| 15:21.31 | starseeker | mpictor: would you mind terribly if I did some whitespace and case tweaking on the CMakeLists.txt files? I know that would complicate some branch merges that are in the future, so I won't do it without your OK, but if it's OK with you i'd like to have a go at it |
| 15:43.03 | mpictor | starseeker: have at it |
| 15:44.28 | mpictor | are the large indents bugging you? for example, https://github.com/stepcode/stepcode/blob/mp/config-parse/cmake/SC_CXX_schema_macros.cmake#L51 |
| 15:45.37 | mpictor | if you're going to work on the code on github, I'd start with that branch, mp/config-parse, because it has a lot of cmake changes |
| 15:46.52 | mpictor | and if you have ideas for better names for any of cmake/SC_*, let me know - I was running low on creativity by the time I got done ;) |
| 18:31.40 | starseeker | mpictor: is the plan to offer both single-file and multi-file outputs for exp2cxx, or (once the headers are broken out as well) to just go with the multi-file? |
| 18:37.24 | starseeker | is pondering allowing for both single and multi-file, but if the plan is to hammer the multi-file fully into submission that might end up making the single file outputs completely obsolete |
| 18:49.20 | mpictor | davyw was wondering the same thing, but I wonder if it's too much of a pain |
| 18:51.17 | mpictor | another option would be to write a .cc file that #includes all the entities and types |
| 18:58.10 | starseeker | the latter probably makes the most sense |
| 18:59.11 | starseeker | that gives people the "include/build everything" option when doing initial development, and then they can refine things over time |
| 18:59.53 | starseeker | easy to autogenerate too |
| 18:59.59 | mpictor | yea |
| 19:00.36 | starseeker | when I fork on github, can I fork one branch or do I fork the whole thing? |
| 19:00.59 | mpictor | apparently, it will make it build quite a bit faster - it's mentioned on SO as a solution to long build times |
| 19:01.05 | mpictor | you have to for the whole thing |
| 19:01.11 | mpictor | you should have commit access, though |
| 19:01.29 | mpictor | just create your own branch in the main repo if you like |
| 19:01.34 | starseeker | ah point |
| 19:02.09 | starseeker | is it the header includes now that are dictating the build time in the new branch? |
| 19:02.44 | mpictor | I haven't really looked |
| 19:03.02 | mpictor | davy told me that he worked on the includes and didn't see a difference |
| 19:03.09 | starseeker | nods - probably a little tricky to profile for than in any case... |
| 19:03.39 | mpictor | if he pushed his changes to github, I didn't see them |
| 19:04.00 | mpictor | maybe he's waiting until he has something which shows an improvement |
| 19:04.15 | mpictor | I'm working on the helper functions |
| 19:04.25 | starseeker | breaking out the include files would involve both defining the equalivent of the .cc file for each type plus determing what definitions that .cc file needs included? |
| 19:04.47 | mpictor | once that's done, I'll try one more build of everything and then I think we're good to go |
| 19:04.55 | mpictor | headers are already written |
| 19:05.06 | starseeker | ah, interesting |
| 19:05.33 | starseeker | needs to study the output more closely, sorry - was too focused on CMake that first go-around |
| 19:05.44 | mpictor | they'll be in the same dir as code, schemas/$schema/entity and /type |
| 19:32.51 | *** join/#stepcode infobot (~infobot@rikers.org) |
| 19:32.51 | *** topic/#stepcode is STEPcode - data exchange with the technologies of ISO10303 || http://stepcode.org || LOG http://ibot.rikers.org/%23stepcode/ |
| 20:54.54 | mpictor | 'make -j9' takes me 67 minutes. I hope the #include trick helps! |
| 20:57.18 | mpictor | looks like the libs have ballooned as well - libsdai_cd209 is 134MB in debug mode |
| 20:57.31 | mpictor | hopes it's much smaller when stripped |