc/board-of-directors/alexjoni) [NETSPLIT VICTIM] |
| 11:17.19 | *** mode/#brlcad [+o ChanServ] by irc.freenode.net |
| 11:18.56 | *** join/#brlcad punkrockgirl (n=Pandora@c-69-243-244-154.hsd1.mo.comcast.net) |
| 11:18.57 | *** join/#brlcad MinuteEl1ctron (n=MinuteEl@bz.bzflag.bz) |
| 11:18.57 | *** join/#brlcad vedge (n=vedge@205-237-251-204.ilesdelamadeleine.ca) |
| 11:23.37 | *** join/#brlcad elite01 (n=elite01@dslb-088-071-040-234.pools.arcor-ip.net) [NETSPLIT VICTIM] |
| 11:23.37 | *** join/#brlcad dtidrow (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net) [NETSPLIT VICTIM] |
| 11:35.16 | *** join/#brlcad vedge (n=vedge@205-237-251-204.ilesdelamadeleine.ca) |
| 12:07.43 | *** join/#brlcad elite01 (n=elite01@dslb-088-071-040-234.pools.arcor-ip.net) [NETSPLIT VICTIM] |
| 12:07.43 | *** join/#brlcad dtidrow (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net) [NETSPLIT VICTIM] |
| 12:14.46 | CIA-22 | BRL-CAD: 03homovulgaris * r31550 10/brlcad/trunk/ (1124 files in 80 dirs): Boost (header only include library) initial import, using bcp |
| 12:32.59 | brlcad | heh, wow, that's big for header only :) |
| 12:33.14 | brlcad | initial import of what? |
| 12:42.27 | mafm | brlcad: did you have the chance to test the branch |
| 12:42.28 | mafm | ? |
| 13:19.05 | *** join/#brlcad docelic (n=docelic@78.134.197.96) |
| 13:21.07 | *** join/#brlcad pacman87 (n=Timothy@71.170.63.120) |
| 13:27.52 | pacman87 | morning, all |
| 13:31.28 | mafm | 'allo pacman87 :) |
| 13:36.24 | d_rossberg | morning pacman97 |
| 13:37.54 | pacman87 | d_rossberg: i've looked into the sketch primitive, and there's 4 types of segments to deal with: line segment, circular arc, bezier spline, and nurbs |
| 13:40.04 | pacman87 | for now, i'm thinking it would be best to limit the order of the splines to be able to solve intersetions analytically |
| 13:40.20 | d_rossberg | right, and in the extrude the nurb isn't implemented, probable because it isn't trivial |
| 13:41.20 | pacman87 | so my overall plan is to start with just line segments, and get them working, before moving on the the more complex ones |
| 13:41.23 | d_rossberg | i would say you have to multiply the degree of the sketch with the one of the spline to get the order of the surface |
| 13:41.48 | pacman87 | 'degree of the sketch'? |
| 13:42.28 | d_rossberg | i mean the degree of the segments (line is 1, arc is 2 etc.) |
| 13:43.09 | pacman87 | and the spline is for sweeping |
| 13:43.20 | pacman87 | sorry, i'm still just thinking revolve at the moment |
| 13:44.17 | d_rossberg | it depends on where your priorities are, i would start with a degree 2 sketch and degree 2 spline (or revolve) |
| 13:44.31 | d_rossberg | i.e. both build from 3 point beziers |
| 13:45.02 | pacman87 | so do you think revolve should only be a subset of sweep? |
| 13:45.35 | pacman87 | like how sph is really an ell? |
| 13:45.50 | d_rossberg | you can reach a revolve effect from a sweep on a circle |
| 13:46.13 | d_rossberg | however, i like the idea to have a distinct revolve primitive |
| 13:46.38 | pacman87 | right, because then i can make revolve-specific performance enhancements |
| 13:46.47 | pacman87 | that dont' necessarliy apply to all sweeps |
| 13:48.38 | CIA-22 | BRL-CAD: 03mafm * r31551 10/rt^3/trunk/src/other/rbgui/Gui/Source/PosixPlatformManager.cpp: Fix for lame error preventing the fix for autorepeat behaviour to work -- the huge amounts of unimportant warnings were masking this. |
| 13:50.19 | d_rossberg | if you once got the revolve you could perhaps develop the sweep from it like: "revolves" along a spline |
| 13:50.28 | pacman87 | one problem i see is determine which segments to test for intersection, or to test all of them |
| 13:51.12 | d_rossberg | in a first attempt: all segments |
| 13:52.00 | pacman87 | my general plan is to essentially take the 'reverse' of the sweep path, and apply that transformation to the ray, and project the ray onto the 2d sketch surface |
| 13:52.45 | pacman87 | ie, if you 'unbend' a revolve, the ray becomes a hyperbola |
| 13:53.56 | d_rossberg | your plan sounds still good :) |
| 13:54.02 | pacman87 | and looking toward sweep, is there already a way to store a 3d spline, or will i have to creat something for that specifically? |
| 13:57.09 | pacman87 | should the 3d path spline be created/edited seperately, or inside the sweep only? |
| 13:58.26 | d_rossberg | store the spline: hard to say, it depends on the spline, there is the bu_list but this may be not optimal |
| 13:58.51 | pacman87 | could that be its own primitive, like sketch? |
| 13:59.55 | pacman87 | so if you want two parallel sweeps, you specify the same sketch and path, and change the vertex |
| 14:00.30 | d_rossberg | it could be an own primitive but this also means that it is its own live in some sense |
| 14:00.50 | d_rossberg | (3D sketch) |
| 14:01.32 | d_rossberg | you could get two parallel sweeps by copying the primitive ... |
| 14:02.54 | pacman87 | if you copy, and have to edit the path, will those changes affect both? |
| 14:03.13 | d_rossberg | no |
| 14:04.09 | pacman87 | and if you wanted that behavior, you could copy the path |
| 14:04.13 | d_rossberg | but you could add a primitive twice to the geometry with different matrices |
| 14:06.37 | pacman87 | what problems do you see with storing the path seperately? |
| 14:07.06 | d_rossberg | but you have a point there: i would be a nice feature to be able to sweep different shapes along a sweep line |
| 14:07.41 | d_rossberg | with the sweep path the problem might not be so big: |
| 14:08.24 | d_rossberg | because you define it, you may define it in a way you won't get any problems with your sweep primitive |
| 14:09.08 | d_rossberg | i.e. "a 3D sketch has at most degree 2 spline segments" |
| 14:09.58 | pacman87 | would there be a practical use for a 3d sketch besides the sweep? |
| 14:11.18 | d_rossberg | i would expect it to have a practical use because the large CAD packages have this |
| 14:12.22 | pacman87 | the only thing i remember using 3d sketches for is for sweeps/lofts |
| 14:13.33 | pacman87 | but if there are other uses, i'd be better to not restrict the path, and just have a flag for isSweepable |
| 14:14.27 | d_rossberg | the sketches in BRL-CAD have a "comment" role, but i can't see why these "comments" should be 2D only |
| 14:14.57 | pacman87 | wasnt' that on the list of improvements? |
| 14:17.25 | d_rossberg | i don't like the idea with the flag, this would make it possible to destroy the sweeps inadvertently |
| 14:18.22 | pacman87 | while editing the path? |
| 14:19.03 | d_rossberg | i'm not sure but could the parametrics/constraint help? |
| 14:19.58 | pacman87 | i think the p/c stuff is about physical relationships, and the problem here is with the math behind it all |
| 14:21.02 | pacman87 | it's also possible to set the isSweepable flag when you create the path, and require the user to explicitly change it |
| 14:21.10 | pacman87 | so if it's set, it doesnt allow changes to break it |
| 14:21.33 | pacman87 | but then that becomes close to a whole different primitive |
| 14:21.45 | pacman87 | ie, path vs 3dsketch |
| 14:22.32 | d_rossberg | the is-a relationship works only in one direction |
| 14:23.02 | pacman87 | path is a sketch3d |
| 14:23.13 | d_rossberg | right |
| 14:23.16 | pacman87 | with additional restrictions |
| 14:25.24 | d_rossberg | an other solution would be to have default replacements for "unknown" segments (e.g. build a spline of line segments from a nurb curve) |
| 14:26.42 | pacman87 | using a bezier curve would require fewer segments |
| 14:26.54 | pacman87 | and keep the surface curvature better |
| 14:27.35 | pacman87 | since i'll have to check each segment in the sketch |
| 14:27.45 | d_rossberg | only if you have an algorithm for converting a nurb into a bezier spline |
| 14:27.56 | pacman87 | true :) |
| 14:29.33 | pacman87 | the other issue with is-a is that C doesnt really do inheritance |
| 14:30.22 | pacman87 | i suppose it'd be done using the functab |
| 14:31.06 | d_rossberg | you may get the beziers if you attach the border conditions to the line segments |
| 14:32.06 | d_rossberg | functab: this is more an replacement for a switch ... case |
| 14:33.05 | d_rossberg | i.e. switch (object_type) ... case hyp: ... |
| 14:34.14 | pacman87 | well, if path just references the sketch3d functions when it doesnt' need to change them, and references its own functions when there are changes |
| 14:38.38 | d_rossberg | this would only mean you are reusing code but actual you have two diferent primitives |
| 14:39.47 | pacman87 | is that a problem? |
| 14:41.45 | d_rossberg | reusing code isn't a problem but it looks like you should care for the path and if there is a 3dsketch it might reuse your code |
| 14:42.51 | d_rossberg | we would need something like a validator connected to the primitives |
| 14:43.14 | clock_ | win 12 |
| 14:45.14 | *** join/#brlcad prasad_ (n=psilva@h-67-103-183-185.mclnva23.covad.net) |
| 14:45.57 | pacman87 | so just get the path working now, and worry about sketch3d later |
| 14:46.07 | d_rossberg | this validator could be implemented as a new functab method (if this don't go to far away) |
| 14:46.21 | d_rossberg | path vs. sketch3d: yes |
| 14:46.25 | pacman87 | what would be validated? |
| 14:46.34 | d_rossberg | the primitive |
| 14:46.37 | pacman87 | order of sketch and path? |
| 14:47.09 | d_rossberg | a validator is a funktion which returns for a primitive true or false |
| 14:47.25 | d_rossberg | true if it's ok, false if not |
| 14:47.31 | pacman87 | ah, ok |
| 14:48.12 | d_rossberg | i.e. an editor has to call the validator to test the changed primitive |
| 14:48.24 | pacman87 | like how primitives do basic checking before prep() and others |
| 14:48.40 | pacman87 | but moved to it's own function |
| 14:48.43 | pacman87 | make sense |
| 14:48.46 | pacman87 | makes* |
| 14:50.56 | d_rossberg | but in practice it's not so easy e.g. there should be a hint to the user if the validator fails |
| 14:57.59 | pacman87 | right |
| 14:58.43 | pacman87 | i think i've answered your email question, is there anything else i missed? |
| 14:59.51 | pacman87 | i'll update the wiki for revolve, and create a page for sweep to outline everything |
| 15:06.12 | d_rossberg | this is what i wanted to ask you for: document your plans for revolve and sweep on the brlcad.org wiki |
| 15:07.09 | d_rossberg | i'm full of hope :) |
| 15:15.45 | *** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6) |
| 16:15.57 | CIA-22 | BRL-CAD: 03mafm * r31552 10/rt^3/trunk/src/g3d/ (5 files): Adding window which will act as taskbar. Only the skeleton, at the moment. |
| 16:16.20 | *** join/#brlcad archivist (n=archivis@host81-149-119-172.in-addr.btopenworld.com) |
| 16:22.38 | CIA-22 | BRL-CAD: 03mafm * r31553 10/rt^3/trunk/src/g3d/GuiTaskbar.h: Fixing doxygen comments. |
| 16:45.23 | *** join/#brlcad prasad1 (n=psilva@h-72-245-122-226.mclnva23.covad.net) |
| 16:58.50 | *** join/#brlcad pacman87 (n=timothy@71.170.63.120) |
| 17:14.56 | *** join/#brlcad dtidrow_ (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net) |
| 17:39.30 | CIA-22 | BRL-CAD: 03mafm * r31554 10/rt^3/trunk/src/g3d/CMakeLists.txt: Removing debugging messages |
| 17:40.36 | CIA-22 | BRL-CAD: 03starseeker * r31555 10/brlcad/trunk/doc/docbook/ (17 files in 4 dirs): Add includes and graphics through Lesson 5 |
| 17:41.42 | CIA-22 | BRL-CAD: 03starseeker * r31556 10/brlcad/trunk/doc/docbook/lessons/mged/learning_boolean_expressions.xml: Whoopsie, add the xml file as well as the images |
| 17:48.46 | CIA-22 | BRL-CAD: 03mafm * r31557 10/rt^3/trunk/src/g3d/ (8 files): |
| 17:48.46 | CIA-22 | BRL-CAD: Introducing Window Base class, so the rest of the windows are derived from it. |
| 17:48.46 | CIA-22 | BRL-CAD: This allows to have some operations factored out, and to have lists of windows |
| 17:48.46 | CIA-22 | BRL-CAD: present and so on (which is immediately useful for the taskbar, in example). |
| 18:05.28 | CIA-22 | BRL-CAD: 03mafm * r31558 10/rt^3/trunk/src/g3d/ (6 files): Delegating more functionality in the windows base class. |
| 18:08.34 | mafm | brlcad, are you back? |
| 18:15.19 | pacman87 | what's the best (ie, user friendly) way to specify the start/end planes of the revolve? |
| 18:17.46 | starseeker | Are your start/end planes constrained relative to each other? |
| 18:18.18 | pacman87 | i know i need the revolve axis (point + vector) |
| 18:18.37 | pacman87 | and a way to specify which plane passing though that axis is the start plane |
| 18:20.08 | pacman87 | they could be constrained relativly by giving the first plane, and an agnle |
| 18:20.09 | pacman87 | angle |
| 18:20.26 | starseeker | Hmm - I'd guess for full generality you'd need to specify another vector perpendicular to the revolve axis, but that's just off the cuff |
| 18:20.36 | pacman87 | that's what i was thinking too |
| 18:20.46 | starseeker | yes, once the start plane is identified the end plane could be identified by an angle |
| 18:21.03 | pacman87 | would you perfer an angle, or a second vector to define the end plane? |
| 18:21.08 | starseeker | angle |
| 18:21.09 | pacman87 | ok |
| 18:21.38 | starseeker | you might optionally take a second vector, if there is a way to work it |
| 18:22.11 | pacman87 | i dont' think you can overload 'in' |
| 18:22.53 | pacman87 | though i suppose once the p/c is fully integrated with the primitives, you could jsut use that |
| 18:23.11 | *** join/#brlcad dtidrow_ (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net) |
| 18:23.40 | starseeker | well, you could do like ell1 and ell2 |
| 18:24.05 | starseeker | er ell and ell1 rather |
| 18:24.16 | starseeker | have a sweep and sweep1 primitive |
| 18:24.33 | pacman87 | what should i name the revolve? |
| 18:24.35 | pacman87 | rev? |
| 18:24.54 | starseeker | iirc, ell and ell1 exist because they take different parameters for the same geometric shape |
| 18:25.33 | starseeker | hmm. rev is less typing, but extrude uses the full name... |
| 18:25.49 | pacman87 | but ext is a filesystem |
| 18:25.59 | starseeker | and rev is revision :-) |
| 18:26.03 | pacman87 | right |
| 18:26.17 | pacman87 | rvl |
| 18:26.35 | starseeker | It's a brlcad question, but my sense would be to go ahead and use revolve to keep it consistent with extrude |
| 18:26.49 | pacman87 | and sweep instead of swp |
| 18:26.57 | pacman87 | swp looks like swap |
| 18:27.18 | starseeker | right - extrude needs a sketch, and so will revolve/sweep yes? |
| 18:27.23 | pacman87 | yes |
| 18:28.25 | pacman87 | i'll have to specify the point/axis in both the 3D side, and the 2D sketch side |
| 18:28.30 | starseeker | I think pipe is the only non-sketch based primitive with a full name... |
| 18:28.42 | starseeker | yeah, check with brlcad |
| 18:28.44 | pacman87 | sketch is too |
| 18:29.06 | starseeker | heh - is sketch sketch based? :-P |
| 18:29.41 | starseeker | probably not, strictly speaking |
| 18:29.49 | pacman87 | no, it doesnt' reference a sketch |
| 18:30.07 | starseeker | My instinct is to go with full names as being more descriptive, now that we have tab completion |
| 18:30.26 | starseeker | although I don't know if that works for primitive type, come to think of it |
| 18:32.52 | CIA-22 | BRL-CAD: 03mafm * r31559 10/rt^3/trunk/src/g3d/ (GuiConsole.cxx GuiConsole.h): Properly center console vertically |
| 18:57.08 | CIA-22 | BRL-CAD: 03mafm * r31560 10/rt^3/trunk/src/g3d/ (GuiTaskbar.cxx GuiTaskbar.h): Implementing the buttons in the task bar (not complete) |
| 18:57.19 | mafm | I have to go now, see you guys |
| 18:58.40 | CIA-22 | BRL-CAD: 03starseeker * r31561 10/brlcad/trunk/doc/docbook/ (10 files in 3 dirs): Add lesson 6 |
| 19:04.16 | CIA-22 | BRL-CAD: 03bob1961 * r31562 10/brlcad/trunk/ (12 files in 5 dirs): Remove tcl.c from libged. Added ged_obj.c to libtclcad. Call Ged_Init from Tclcad_Init. Added struct ged_obj to tclcad.h. |
| 19:19.06 | CIA-22 | BRL-CAD: 03bob1961 * r31563 10/brlcad/trunk/src/libged/ (CMakeLists.txt Makefile.am): Added ged.c to Makefile.am. Added erase.c and erase_all.c to CMakeLists.txt |
| 19:21.08 | CIA-22 | BRL-CAD: 03starseeker * r31564 10/brlcad/trunk/doc/docbook/ (7 files in 3 dirs): Add Lessson 7 |
| 19:40.57 | *** join/#brlcad elite01 (n=elite01@dslb-088-071-040-234.pools.arcor-ip.net) |
| 19:51.53 | *** join/#brlcad dtidrow (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net) |
| 20:04.10 | CIA-22 | BRL-CAD: 03starseeker * r31565 10/brlcad/trunk/doc/docbook/ (9 files in 3 dirs): Add Lesson 8 |
| 20:33.39 | CIA-22 | BRL-CAD: 03starseeker * r31566 10/brlcad/trunk/doc/docbook/ (9 files in 3 dirs): Add Lesson 9 |
| 20:33.58 | *** join/#brlcad Elperion (n=Bary@p5B14CA61.dip.t-dialin.net) |
| 20:34.24 | starseeker | attempts to uncross eyeballs |
| 20:51.41 | CIA-22 | BRL-CAD: 03starseeker * r31567 10/brlcad/trunk/doc/docbook/ (4 files in 3 dirs): Add Lesson 10 |
| 20:55.28 | CIA-22 | BRL-CAD: 03bob1961 * r31568 10/brlcad/trunk/ (6 files in 4 dirs): Split out the ls command from wdb_obj. Added the ls command to ged_obj. |
| 21:06.40 | CIA-22 | BRL-CAD: 03starseeker * r31569 10/brlcad/trunk/doc/docbook/ (5 files in 3 dirs): Add Lesson 11 |
| 22:05.00 | CIA-22 | BRL-CAD: 03starseeker * r31570 10/brlcad/trunk/doc/docbook/ (8 files in 3 dirs): Add Lesson 12 |
| 22:37.58 | CIA-22 | BRL-CAD: 03starseeker * r31571 10/brlcad/trunk/doc/docbook/ (12 files in 3 dirs): Add Lesson 13 |
| 23:09.43 | CIA-22 | BRL-CAD: 03starseeker * r31572 10/brlcad/trunk/doc/docbook/ (11 files in 3 dirs): Add Lesson 14 |
| 23:30.32 | pacman87 | so there's a 2d point, but no 2d vector? |
| 23:46.46 | pacman87 | make[2]: *** No rule to make target `solver_test.c', needed by `solver_test.o'. Stop. |
| 23:49.21 | poolio | pacman87: distclean :) I had that same issue |
| 23:49.36 | pacman87 | poolio: thanks |
| 23:53.33 | CIA-22 | BRL-CAD: 03starseeker * r31573 10/brlcad/trunk/doc/docbook/ (15 files in 3 dirs): Add Lesson 15 |