| 00:33.16 | starseeker | Huh: http://www.rhino3d.com/download/opennurbs/5.0/opennurbs5samples |
| 00:38.15 | starseeker | Hah! the ON_Brep structure diagram is still on the web: http://wiki.mcneel.com/developer/brepstructure |
| 00:50.46 | *** join/#brlcad FreezingDroid (~FreezingC@135.0.41.14) |
| 01:03.40 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 01:10.21 | *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14) |
| 01:21.44 | starseeker | brlcad: do you think fortran to C/C++ conversion might be something that could come under the GCI heading, or is that a but much for such tasks? |
| 01:22.11 | starseeker | (thinking of http://www.pdas.com/naca456.html) |
| 03:05.02 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 03:29.42 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 03:46.00 | brlcad | starseeker: how many lines of code is it? |
| 03:51.04 | brlcad | gut feeling is that a GCI task would be somewhere under 5k LOC, so it'd need to be less or easily segmented into testable sets around 1-3k per task |
| 03:52.23 | brlcad | and probably on the lower side of that range |
| 04:50.55 | starseeker | cat *.f90|wc -l says 4926 lines |
| 04:53.10 | starseeker | https://github.com/starseeker/naca is the repo I set up, but haven't had any time to do the translation |
| 04:54.04 | starseeker | grants it's a bit off-topic - possible relatedness to BRL-CAD would be in the form of a shapes program to generate wings using it |
| 05:09.58 | starseeker | I've stubbed in some (basically) empty cxx files and CMake logic, so it should be a question of translating a routine and hitting make to see if it builds |
| 05:10.35 | starseeker | eventually of course it'll need to be checked against the outputs from the fortran code |
| 05:10.52 | starseeker | also good tasks if a complete translation is achieved |
| 05:13.27 | starseeker | actually, that loc is too large - the avd file isn't necessary for the core naca456 program |
| 05:14.36 | starseeker | so it's 3018 lines total |
| 05:15.14 | starseeker | of the four, only one file is longer than 1k and it's at 1350 - so maybe one file per task... |
| 08:18.22 | *** join/#brlcad ries (~ries@D979EA84.cm-3-2d.dynamic.ziggo.nl) |
| 09:47.21 | *** join/#brlcad LordOfBikes (~armin@dslb-092-074-235-144.092.074.pools.vodafone-ip.de) |
| 11:22.41 | *** join/#brlcad ishwerdas (75d6ca51@gateway/web/cgi-irc/kiwiirc.com/ip.117.214.202.81) |
| 12:09.14 | *** join/#brlcad ishwerdas (75d6ca51@gateway/web/cgi-irc/kiwiirc.com/ip.117.214.202.81) |
| 12:50.12 | ``Erik | starseeker: f2c generated code has horrible function call style (copying arrays instead of passing pointers) and typically requires their lib... libwa started life as fortran, you can check the git repo to see the major steps to convert the f2c output to normalish C O.o (original program is something weird like docs/sphere.f) |
| 13:56.03 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 14:02.19 | *** join/#brlcad teepee (~teepee@gateway/tor-sasl/teepee) |
| 15:17.36 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 16:26.35 | starseeker | ``Erik: that's why I'm suggesting something along the lines of what Sean had mentioned in earlier discussions - a manual translation of the logic, rather than an f2c conversion |
| 17:00.36 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 18:19.22 | *** join/#brlcad Ch3ck_ (~Ch3ck@66-118-151-70.static.sagonet.net) |
| 18:19.24 | *** join/#brlcad tofu_ (~sean@66-118-151-70.static.sagonet.net) |
| 18:35.02 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 22:41.14 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 23:45.20 | Notify | 03BRL-CAD:stefanmirea * 63548 brlcad/trunk/src/librt/primitives/revolve/revolve.c: Fixed rt_revolve_plot() to work with negative angles too (inspired from bug 356 on the tracker). The other rt_revolve_*() functions also need to be modified to support negative angles. |
| 23:51.33 | *** join/#brlcad stefanmirea (~stefan@p16.eregie.pub.ro) |