| 00:21.38 | Notify | 03BRL-CAD:starseeker * 60032 (brlcad/trunk/src/libbu/bitv.c brlcad/trunk/src/libbu/bu_internals.h and 72 others): Thin out the bu header inclusions in libbu itself. |
| 00:32.29 | starseeker | meh - for a single threaded build of just libbu - 7.24.0: 8.419s, r60032: 7.578s |
| 00:34.56 | starseeker | keeps chipping away |
| 00:49.58 | Notify | 03BRL-CAD:starseeker * 60033 (brlcad/trunk/include/CMakeLists.txt brlcad/trunk/include/bu/CMakeLists.txt and 2 others): move vfont-if.h to bu/ |
| 00:51.55 | Notify | 03BRL-CAD:starseeker * 60034 brlcad/trunk/include/bu/vfont-if.h: Don't include all of bu.h in vfont-if.h |
| 00:53.42 | Notify | 03BRL-CAD:starseeker * 60035 (brlcad/trunk/include/bu/vfont-if.h brlcad/trunk/src/libbu/vfont.c): Actually, only need the defines.h in the header itself. |
| 00:55.23 | Notify | 03BRL-CAD:starseeker * 60036 brlcad/trunk/include/vmath.h: Only include the bu sub-header defining fastf_t for vmath.h |
| 00:57.51 | Notify | 03BRL-CAD:starseeker * 60037 (brlcad/trunk/src/adrt/adrt_struct.h brlcad/trunk/src/conv/shp/shapelib/shpopen.c): Correct code that was using vmath.h for more of libbu than just fastf_t |
| 01:06.17 | Notify | 03BRL-CAD:starseeker * 60038 (brlcad/trunk/include/bn.h brlcad/trunk/src/libbn/chull.c and 2 others): Reduce bn.h inclusions from bu to those actually needed in the header, and add extra local includes where needed in bn. |
| 01:22.54 | Notify | 03BRL-CAD:starseeker * 60039 (brlcad/trunk/src/libbn/anim.c brlcad/trunk/src/libbn/bntester.c and 17 others): Break up libbn's use of bu.h |
| 01:27.26 | Notify | 03BRL-CAD:starseeker * 60040 brlcad/trunk/include/brep.h: Don't need bu.h in brep.h |
| 01:53.58 | *** join/#brlcad kintel_ (~kintel@unaffiliated/kintel) |
| 02:21.08 | Notify | 03BRL-CAD:starseeker * 60041 (brlcad/trunk/include/db.h brlcad/trunk/include/dm.h and 20 others): More 'use only what's needed' for bu.h components |
| 02:38.13 | brlcad | starseeker: that's 10% for just libbu, which everything includes so even if all you did was extend that to the rest of the package, that'd be 10% across the board |
| 02:39.14 | brlcad | say you get another 10% from libbn, extend to callers .. now it approaches 10-20% across the board (only because not everything uses bn) |
| 02:39.49 | starseeker | nods |
| 02:39.54 | brlcad | and again and again, it has the potential to cut the build time in half |
| 02:40.22 | starseeker | raytrace.h is going to be hard to break apart - it doesn't seem to be catagorized like libbu/bn |
| 02:40.32 | brlcad | nods |
| 02:40.40 | starseeker | unless I do something like one file per struct, which seems a tad excessive |
| 02:42.00 | brlcad | there's at least a half-dozen logical groupings |
| 02:42.45 | starseeker | was hoping brlcad would group it - it'll save time as compared to me doing it wrong and you fixing it ;-) |
| 02:42.52 | brlcad | heh |
| 02:43.18 | starseeker | bu.h got included *everywhere* |
| 02:43.23 | starseeker | it's kind of amazing |
| 02:43.57 | brlcad | by design, great reuse |
| 02:44.59 | brlcad | I can take a stab at groupings, but that's probably one of the last libs if they're attacked in order |
| 02:45.11 | starseeker | nods - I know it's not a priority |
| 02:45.23 | starseeker | and I'm understanding the merits of going in order better now |
| 02:45.59 | starseeker | bu.h is going to be a few weeks of off and on pecking to get all of the uses handled |
| 02:46.28 | brlcad | yeah |
| 02:47.57 | Notify | 03BRL-CAD:starseeker * 60042 brlcad/trunk/include/light.h: Include only the needed bits for light.h |
| 02:49.03 | starseeker | it's weird - when I switch mater.h from bu.h to bu/vls.h, I suddenly get undefined link errors when anything tries to link libged |
| 02:51.23 | starseeker | digs |
| 03:15.33 | *** join/#brlcad KimK (~Kim__@ip24-255-223-153.ks.ks.cox.net) |
| 03:19.18 | brlcad | means some file in src/libged was including only mater.h but actually needed bu.h |
| 03:19.34 | brlcad | some preprocessor symbol |
| 03:22.12 | brlcad | quick survey shows it's one or more of these: for file in `grep mater.h * | cut -f 1 -d:` ; do if ! grep -q bu.h $file ; then echo $file ; fi ; done |
| 03:22.21 | brlcad | (in src/libged dir) |
| 03:22.53 | brlcad | color.c ged_private.h polybinout.c prcolor.c vdraw.c wdb_vdraw.c |
| 03:24.17 | brlcad | they all use bu.h, so they all need it anyways |
| 03:27.11 | brlcad | starseeker: note that headers you move out of include to include/bu need to be marked in CHANGES (minimally impacting, but the regex needed is a little ugly) |
| 04:06.33 | *** join/#brlcad hsrai (~hsrai@202.164.53.116) |
| 04:30.22 | Notify | 03BRL-CAD Wiki:Vegantrex6 * 0 /wiki/User:Vegantrex6: |
| 04:31.59 | *** join/#brlcad hsrai (~hsrai@202.164.53.116) |
| 04:55.24 | *** join/#brlcad vegantrex6 (925fe7a4@gateway/web/freenode/ip.146.95.231.164) |
| 05:20.26 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 05:42.38 | *** join/#brlcad vegantrex6 (925f13cb@gateway/web/freenode/ip.146.95.19.203) |
| 05:42.59 | vegantrex6 | Hullo, Im Gabriela Voll |
| 05:43.54 | vegantrex6 | I realize it really late but im just introducing myself as I become more familiar with brl-cad in preparation for GSoc proposal |
| 05:44.43 | vegantrex6 | interested in the STEP exporter and STEP importer improvements at the moment, i'll be lurkin |
| 06:46.57 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 08:04.31 | *** join/#brlcad KimK (~Kim__@ip24-255-223-153.ks.ks.cox.net) |
| 08:44.15 | *** join/#brlcad merzo (~merzo@user-94-45-58-138-1.skif.com.ua) |
| 09:54.57 | *** join/#brlcad KimK (~Kim__@ip24-255-223-153.ks.ks.cox.net) |
| 10:34.07 | *** join/#brlcad KimK (~Kim__@ip24-255-223-153.ks.ks.cox.net) |
| 11:04.31 | *** join/#brlcad KimK (~Kim__@ip24-255-223-153.ks.ks.cox.net) |
| 11:34.00 | *** join/#brlcad KimK (~Kim__@ip24-255-223-153.ks.ks.cox.net) |
| 12:25.44 | *** join/#brlcad funkey_monkey (~funkey_mo@202.78.172.162) |
| 12:25.45 | starseeker | brlcad: gah. Can I just stub in headers that include their bu version and avoid the issue for now? |
| 12:26.22 | starseeker | doubts he could get a regex like that right |
| 12:30.37 | *** join/#brlcad teepee_ (bc5c2134@gateway/web/freenode/ip.188.92.33.52) |
| 12:31.12 | *** join/#brlcad KimK (~Kim__@ip24-255-223-153.ks.ks.cox.net) |
| 12:37.44 | *** join/#brlcad ries (~ries@190.9.171.121) |
| 13:24.42 | *** join/#brlcad funkey_monkey (~funkey_mo@202.78.172.162) |
| 13:29.54 | ``Erik | if I grok, bu provided a lot of posix style standardization before there was posix, so yeh, all over the place... |
| 13:36.26 | ``Erik | hm, walmart is demoing a truck that uses a microturbine driving a generator and electric motors O.o |
| 13:45.48 | *** join/#brlcad KimK (~Kim__@ip24-255-223-153.ks.ks.cox.net) |
| 14:09.22 | brlcad | starseeker: regex is something like 's/\(#[[:space:]]*include[[:space:]]+["<]\)FILE\.h/\1bu\/FILE.h/g' |
| 14:14.46 | *** join/#brlcad KimK (~Kim__@ip24-255-223-153.ks.ks.cox.net) |
| 14:15.18 | Notify | 03BRL-CAD:carlmoore * 60043 brlcad/trunk/src/libbrep/intersect.cpp: remove trailing blank or tab |
| 14:17.25 | *** join/#brlcad Anaphaxeton (~george@unaffiliated/anaphaxeton) |
| 14:36.58 | *** join/#brlcad deepak (~chatzilla@59.91.117.181) |
| 14:41.51 | *** join/#brlcad merzo (~merzo@user-94-45-58-138-1.skif.com.ua) |
| 15:00.34 | *** join/#brlcad KimK (~Kim__@ip24-255-223-153.ks.ks.cox.net) |
| 15:10.48 | *** join/#brlcad kesha (~kesha@14.139.122.114) |
| 15:17.24 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 15:24.03 | Notify | 03BRL-CAD:starseeker * 60044 brlcad/trunk/CHANGES: Building on Sean's suggestion, add regex lines for headers that have moved to bu subdirectory. |
| 15:28.20 | Notify | 03BRL-CAD:starseeker * 60045 brlcad/trunk/CHANGES: Want to eventually rename all the hist stuff to histogram - queue it up in CHANGES. |
| 15:28.41 | *** join/#brlcad kintel_ (~kintel@unaffiliated/kintel) |
| 15:29.37 | starseeker | brlcad: that regex wasn't too bad - when *you* say a regex is "a little ugly" I'm expecting something like the "remove all the spaced out comments" fun |
| 15:30.07 | *** part/#brlcad andrei__ (~IceChat77@188.25.171.245) |
| 15:38.39 | kesha | brlcad: (y). How abt sending out a mail on mailing list giving initial pointers and asking if anyone would like to take it up and make it ? :) |
| 15:50.49 | *** join/#brlcad andrei__ (~IceChat77@188.25.171.245) |
| 15:56.15 | *** join/#brlcad javampire (~ncsaba@p4FF74765.dip0.t-ipconnect.de) |
| 15:56.53 | Notify | 03BRL-CAD:starseeker * 60046 brlcad/trunk/src/libged/search.c: Tweak search usage |
| 16:19.50 | Notify | 03BRL-CAD:starseeker * 60047 (brlcad/trunk/include/rt/search.h brlcad/trunk/src/libged/search.c brlcad/trunk/src/librt/search.c): the Q (quiet) flag wasn't working properly for search - fix. |
| 16:29.35 | *** join/#brlcad harman_ (~harman@198.199.108.236) |
| 16:30.37 | *** part/#brlcad harman_ (~harman@198.199.108.236) |
| 16:36.16 | brlcad | starseeker: not too bad means it can be figured out and written in just a few minutes ... no correlation to length or fugliness |
| 16:36.49 | brlcad | even the spaced out comment one, while long, didn't take long at all to write |
| 16:37.12 | brlcad | there are some regexes that really are *hard* to compose, difficult concepts |
| 16:38.01 | brlcad | note that you only need to escape \. on the match expression, not the replacement text |
| 16:38.22 | brlcad | doesn't hurt, but some regex engines may insert a literal \. |
| 16:38.31 | brlcad | in which case it might hurt ;) |
| 16:42.22 | Notify | 03BRL-CAD:brlcad * 60048 brlcad/trunk/CHANGES: you don't want to escape dot in the replacement text, '.' only has special meaning in the match |
| 16:47.02 | *** join/#brlcad javampire (~ncsaba@p4FF74765.dip0.t-ipconnect.de) |
| 16:58.09 | *** join/#brlcad javampire (~ncsaba@p4FF74765.dip0.t-ipconnect.de) |
| 17:10.33 | *** join/#brlcad deepak (~chatzilla@59.91.117.181) |
| 17:11.59 | deepak | brlcad: Hi, I have just submitted the patch, here is the link https://sourceforge.net/p/brlcad/patches/255/ please test it. In this patch I have reduced the line of code. |
| 17:32.04 | brlcad | deepak: awesome, thanks |
| 17:34.22 | deepak | brlcad: Thanks :) |
| 17:35.23 | brlcad | deepak: took a quick look and a few issues |
| 17:36.58 | *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14) |
| 17:37.00 | deepak | brlcad: Please tell me the issues. |
| 17:38.06 | brlcad | deepak: it's a reasonable refactoring, decent 3:1 ratio |
| 17:39.24 | brlcad | some of the issues I see: 1) icv_val() name shouldn't have icv_ prefix and should be marked HIDDEN (static) |
| 17:40.09 | brlcad | 2) the places it's called should,not,have,args,squished per our HACKING guidelines (we take great priority in consistent style) |
| 17:41.08 | brlcad | 3) should be "switch (switch_icv_val) {" ... switch is not a function |
| 17:41.35 | brlcad | 4) seemingly unrelated code removals in src/mged/mged.c |
| 17:42.23 | brlcad | that's all I see right now |
| 17:49.22 | *** join/#brlcad rotad (~user@unaffiliated/rotad) |
| 17:50.43 | deepak | brlcad: Okay i'll make changes and transform it according to HACKING guidelines. I didn't get you, you said "should be "switch (switch_icv_val) {" ... switch is not a function" can you explain it again. |
| 17:50.45 | deepak | <PROTECTED> |
| 17:58.56 | *** join/#brlcad krushia (~krushia@saturn.pcsrvc.com) |
| 18:14.04 | deepak | gotch'a what does that mean :). |
| 18:21.52 | brlcad | deepak: HACKING talks about switch too |
| 18:22.22 | brlcad | I know it's a big doc, but coding is complicated ;) it's our best guide for working together |
| 18:23.34 | deepak | brlcad: Exploring HACKING guidelines, I must say it's a best thing I have ever done in my life :) |
| 18:29.04 | brlcad | what's the best thing? that patch? |
| 18:29.20 | brlcad | working on open source? |
| 18:29.35 | brlcad | reading a dev guide? |
| 18:31.43 | *** join/#brlcad trailhand (46b3bd91@gateway/web/freenode/ip.70.179.189.145) |
| 18:32.40 | trailhand | Howdy |
| 18:37.02 | trailhand | I had a question and a comment. |
| 18:39.02 | trailhand | Question: I have a BRL-CAD model with a TGC I am trying to convert to STL with g-stl. How do you control facet resolution with g-stl? The cone has 10 sides, I need more. |
| 18:40.03 | trailhand | Comment: I found an interesting way to add 3D (and possibly 2D) text to a BRL-CAD model with OpenInventor's font system. |
| 18:42.17 | deepak | brlcad: Everything :D. Working on open source, exploring/leaning new stuff in C language (that help me to make my concept more clearer and my love toward C is growing day by day). I enjoyed alot when I made that patch, learning what is happening in that file, then finding similarity in code, applying concept if this not then what, keep on questioning what are the pros and cons. I just... |
| 18:42.18 | deepak | ...incredible :). |
| 18:47.36 | *** join/#brlcad witness___ (uid10044@gateway/web/irccloud.com/x-juuaoglrevsqtjlr) |
| 19:12.23 | brlcad | trailhand: howdy! |
| 19:27.29 | brlcad | trailhand: g-stl has a number of tolerance options |
| 19:27.38 | brlcad | those control tessellation quality |
| 19:28.09 | brlcad | e.g., -a 1.0 will set the absolute tolerance to 1mm ... probably a find mesh |
| 19:28.26 | brlcad | or -n 1.0 to set a 1-degree normal tolerance |
| 19:36.27 | brlcad | deepak: glad you're having fun :) |
| 19:42.05 | deepak | brlcad: Thanks, I feel myself luck that i'm part of this community. I have one question, i'm searching for a file which create sphere when type "make sph sph" on mged. I mean when I type command "make sph sph" on mged, then what file called? Is there any kind of documentation or procedure which can help us. |
| 19:44.38 | trailhand | @brlcad: Thanks, I'll look more at the tolerance options . |
| 19:45.38 | trailhand | @brlcad: Would the BRL-CAD team be interested in us developing a 3D/2D primitive? |
| 19:48.08 | trailhand | @brlcad: and one more question: Could there be a Google Summer of Code project to start moving ray-tracing to the GPU? |
| 19:49.10 | trailhand | @brlcad: Could one start with a small enough project? I know some students that would be interested in this. |
| 19:49.44 | n_reed | deepak: the file containing the source code that implements the make command (ged_make()) is src/libged/make.c |
| 19:56.48 | trailhand | @brlcad: Correction, that is a 3D/2D "text" primitive. |
| 19:57.45 | deepak | n_reed: Is that file also responsible for making sphere. I mean when we write "make sph1 sph", make.c is for make command but what about sph1 and sph? Where can I find code of sphere? Where can I find that sph stored i.e sph1? |
| 20:03.10 | Notify | 03BRL-CAD Wiki:Carlgmoore * 0 /wiki/User:Carlgmoore: |
| 20:08.41 | trailhand | @deepak: The code for adding a sphere to the internal database is in ./src/libwdb/wdb.c, I think |
| 20:09.30 | trailhand | @deepak: and a sphere is added as an ellipsoid, IIRC, so look for the mk_ell function in that file. The raytracing code |
| 20:10.45 | trailhand | @deepak: in librt has validation checking for different primitives, so look there too for the ellipsoid (a sphere is a specialized ellipsoid) |
| 20:13.14 | deepak | trailhand: Okay i'll start exploring in these files. Thanks. |
| 20:22.54 | trailhand | @deepak: I looked at wdb.c and there is a mk_sph (for the sphere) but if you notice, it is added as an ID_ELL (ellipsoid) |
| 20:24.40 | *** join/#brlcad merzo (~merzo@208-4-132-95.pool.ukrtel.net) |
| 20:25.52 | *** join/#brlcad yizhen_ (~yizhen@c-98-206-167-91.hsd1.il.comcast.net) |
| 20:26.49 | *** join/#brlcad kesha (~kesha@14.139.122.114) |
| 20:32.34 | *** part/#brlcad trailhand (46b3bd91@gateway/web/freenode/ip.70.179.189.145) |
| 20:58.02 | Notify | 03BRL-CAD:starseeker * 60049 brlcad/trunk/src/libdm/dm-osg.cpp: Try to ditch the trackball manipulator and directly use the provided matrices. Not working yet. |
| 21:00.00 | brlcad | ah, he left |
| 21:04.19 | Notify | 03BRL-CAD:n_reed * 60050 brlcad/trunk/src/libbrep/intersect.cpp: move subsurface intersection test to separate function |
| 21:09.40 | Notify | 03BRL-CAD:n_reed * 60051 brlcad/trunk/src/libbrep/intersect.cpp: mark Overlapevent members const |
| 21:11.16 | Notify | 03BRL-CAD:n_reed * 60052 brlcad/trunk/src/libbrep/intersect.cpp: make unmodified parameter const |
| 21:16.07 | Notify | 03BRL-CAD Wiki:Sean * 6530 /wiki/Summer_of_Code/Application_Guidelines: make more org agnostic |
| 21:17.18 | ``Erik | http://www.arcimoto.com/product (starseeker may get a kick out of this one) |
| 21:47.33 | brlcad | their enclosed versions are more interesting for quasipractical use |
| 21:55.16 | Notify | 03BRL-CAD:starseeker * 60053 (brlcad/trunk/src/libdm/dm-ogl.c brlcad/trunk/src/libdm/dm-osg.cpp): Start digging in to figure out how to mimic ogl's matrix settings in views/cameras. |
| 22:10.36 | *** join/#brlcad ejn0 (~ejno@unaffiliated/kazaik) |
| 22:29.26 | Notify | 03BRL-CAD Wiki:Sean * 6531 /wiki/Summer_of_Code/Application_Guidelines: tighten up a little bit |
| 22:35.24 | Notify | 03BRL-CAD Wiki:Sean * 6532 /wiki/Summer_of_Code/Application_Guidelines: reorder for clarity |
| 22:35.49 | brlcad | one more down, one to go |
| 22:56.05 | ``Erik | the most recent generation is all enclosed, I think? seemed like a nifty and plausible successor to the aptera |
| 22:57.02 | ``Erik | speaking of; http://www.gizmag.com/aptera-independent-production-us/27868/ |
| 23:11.26 | brlcad | I wouldn't mind driving that kind of "motorcycle" on a daily basis |
| 23:20.10 | *** join/#brlcad zxq9 (~ceverett@FL9-125-199-207-150.okn.mesh.ad.jp) |
| 23:24.47 | Notify | 03BRL-CAD Wiki:198.207.222.136 * 6533 /wiki/Google_Summer_of_Code/Project_Ideas: /* OpenSCAD Projects */ removing what i had added earlier, as it does not cross-pollinate with other cad programs (per GSOC intro) |
| 23:43.22 | *** join/#brlcad kintel_ (~kintel@unaffiliated/kintel) |