Commit Graph

50 Commits

Author SHA1 Message Date
ThePhD
67116a67f9 Fixes #588
Tentatively supports (?) #589 with sol::yielding and sol::yielding_t
Added Catch as a cmake module
2018-02-17 00:18:26 -05:00
ThePhD
eca9ec46fd Beef up testing suite significantly and make sure that the interop examples compile, at least on Linux
These damn libraries definitely don't compile clean on Windows... but then again, most people don't give a damn about Windows :<
2018-02-12 17:01:30 -05:00
ThePhD
8d2a25988c Maybe this will finally fix all the example errors 2018-02-12 03:55:14 -05:00
ThePhD
72c7959403 Give up on x86 builds, since they're quite literally ruining most things! 2018-02-11 01:13:42 -05:00
ThePhD
6f80e271d7 these cmake details are starting to get unruly... 2018-02-11 00:53:53 -05:00
ThePhD
2e648638ef Fixes #580
Adds some code for #579 to warn of bad errors
Update documentation and examples, as well as fix fPIC with static builds on Linux
2018-02-08 23:19:45 -05:00
ThePhD
db9c5d64c6 update single, use special handling for find_package 2018-01-23 16:18:17 -05:00
ThePhD
63ec47bf6e Make base class type checks and lookup stable across DLLs 2018-01-23 12:07:22 -05:00
ThePhD
0afd822644 check platform variable 2018-01-20 12:10:50 -05:00
ThePhD
6202b91b40 x86 fixes 2018-01-20 12:02:06 -05:00
ThePhD
bfe696d2d7 put things in place for 32-bit testing and builds on both appveyor and travis
pray it doesn't shatter
2018-01-19 23:59:43 -05:00
ThePhD
1a64ba0fe5 Improve the docs, set up LuaJIT, and try to figure out why we can't build a DLL as an external projcet target with LuaJIT... 2018-01-19 15:03:22 -05:00
ThePhD
71963f9678 Try to always use the dedicated library built for the docker image by the LLVM team, otherwise everything goes to hell.... 2018-01-10 08:40:37 -05:00
ThePhD
8cf7f49ec7 commit the fun, see how it all goes down... 2018-01-08 20:33:39 -05:00
ThePhD
1b94182c8d ¯\_(ツ)_/¯ 2018-01-01 20:03:18 -05:00
ThePhD
962ff6cf61 master the use of docker. maybe 2017-12-31 21:15:43 -05:00
ThePhD
0051548dfa big sigh; one day we'll get this all right... 2017-12-27 07:42:37 -05:00
ThePhD
70b2fa0b41 unless the LLVM and MINGW builds once again 2017-12-26 20:42:40 -05:00
ThePhD
d05769b80e fix up examples and use build log to get neat messages 2017-12-26 08:04:54 -05:00
ThePhD
6000dcd490 update assert implementation 2017-12-26 01:11:09 -05:00
ThePhD
002303d52b update tests and single again
add assert.hpp for better code understanding
prepare to rewrite all the damn docs, and update the tutorials...
2017-12-25 23:27:22 -05:00
ThePhD
6d2100e814 Prepare for full appveyor build.
Add abort_clean so that CI does not get stuck on a message box that can never be answered
alignment rules do not apply to stack variables, so the tests cannot have the alignment check in the destructor: only for memory pushed into and coming straight out of Lua
2017-12-25 11:44:04 -05:00
ThePhD
f2db7e3a6f modify vanilla build to never use internal makefile
appveyor will build both versions on x64 since it can (maybe) run fast enough?
generate lua.hpp for versions 5.1, since it doesn't roll its own
2017-12-24 17:59:45 -05:00
ThePhD
83013333d7 Let it rip! 2017-12-24 12:32:23 -05:00
ThePhD
7fae78b23b update cmake build and try out new appveyor commands 2017-12-24 01:44:51 -05:00
ThePhD
453a0daaaf fix VC++ build of examples and try to get appveyor to compile and run... 2017-12-24 01:30:55 -05:00
ThePhD
9a009e5cf1 gate features behind check 2017-12-23 21:29:29 -05:00
ThePhD
9524fd061e All these platforms are so hard to deal with... 2017-12-23 20:34:34 -05:00
ThePhD
0e19156d52 update everything for the new cmake build, prepare for new .travis.yml and new appveyor.yml
Add name of project at top of license
fix container_traits with respect to boost::property_tree: Fixes #554
fix as_table with a vector of tuples (forward_get ignored constness): Fixes #552
2017-12-20 04:58:50 -05:00
ThePhD
2f3d5e6e11 update bootstrap.py for the last time...!
then, its on to do harness testing with CMake
2017-12-11 13:23:32 -05:00
ThePhD
e54b775003 CAPITAL LETTERS
BECAUSE COBOL
2017-12-11 12:34:54 -05:00
ThePhD
c0660b454c fix multiply-declared forward declaration values
transition to CATCH2 (no changes really)
fix diagnostic printout for single.py
update ignores
update CMake to download CATCH single
update single
2017-12-11 12:31:30 -05:00
ThePhD
cf81d815a9 update cmake lists for test running 2017-12-10 17:07:59 -05:00
ThePhD
e06d2fb581 add single, add examples 2017-12-10 15:56:49 -05:00
ThePhD
c3c7f42d91 update cmakelists for new tests
fix problematic macro expansion issues
2017-12-10 01:34:48 -05:00
Nicolas Cornu
96fc84b4ef tests files have been moved 2017-09-22 17:16:48 -04:00
Nicolas Cornu
46e05b92e8 Add options to disable build of tests and examples
Add a source gorup for headers.
Detect all cases for "LuaJit"
2017-09-22 17:16:48 -04:00
Nicolas Cornu
242ed15777 Use glob 2017-09-22 17:16:48 -04:00
Nicolas Cornu
bb41196048 Last changes? 2017-09-22 17:16:48 -04:00
Kevin Brightwell
0d2c125cbb Add FindLua/FindLuaJIT that behave better
Better FindLua modules as per CMake devs
2017-09-22 17:16:48 -04:00
Nicolas Cornu
cbe7d765ca Generate single_sol in source dir 2017-09-22 17:16:48 -04:00
Nicolas Cornu
2d1abc9ae7 doing shit to create sol.hpp in single directory. 2017-09-22 17:16:48 -04:00
Nicolas Cornu
412df84d58 Add message for single_sol target generation or not. 2017-09-22 17:16:48 -04:00
Nicolas Cornu
0234385f7b Fix, it was for testing purpose 2017-09-22 17:16:48 -04:00
Nicolas Cornu
98b65e6c27 Make use of python3 optionnal and generating single_sol too 2017-09-22 17:16:48 -04:00
Nicolas Cornu
084bbc0fb8 Fix different things 2017-09-22 17:16:48 -04:00
Nicolas Cornu
bba8b3fdbd Continue fixing things 2017-09-22 17:16:48 -04:00
Nicolas Cornu
368f575084 Stop globbing and add flags 2017-09-22 17:16:48 -04:00
Nicolas Cornu
a0eb0b4313 Rework CMakeLists.txt following comments 2017-09-22 17:16:48 -04:00
Nicolas Cornu
e834646dfb [WIP] Add a CMakeLists.txt 2017-09-22 17:16:48 -04:00