| 00:00.01 | mpictor | abort trap 6 does not sound familiar |
| 00:01.15 | mpictor | in step-g or g-step, have you had a need for const or non-const attribute setters/getters that weren't provided in the generated code? |
| 00:01.47 | mpictor | that's one of the things I'm adding in the inverse attr branch |
| 00:02.12 | mpictor | err, getters only |
| 00:02.23 | mpictor | const setters don't make much sense :) |
| 00:10.47 | starseeker | um. probably, but I'm not sure I'd realize it - I was mostly focused on finding *some* way to get the outputs I wanted |
| 00:10.57 | mpictor | heh |
| 00:11.11 | starseeker | from what I recall, accessing the contents of complex entities is a pain... |
| 00:11.19 | starseeker | did the backtrace help any? |
| 00:11.57 | mpictor | I needed it for some change I was making, so I'm making sure that there are const and non-const getters for everything |
| 00:12.03 | starseeker | cool |
| 00:12.15 | mpictor | oh! I know - I'm creating objects if the pointer is null |
| 00:12.37 | mpictor | I haven't looked at the backtrace yet |
| 00:12.51 | starseeker | np - let me know if you need me to try something else |
| 00:12.56 | mpictor | ok |
| 00:13.09 | mpictor | complex entities are on my todo list |
| 00:16.13 | starseeker | i think my favorite weirdness, aside from that "undertype" for units you helped figure out, was the name setting on the CARTESIAN_TRANSFORMATION_OPERATOR_3D |
| 00:16.45 | mpictor | that undertype is another thing I keep meaning to look at |
| 00:16.54 | mpictor | name setting? |
| 00:19.28 | starseeker | http://sourceforge.net/p/brlcad/code/HEAD/tree/brlcad/trunk/src/conv/step/g-step/Assembly_Product.cpp |
| 00:20.08 | starseeker | line 188 |
| 00:20.51 | mpictor | oh yea |
| 00:20.54 | mpictor | that is weird |
| 00:20.56 | starseeker | it works OK, but yeish |
| 00:21.01 | mpictor | heh |
| 00:21.34 | starseeker | guess I could have done http://sourceforge.net/p/brlcad/code/HEAD/tree/brlcad/trunk/src/conv/step/g-step/Assembly_Product.cpp#l188 |
| 00:23.11 | mpictor | dunno why sourceforge insists on making the page so narrow |
| 00:23.29 | mpictor | at least they have a button to change it |
| 01:53.37 | mpictor | I'm surprised this hasn't bitten us before: https://github.com/stepcode/stepcode/blob/mp/config-parse-rebase/src/exp2cxx/class_strings.c#L37 |
| 01:54.10 | mpictor | strcpy at line 56 copies retval into retval |
| 01:56.18 | mpictor | that's where your crash is |
| 03:44.02 | starseeker | erm. I take it that shouldn't be static? |
| 03:44.58 | starseeker | probably should be malloced and caller freed, in fact... |
| 03:47.03 | starseeker | 'cept we have a bunch of hardwired strings too... |
| 03:47.05 | starseeker | hmm |
| 03:51.51 | starseeker | maybe return a C++ string? |
| 09:02.02 | *** join/#stepcode kesha (~kesha@14.139.122.114) |
| 13:44.33 | *** join/#stepcode kesha (~kesha@14.139.122.114) |
| 14:27.23 | mpictor | starseeker: it's C, so std:string is out unless we rewrite some stuff |
| 14:28.18 | mpictor | I was thinking of eliminating the strcpy |
| 14:32.26 | mpictor | or, create the buffer in the caller and pass a pointer |
| 15:20.59 | starseeker | mpictor: sounds like a plan |
| 17:13.11 | *** join/#stepcode kesha (~kesha@14.139.122.114) |
| 17:37.23 | *** join/#stepcode hoiji (~hoiji@59.89.17.151) |
| 18:30.05 | *** join/#stepcode kesha (~kesha@14.139.122.114) |
| 18:33.20 | *** join/#stepcode kesha (~kesha@14.139.122.114) |
| 21:36.32 | *** join/#stepcode kesha_ (~kesha@14.139.122.114) |