| 00:44.18 | maths22 | the server has uncrazyified |
| 01:05.49 | Ignacio` | Hi, any gci winner here? :P |
| 01:20.00 | *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14) | |
| 01:59.43 | *** part/#brlcad Ignacio` (~IgnacioUy@unaffiliated/ignaciouy) | |
| 02:30.22 | starseeker | brlcad: am I correct that the bu_external also holds all the attributes for a given object? |
| 02:31.11 | starseeker | or do I need to pull those separately? (doing a basic, fast diff/no-diff test for db_diff) |
| 02:44.11 | Notify | 03BRL-CAD:starseeker * 59384 brlcad/trunk/src/librt/CMakeLists.txt: Start working out how to do db_diff version of diff functionality. |
| 02:45.37 | Notify | 03BRL-CAD:starseeker * 59385 (brlcad/trunk/include/db_diff.h brlcad/trunk/src/librt/db_diff.c): Include the header to make sure we're following the template |
| 04:21.54 | *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14) | |
| 05:21.41 | *** join/#brlcad KimK (~Kim__@ip24-255-223-153.ks.ks.cox.net) | |
| 05:48.33 | brlcad | starseeker: yes, they're serialized after the object data |
| 05:54.45 | *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14) | |
| 06:46.59 | *** join/#brlcad Ch3ck_ (~Ch3ck@195.24.220.16) | |
| 08:35.22 | *** join/#brlcad merzo (~merzo@125-81-132-95.pool.ukrtel.net) | |
| 09:18.14 | *** join/#brlcad merzo (~merzo@8-152-132-95.pool.ukrtel.net) | |
| 09:37.03 | *** join/#brlcad Anaphaxeton (~george@unaffiliated/anaphaxeton) | |
| 10:47.41 | *** join/#brlcad chick_ (~chick_@195.24.220.16) | |
| 12:28.01 | Notify | 03BRL-CAD:indianlarry * 59386 (brlcad/branches/nurbs/AUTHORS brlcad/branches/nurbs/BUGS and 3989 others): Merging changes from trunk into branch 'nurbs' r:58142:59385 |
| 14:12.59 | *** join/#brlcad jschulte (5dc6fced@gateway/web/freenode/ip.93.198.252.237) | |
| 14:26.20 | brlcad | maths22: how is fisheye started? |
| 14:26.40 | brlcad | need to set it up as a proper system service so that it can be started/stopped on demand |
| 14:27.06 | brlcad | killed it yesterday while debugging the server connectivity issue (ended up being a bad router at the ISP) |
| 15:39.25 | starseeker | is buildbot still up and working on the server? |
| 15:39.48 | starseeker | may take a stab at experimenting with it sometime in the next week or so if it's still up |
| 15:58.53 | Notify | 03BRL-CAD:starseeker * 59387 brlcad/trunk/CMakeLists.txt: If we ever enable Bob's osg work or do similar work, we'll need the find_package for OpenSceneGraph |
| 15:59.44 | Notify | 03BRL-CAD:starseeker * 59388 (brlcad/trunk/include/db_diff.h brlcad/trunk/src/librt/db_diff.c): If I'm not mistaken, there's no point in passing around a void pointer to data in the visiting functions if we can't introduce that data in the db_diff function itself. |
| 16:09.32 | Notify | 03BRL-CAD:brlcad * 59389 brlcad/trunk/src/libgcv/NOTES: add my thoughts on architecture design tenants |
| 16:22.29 | Notify | 03BRL-CAD:brlcad * 59390 brlcad/trunk/src/libgcv/NOTES: add a few comments on the design. probably the start of a library foreward. |
| 16:22.57 | brlcad | starseeker: ah right, there needed to be a void* there as well |
| 16:38.27 | *** join/#brlcad Ch3ck_ (~Ch3ck@41.205.13.138) | |
| 16:41.17 | starseeker | brlcad: re: libgcv NOTES - how do we specify a specific format for input and output streams to a generic convert function without having any defines or enums that identify those formats? take a char* string and do some kind of internal string->int mapping? |
| 16:44.48 | Notify | 03BRL-CAD:starseeker * 59391 brlcad/trunk/src/libgcv/NOTES: Note that Coin3D is now BSD licensed for the inventor section |
| 16:50.46 | maths22 | brlcad: /usr/home/www/fisheye/bin/start.sh |
| 16:57.32 | Notify | 03BRL-CAD:brlcad * 59392 brlcad/trunk/src/libgcv/NOTES: there's easily dozens more that would be interesting to incorporate |
| 17:02.53 | brlcad | starseeker: the MagickCore library (Image Magick's core library) pretty much set the stage when it comes to conversion convention |
| 17:04.02 | starseeker | nods - OK, I'll have a look at some point to see how they handle overriding file extensions |
| 17:04.47 | brlcad | basically, the type is inferred by the "file name" and there's a simple override convention for the case where a choice is ambiguous or different than what it seems |
| 17:05.40 | brlcad | iirc, it's a simple prefix that can be implicit or explicit like: convert file.png file2.jpg |
| 17:06.11 | brlcad | or: convert jpg:my_jpeg_file png:file.jpg |
| 17:06.25 | starseeker | nods - my concern was how to correctly interpret an override if we don't have advance knowledge of what file formats are handled - is that where the plugin system comes in? |
| 17:06.36 | brlcad | each format will still have to register itself |
| 17:06.41 | starseeker | ah, right |
| 17:07.00 | brlcad | included in that registration is some label/extension or other way to name that type |
| 17:07.56 | brlcad | I see it going down that each format just defines a unique label for that extension (which represents any number of formats it supports), which is the label you'd use |
| 17:08.13 | starseeker | makes sense |
| 17:08.14 | brlcad | note that many geometry file formats are directories of files too, not just a single file |
| 17:09.42 | starseeker | (if anybody feels like a small code cleanup/reduction chore, it looks like libged's get_types.c could benefit from replacing that big switch statement (in most cases) with callling intern.idb_meth->ft_label ) |
| 17:09.57 | brlcad | I could even see creating a "zip" format extension, for example, that just unpacks a zipfile and passes it back again to the library for open so you could do things like: gcv myfile.zip file.g |
| 17:11.42 | brlcad | the only complexity that will eventually be required is all of the various options that one might want to specify and how to get that information into the library as conversion preferences/rules |
| 17:12.12 | starseeker | nods |
| 17:12.17 | brlcad | like conversion tolerances, whether to preserve solidity, what representation format to export if multiple options are available, etc |
| 17:12.52 | brlcad | later can look into handling units, coordinate systems, feature/object suppression, healing, ... |
| 17:13.27 | brlcad | there's barely enough time to get: gcv file.g file.obj working cleanly for the five formats identified |
| 17:27.19 | *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14) | |
| 17:41.40 | *** join/#brlcad Ch3ck_ (~Ch3ck@41.205.13.38) | |
| 17:44.17 | ``Erik | starseeker: last I checked, both buildbot and jenkins were chugging along... if not, lemme know O.o (as far as using them, you're on your own, but I'll do the sysadminy stuff) |
| 17:46.37 | starseeker | ``Erik: do I have the permission I would need to work with them? |
| 17:50.13 | ``Erik | uhhhhhh, idunno? O.o |
| 17:51.22 | ``Erik | looks like gcimoin set up jenkins, which is a java based one, so he'd be sme |
| 17:53.00 | ``Erik | buildbot is python based iirc, I'm offering service in the sysadmin role, not the tool usage role :D ya'll gotta figure out how to use it, I'll just handle permissions and making sure it stays in the portupgrade path |
| 18:14.27 | Notify | 03BRL-CAD:starseeker * 59393 (brlcad/trunk/src/libanalyze/CMakeLists.txt brlcad/trunk/src/libanalyze/tests/CMakeLists.txt): Remove the diff code from libanalyze |
| 18:15.52 | Notify | 03BRL-CAD:starseeker * 59394 (brlcad/trunk/include/db_diff.h brlcad/trunk/src/librt/db_diff.c brlcad/trunk/src/librt/tests/CMakeLists.txt): Implement db_compare. Rework the diff test to use the new approach. |
| 18:16.11 | starseeker | brlcad: that appears to be a successful proof-of-concept for your diff approach |
| 18:18.07 | Notify | 03BRL-CAD:starseeker * 59395 brlcad/trunk/src/librt/db_diff.c: Hide a couple more functions |
| 18:20.31 | brlcad | starseeker: cool, any room for improvement now that you've used it? something else worth generalizing or good enough? |
| 18:21.27 | *** join/#brlcad ``Erik (~erik@pool-74-103-94-19.bltmmd.fios.verizon.net) | |
| 18:23.13 | Notify | 03BRL-CAD:starseeker * 59396 brlcad/trunk/src/librt/tests/test_diff.c: Sort out attributes from parameters |
| 18:23.52 | starseeker | brlcad: probably good enough - need to work out a few more kinks and maybe try replicating gdiff |
| 18:26.39 | starseeker | one question is whether the avs diffing function should move to libbu |
| 18:26.40 | Notify | 03BRL-CAD:starseeker * 59397 brlcad/trunk/src/librt/db_diff.c: Fix bu_avs_diff logic |
| 18:30.42 | Notify | 03BRL-CAD:starseeker * 59398 brlcad/trunk/src/librt/db_diff.c: Use the correct directory pointer when adding objects... |
| 18:33.21 | Notify | 03BRL-CAD:starseeker * 59399 brlcad/trunk/src/librt/db_diff.c: Comment typo |
| 18:42.01 | Notify | 03BRL-CAD:starseeker * 59400 brlcad/trunk/src/librt/tests/test_diff.c: Parameters, then attributes |
| 18:51.13 | Notify | 03BRL-CAD:brlcad * 59401 brlcad/trunk/include/db_diff.h: doxy comments get associated with the symbol that they immediately precede, so db_compare()'s comment needs to come after the typedef. added another comment just for the typedef so it's not dangling undocumented API. |
| 18:57.38 | *** join/#brlcad Ch3ck_ (~Ch3ck@195.24.220.16) | |
| 19:15.07 | *** join/#brlcad Ch3ck_ (~Ch3ck@195.24.220.16) | |
| 19:18.28 | *** join/#brlcad FreezingCold (~FreezingC@205.211.50.163) | |
| 19:25.22 | *** join/#brlcad Ch3ck_ (~Ch3ck@195.24.220.16) | |
| 19:58.21 | *** join/#brlcad Ch3ck_ (~Ch3ck@195.24.220.16) | |
| 20:02.07 | Notify | 03BRL-CAD:carlmoore * 59402 brlcad/trunk/src/util/imgdims.c: remove single-quote marks from Usage |
| 20:23.37 | *** join/#brlcad deepak (~chatzilla@117.214.201.55) | |
| 20:34.58 | Notify | 03BRL-CAD:starseeker * 59403 brlcad/trunk/src/librt/db_diff.c: Don't try to call a function we don't have |
| 20:36.43 | Notify | 03BRL-CAD:starseeker * 59404 brlcad/trunk/include/db_diff.h: comment tweak |
| 20:44.56 | Notify | 03BRL-CAD:brlcad * 59405 brlcad/trunk/src/librt/bezier_2d_isect.c: apply sf patch 253 (submitted to bug tracker as #355) from huskmate13 which fixes a bug where extrude objects of sketches containing bezier curves where the first and last points contain the same was resulting in a division by zero (crash). nice catch. |
| 20:47.23 | Notify | 03BRL-CAD:brlcad * 59406 brlcad/trunk/src/librt/bezier_2d_isect.c: ws |
| 20:50.06 | Notify | 03BRL-CAD:brlcad * 59407 brlcad/trunk/src/librt/bezier_2d_isect.c: mark a few functions HIDDEN that don't need to be exposed |
| 20:52.37 | Notify | 03BRL-CAD:brlcad * 59408 brlcad/trunk/src/librt/bezier_2d_isect.c: extrude directly calls Bezier |
| 20:53.37 | Notify | 03BRL-CAD:brlcad * 59409 brlcad/trunk/include/nurb.h: remove the bezier_2d_isect functions that didn't need to be public. |
| 21:10.52 | *** join/#brlcad Ch3ck_ (~Ch3ck@195.24.220.16) | |
| 21:15.38 | Notify | 03BRL-CAD:brlcad * 59410 (brlcad/trunk/include/nurb.h brlcad/trunk/src/librt/bezier_2d_isect.c and 3 others): rename several of the bezier functions including subdivide_bezier, FindRoots, and Bezier() to be more consistent and since they don't need to be public API (and they were never publicly documented), remove them from the public header to reduce their scope. |
| 21:16.38 | Notify | 03BRL-CAD:brlcad * 59411 brlcad/trunk/src/librt/CMakeLists.txt: rename bezier_2d_isec.c to just bezier.c |
| 21:25.30 | Notify | 03BRL-CAD:brlcad * 59412 brlcad/trunk/src/librt/bezier.c: update the file header and fix the functions that don't conform with our naming convention |
| 21:27.15 | Notify | 03BRL-CAD:brlcad * 59413 (brlcad/trunk/src/librt/bezier.c brlcad/trunk/src/librt/librt_private.h): move the function documentation over with the declarations |
| 21:37.45 | Notify | 03BRL-CAD:brlcad * 59414 (brlcad/trunk/AUTHORS brlcad/trunk/NEWS): credit huskmate13 with his bug fix (sf patch 253, bug 354) that fixes a divide-by-zero crash when you tried to evaluate (raytrace) extrude objects referencing sketch objects containing bezier curves where the first and last point of the curve have the same Y (e.g., 0,0,0; 50,100,0; 100,0,0) |
| 21:42.58 | *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14) | |
| 21:48.23 | brlcad | starseeker: that case logic collapses in r59403 easily with if (add_func && add_func(...)) |
| 21:54.08 | starseeker | brlcad: ah, sounds good |
| 21:54.37 | starseeker | brlcad: just saw the comment about allowing NULL to skip a step, did the first thing that came to mind |
| 21:56.39 | starseeker | oh, wait - I wanted to decrement error if something went wrong when the function call was run, but not when NULL was passed in |
| 21:56.52 | starseeker | perhaps there was a method to my madness |
| 21:57.42 | brlcad | if (func) if (func()) is equivalent to if (func && func()) |
| 21:57.42 | Notify | 03BRL-CAD:carlmoore * 59415 brlcad/trunk/src/sig/imod.c: 'cosmetic' changes to look like the smod.c program, and remove incorrect references to bwmod |
| 21:59.00 | starseeker | ok, I think i see it |
| 22:00.27 | Notify | 03BRL-CAD:starseeker * 59416 brlcad/trunk/src/librt/db_diff.c: collapse if conditionals per Sean's suggestion |
| 22:03.31 | Notify | 03BRL-CAD:starseeker * 59417 (brlcad/trunk/include/ged.h brlcad/trunk/src/libged/CMakeLists.txt and 3 others): No function as yet, but stub in a gdiff command for MGED and archer |
| 22:13.08 | Notify | 03BRL-CAD:carlmoore * 59418 (brlcad/trunk/src/sig/imod.c brlcad/trunk/src/sig/smod.c): make imod and smod programs more alike; remove '?' comment because 'default:' picks up other stuff; add an 'else' to imod |
| 23:11.06 | *** join/#brlcad hickoryk1oll (~hickorykn@66-118-151-70.static.sagonet.net) | |
| 23:16.00 | *** join/#brlcad Ch3ck_ (~Ch3ck@66-118-151-70.static.sagonet.net) | |