kevin/neovim
1
Fork 0
mirror of https://github.com/neovim/neovim.git synced 2024-12-23 20:55:18 -07:00
Code Issues Packages Projects Releases Wiki Activity
e70d6283b6
neovim/CMakeLists.txt

389 lines
13 KiB
CMake
Raw Normal View History

CMake: Set -Wconversion by default.
2014-11-18 06:01:24 -07:00
cmake_minimum_required(VERSION 2.8.7)
project(NEOVIM)
Import vim from changeset v5628:c9cad40b4181 - Cleanup source tree, leaving only files necessary for compilation/testing - Process files through unifdef to remove tons of FEAT_* macros - Process files through uncrustify to normalize source code formatting. - Port the build system to cmake
2014-01-31 06:39:15 -07:00
add custom CMake module path
2014-02-24 13:54:45 -07:00
# Point CMake at any custom modules we may ship
build: fix CMAKE_MODULE_PATH usage Fixes #1447. `CMAKE_MODULE_PATH` is meant to be a list of directories, and as such, is not the proper way to launch our scripts. Let's use `${PROJECT_SOURCE_DIR}/cmake` instead. Also, let's not outright set `CMAKE_MODULE_PATH`, but instead append our location to the list.
2014-11-10 17:26:01 -07:00
list(APPEND CMAKE_MODULE_PATH "${PROJECT_SOURCE_DIR}/cmake")
add custom CMake module path
2014-02-24 13:54:45 -07:00
Revamp the build system. This achieves several goals: * Less reliance on scripts so we have better portability to Windows (though we still have a ways to go for proper Windows support). Luajit, luarocks, moonscript, and busted are all installed via CMake now. * Trying to make use of pkg-config to get the correct libraries. The latest libuv is still broken in this regard, but we'll at least be in a position to use it. * Allow the use of Ninja or make. The former runs faster in many environments, and automatically makes use of parallel builds. This also allows for system installed dependencies--though not through the Makefile just yet--and adds support for FreeBSD. This also make us build libuv and luajit as static libraries only, since we're only concerned about having static libraries for our bundled dependencies.
2014-03-03 08:09:06 -07:00
# Prefer our bundled versions of dependencies.
Allow cmake caller to override DEPS_PREFIX - Caller can override bundled dependency location using DEPS_PREFIX - Cache variable DEPS_PREFIX, using .deps/usr by default - Removed unused variables DEPS_BIN_DIR, DEPS_BUILD_DIR, DEPS_DIR DEPS_INSTALL_DIR - Corner case: if the caller tries to override DEPS_PREFIX after a successful cmake configuration, the caller needs to clear the cache because dependency checks are based on the old value
2014-12-11 10:17:20 -07:00
set(DEPS_PREFIX "${CMAKE_CURRENT_SOURCE_DIR}/.deps/usr" CACHE PATH "Path prefix for finding dependencies")
list(INSERT CMAKE_PREFIX_PATH 0 ${DEPS_PREFIX})
build: make sure our deps area is picked up for pkg-config files
2015-02-24 11:59:29 -07:00
set(ENV{PKG_CONFIG_PATH} "$ENV{PKG_CONFIG_PATH}:${DEPS_PREFIX}/lib/pkgconfig")
Revamp the build system. This achieves several goals: * Less reliance on scripts so we have better portability to Windows (though we still have a ways to go for proper Windows support). Luajit, luarocks, moonscript, and busted are all installed via CMake now. * Trying to make use of pkg-config to get the correct libraries. The latest libuv is still broken in this regard, but we'll at least be in a position to use it. * Allow the use of Ninja or make. The former runs faster in many environments, and automatically makes use of parallel builds. This also allows for system installed dependencies--though not through the Makefile just yet--and adds support for FreeBSD. This also make us build libuv and luajit as static libraries only, since we're only concerned about having static libraries for our bundled dependencies.
2014-03-03 08:09:06 -07:00
cmake: Enable -fdiagnostics-color=auto gcc flag. #3137
2015-08-27 00:56:46 -07:00
# used for check_c_compiler_flag
include(CheckCCompilerFlag)
build: give priority to /sw and /opt/local on Mac OS X Unfortunately, we can't force the specific inclusion of a header file. So if anything add /opt/local/include to the include path--such as libintl--then other dependencies might be drawn from /opt/local at compile time, even though we detected them elsewhere at configure time. This, in turn, causes issues with mixed versions, such as the iconv.h header being pulled from /opt/local/include, but linked against the library in /usr/lib--which can be mismatched versions. So, despite CMake's best effort to treat /sw and /opt/local as just another system area, we really need to give them preferential treatment. To do this, we add them to CMAKE_PREFIX_PATH. This fixes an issue discovered while re-enabling iconv in #1370.
2014-11-08 14:16:39 -07:00
if(CMAKE_SYSTEM_NAME STREQUAL "Darwin")
# CMake tries to treat /sw and /opt/local as extension of the system path, but
# that doesn't really work out very well. Once you have a dependency that
# resides there and have to add it as an include directory, then any other
# dependency that could be satisfied from there must be--otherwise you can end
# up with conflicting versions. So, let's make them more of a priority having
# them be included as one of the first places to look for dependencies.
list(APPEND CMAKE_PREFIX_PATH /sw /opt/local)
# Work around some old, broken detection by CMake for knowing when to use the
# isystem flag. Apple's compilers have supported this for quite some time
# now.
if(CMAKE_COMPILER_IS_GNUCC)
set(CMAKE_INCLUDE_SYSTEM_FLAG_C "-isystem ")
endif()
if(CMAKE_COMPILER_IS_GNUCXX)
set(CMAKE_INCLUDE_SYSTEM_FLAG_CXX "-isystem ")
endif()
CMakeLists: enable USE_FNAME_CASE on Mac
2015-03-26 20:30:45 -07:00
# Enable fixing case-insensitive filenames for Mac.
set(USE_FNAME_CASE TRUE)
build: give priority to /sw and /opt/local on Mac OS X Unfortunately, we can't force the specific inclusion of a header file. So if anything add /opt/local/include to the include path--such as libintl--then other dependencies might be drawn from /opt/local at compile time, even though we detected them elsewhere at configure time. This, in turn, causes issues with mixed versions, such as the iconv.h header being pulled from /opt/local/include, but linked against the library in /usr/lib--which can be mismatched versions. So, despite CMake's best effort to treat /sw and /opt/local as just another system area, we really need to give them preferential treatment. To do this, we add them to CMAKE_PREFIX_PATH. This fixes an issue discovered while re-enabling iconv in #1370.
2014-11-08 14:16:39 -07:00
endif()
version: Add compilation info.
2014-11-09 05:52:15 -07:00
# Set available build types for CMake GUIs.
# A different build type can still be set by -DCMAKE_BUILD_TYPE=...
set_property(CACHE CMAKE_BUILD_TYPE PROPERTY
STRINGS "Debug" "Release" "MinSizeRel" "RelWithDebInfo")
# Set default build type.
if(NOT CMAKE_BUILD_TYPE)
message(STATUS "CMAKE_BUILD_TYPE not given; setting to 'RelWithDebInfo'.")
set(CMAKE_BUILD_TYPE "RelWithDebInfo" CACHE STRING "Choose the type of build." FORCE)
endif()
version: generate "build number" from commit timestamp - cmake: git_timestamp() returns last commit time formatted as `YYYYMMddHHmm`. - Always include commit hash in :version and --version output. `nvim --version` sample output: NVIM 0.0.0-alpha+201410070245 (compiled Oct 7 2014 05:30:45) Commit: f747b2b1ff7bfe7eb00cc2be82d7af87c98f1111
2014-10-06 22:45:05 -07:00
# Version tokens
version: report commit hash as "build number" in long version string
2014-10-04 15:32:47 -07:00
include(GetGitRevisionDescription)
CMake: Use project directory to look for Git revision. If downloading Neovim as a tarball (i.e. without Git data), building Neovim will search parent directories for a .git directory. Explicitly set GIT_DIR to the project directory to avoid that.
2015-09-08 13:54:57 -07:00
file(TO_NATIVE_PATH ${CMAKE_CURRENT_LIST_DIR}/.git GIT_DIR)
version: generate "build number" from commit timestamp - cmake: git_timestamp() returns last commit time formatted as `YYYYMMddHHmm`. - Always include commit hash in :version and --version output. `nvim --version` sample output: NVIM 0.0.0-alpha+201410070245 (compiled Oct 7 2014 05:30:45) Commit: f747b2b1ff7bfe7eb00cc2be82d7af87c98f1111
2014-10-06 22:45:05 -07:00
get_git_head_revision(GIT_REFSPEC NVIM_VERSION_COMMIT)
CMake: Modify _FORTIFY_SOURCE only for release builds. Remove build warning for Debug builds: # warning _FORTIFY_SOURCE requires compiling with optimization (-O)
2014-11-05 03:54:15 -07:00
if(NOT NVIM_VERSION_COMMIT)
cmake: handle missing git or .git/ GetGitRevisionDescription.cmake: we don't need fine-grained failure modes, we only need "yes" or "no". fix #1292
2014-10-15 11:40:18 -07:00
set(NVIM_VERSION_COMMIT "?")
endif()
version: report commit hash as "build number" in long version string
2014-10-04 15:32:47 -07:00
set(NVIM_VERSION_MAJOR 0)
set(NVIM_VERSION_MINOR 0)
set(NVIM_VERSION_PATCH 0)
set(NVIM_VERSION_PRERELEASE "-alpha")
version: generate "build number" from commit timestamp - cmake: git_timestamp() returns last commit time formatted as `YYYYMMddHHmm`. - Always include commit hash in :version and --version output. `nvim --version` sample output: NVIM 0.0.0-alpha+201410070245 (compiled Oct 7 2014 05:30:45) Commit: f747b2b1ff7bfe7eb00cc2be82d7af87c98f1111
2014-10-06 22:45:05 -07:00
# TODO(justinmk): UTC time would be nice here #1071
git_timestamp(GIT_TIMESTAMP)
version: report commit hash as "build number" in long version string
2014-10-04 15:32:47 -07:00
# TODO(justinmk): do not set this for "release" builds #1071
CMake: Modify _FORTIFY_SOURCE only for release builds. Remove build warning for Debug builds: # warning _FORTIFY_SOURCE requires compiling with optimization (-O)
2014-11-05 03:54:15 -07:00
if(GIT_TIMESTAMP)
cmake: handle missing git or .git/ GetGitRevisionDescription.cmake: we don't need fine-grained failure modes, we only need "yes" or "no". fix #1292
2014-10-15 11:40:18 -07:00
set(NVIM_VERSION_BUILD "+${GIT_TIMESTAMP}")
endif()
version: Add compilation info.
2014-11-09 05:52:15 -07:00
set(NVIM_VERSION_BUILD_TYPE "${CMAKE_BUILD_TYPE}")
# NVIM_VERSION_CFLAGS set further below.
Import vim from changeset v5628:c9cad40b4181 - Cleanup source tree, leaving only files necessary for compilation/testing - Process files through unifdef to remove tons of FEAT_* macros - Process files through uncrustify to normalize source code formatting. - Port the build system to cmake
2014-01-31 06:39:15 -07:00
set(CMAKE_EXPORT_COMPILE_COMMANDS ON)
build: default to using -O2 for release builds The -O3 optimization level can often lead to dangerous (and sometimes incorrect) optimizations being performed. So let's use a level that's more stable.
2014-11-07 04:21:54 -07:00
# Default to -O2 on release builds.
string(REPLACE "-O3" "-O2" CMAKE_C_FLAGS_RELEASE "${CMAKE_C_FLAGS_RELEASE}")
CMake: Set -Wconversion by default.
2014-11-18 06:01:24 -07:00
# Enable -Wconversion.
MSVC: use MSVC instead of GCC-like compiler flags. - We can build with /W4, but it gives thousands of warnings. - _CRT_NONSTDC_NO_DEPRECATE disables warnings about POSIX functions. https://msdn.microsoft.com/en-us/library/ms235384.aspx - _CRT_SECURE_NO_WARNINGS https://msdn.microsoft.com/en-us/library/8ef0s5kh.aspx
2015-08-27 19:00:09 -07:00
if(NOT MSVC)
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -Wconversion")
endif()
CMake: Set -Wconversion by default.
2014-11-18 06:01:24 -07:00
build: turn down the _FORTIFY_SOURCE level on gcc >= 4.0
2014-11-04 15:06:18 -07:00
# gcc 4.0 and better turn on _FORTIFY_SOURCE=2 automatically. This currently
# does not work with Neovim due to some uses of dynamically-sized structures.
# See https://github.com/neovim/neovim/issues/223 for details.
build: probe the compiler to see if we need to turn down _FORTIFY_SOURCE Some builds don't use Release (such as Gentoo, though it can be told to). So let's go a bit further and probe the compiler to see if we need to turn down the _FORTIFY_SOURCE level.
2014-11-21 01:42:59 -07:00
include(CheckCSourceCompiles)
build: include the flags for the build type in the _FORTIFY_SOURCE check It turns out the check was being performed without optimizations enabled even when the CMAKE_BUILD_TYPE was set to a release build. This led to _FORTIFY_SOURCE's level not being correctly determined, and us failing to apply the correct workaround. To counter this, we'll take the default flags for the build type and apply them. Also, if options are passed via CFLAGS, they are automatically passed on to the underlying build. So this should cover all the necessary ground. This fixes #1647.
2014-12-16 04:02:42 -07:00
# Include the build type's default flags in the check for _FORTIFY_SOURCE,
# otherwise we may incorrectly identify the level as acceptable and find out
# later that it was not when optimizations were enabled. CFLAGS is applied
# even though you don't see it in CMAKE_REQUIRED_FLAGS.
set(INIT_FLAGS_NAME CMAKE_C_FLAGS_${CMAKE_BUILD_TYPE})
string(TOUPPER ${INIT_FLAGS_NAME} INIT_FLAGS_NAME)
if(${INIT_FLAGS_NAME})
set(CMAKE_REQUIRED_FLAGS "${${INIT_FLAGS_NAME}}")
endif()
build: probe the compiler to see if we need to turn down _FORTIFY_SOURCE Some builds don't use Release (such as Gentoo, though it can be told to). So let's go a bit further and probe the compiler to see if we need to turn down the _FORTIFY_SOURCE level.
2014-11-21 01:42:59 -07:00
check_c_source_compiles("
#if defined(_FORTIFY_SOURCE) && _FORTIFY_SOURCE > 1
#error \"_FORTIFY_SOURCE > 1\"
#endif
int
main(void)
{
return 0;
}
CMake: Fix _FORTIFY_SOURCE detection when using hardening-wrapper. #2788 When checking code with check_c_source_compiles, the "check variable" (for the _FORTIFY_SOURCE detection: previously _FORTIFY_SOURCE_ACCEPTABLE, now HAS_ACCEPTABLE_FORTIFY) is passed to the compiler like -D_FORTIFY_SOURCE_ACCEPTABLE. This throws off hardening-wrapper [1], which then NOT sets _FORTIFY_SOURCE=2 for the check, assuming it had already been set manually as it detected -D_FORTIFY_SOURCE*. Renaming the "check variable" to not match this pattern works around this problem. [1] https://github.com/thestinger/hardening-wrapper
2015-06-04 02:08:26 -07:00
" HAS_ACCEPTABLE_FORTIFY)
build: probe the compiler to see if we need to turn down _FORTIFY_SOURCE Some builds don't use Release (such as Gentoo, though it can be told to). So let's go a bit further and probe the compiler to see if we need to turn down the _FORTIFY_SOURCE level.
2014-11-21 01:42:59 -07:00
CMake: Fix _FORTIFY_SOURCE detection when using hardening-wrapper. #2788 When checking code with check_c_source_compiles, the "check variable" (for the _FORTIFY_SOURCE detection: previously _FORTIFY_SOURCE_ACCEPTABLE, now HAS_ACCEPTABLE_FORTIFY) is passed to the compiler like -D_FORTIFY_SOURCE_ACCEPTABLE. This throws off hardening-wrapper [1], which then NOT sets _FORTIFY_SOURCE=2 for the check, assuming it had already been set manually as it detected -D_FORTIFY_SOURCE*. Renaming the "check variable" to not match this pattern works around this problem. [1] https://github.com/thestinger/hardening-wrapper
2015-06-04 02:08:26 -07:00
if(NOT HAS_ACCEPTABLE_FORTIFY)
message(STATUS "Unsupported _FORTIFY_SOURCE found, forcing _FORTIFY_SOURCE=1.")
Detect and propagate _FORTIFY_SOURCE prefix Get prefix to a -D_FORTIFY_SOURCE string if it is present in CFLAGS and apply the prefix to flags added to redefine _FORTIFY_SOURCE in CFLAGS and CPPFLAGS * fixes 1569
2014-12-12 10:45:21 -07:00
# Extract possible prefix to _FORTIFY_SOURCE (e.g. -Wp,-D_FORTIFY_SOURCE).
STRING(REGEX MATCH "[^\ ]+-D_FORTIFY_SOURCE" _FORTIFY_SOURCE_PREFIX "${CMAKE_C_FLAGS}")
STRING(REPLACE "-D_FORTIFY_SOURCE" "" _FORTIFY_SOURCE_PREFIX "${_FORTIFY_SOURCE_PREFIX}" )
cmake: Build with -fstack-protector-strong if available #2597 If not available, fall back to -fstack-protector + --param=ssp-buffer-size=4 If that isn't available, do nothing. See the following articles for more information: https://lwn.net/Articles/584225/ https://outflux.net/blog/archives/2014/01/27/fstack-protector-strong/
2015-08-26 15:00:33 -07:00
if(NOT _FORTIFY_SOURCE_PREFIX STREQUAL "")
CMake: Fix _FORTIFY_SOURCE detection when using hardening-wrapper. #2788 When checking code with check_c_source_compiles, the "check variable" (for the _FORTIFY_SOURCE detection: previously _FORTIFY_SOURCE_ACCEPTABLE, now HAS_ACCEPTABLE_FORTIFY) is passed to the compiler like -D_FORTIFY_SOURCE_ACCEPTABLE. This throws off hardening-wrapper [1], which then NOT sets _FORTIFY_SOURCE=2 for the check, assuming it had already been set manually as it detected -D_FORTIFY_SOURCE*. Renaming the "check variable" to not match this pattern works around this problem. [1] https://github.com/thestinger/hardening-wrapper
2015-06-04 02:08:26 -07:00
message(STATUS "Detected _FORTIFY_SOURCE Prefix=${_FORTIFY_SOURCE_PREFIX}.")
Detect and propagate _FORTIFY_SOURCE prefix Get prefix to a -D_FORTIFY_SOURCE string if it is present in CFLAGS and apply the prefix to flags added to redefine _FORTIFY_SOURCE in CFLAGS and CPPFLAGS * fixes 1569
2014-12-12 10:45:21 -07:00
endif()
build: turn down the _FORTIFY_SOURCE level on gcc >= 4.0
2014-11-04 15:06:18 -07:00
# -U in add_definitions doesn't end up in the correct spot, so we add it to
# the flags variable instead.
Detect and propagate _FORTIFY_SOURCE prefix Get prefix to a -D_FORTIFY_SOURCE string if it is present in CFLAGS and apply the prefix to flags added to redefine _FORTIFY_SOURCE in CFLAGS and CPPFLAGS * fixes 1569
2014-12-12 10:45:21 -07:00
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${_FORTIFY_SOURCE_PREFIX}-U_FORTIFY_SOURCE ${_FORTIFY_SOURCE_PREFIX}-D_FORTIFY_SOURCE=1")
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${_FORTIFY_SOURCE_PREFIX}-U_FORTIFY_SOURCE ${_FORTIFY_SOURCE_PREFIX}-D_FORTIFY_SOURCE=1")
build: turn down the _FORTIFY_SOURCE level on gcc >= 4.0
2014-11-04 15:06:18 -07:00
endif()
MSVC: use MSVC instead of GCC-like compiler flags. - We can build with /W4, but it gives thousands of warnings. - _CRT_NONSTDC_NO_DEPRECATE disables warnings about POSIX functions. https://msdn.microsoft.com/en-us/library/ms235384.aspx - _CRT_SECURE_NO_WARNINGS https://msdn.microsoft.com/en-us/library/8ef0s5kh.aspx
2015-08-27 19:00:09 -07:00
if(MSVC)
# XXX: /W4 gives too many warnings. #3241
add_definitions(/W3 -D_CRT_SECURE_NO_WARNINGS -D_CRT_NONSTDC_NO_DEPRECATE)
else()
add_definitions(-Wall -Wextra -pedantic -Wno-unused-parameter
Use strict function prototypes #945 `-Wstrict-prototypes` warn if a function is declared or defined without specifying the argument types. This warning disallow function prototypes with empty parameter list. In C, a function declared with an empty parameter list accepts an arbitrary number of arguments when being called. This is for historic reasons; originally, C functions didn't have prototypes, as C evolved from B, a typeless language. When prototypes were added, the original typeless declarations were left in the language for backwards compatibility. Instead we should provide `void` in argument list to state that function doesn't have arguments. Also this warning disallow declaring type of the parameters after the parentheses because Neovim header generator produce no declarations for old-stlyle prototypes: it expects to find `{` after prototype.
2014-07-13 00:03:07 -07:00
-Wstrict-prototypes -std=gnu99)
MSVC: use MSVC instead of GCC-like compiler flags. - We can build with /W4, but it gives thousands of warnings. - _CRT_NONSTDC_NO_DEPRECATE disables warnings about POSIX functions. https://msdn.microsoft.com/en-us/library/ms235384.aspx - _CRT_SECURE_NO_WARNINGS https://msdn.microsoft.com/en-us/library/8ef0s5kh.aspx
2015-08-27 19:00:09 -07:00
endif()
cmake: Build with -fstack-protector-strong if available #2597 If not available, fall back to -fstack-protector + --param=ssp-buffer-size=4 If that isn't available, do nothing. See the following articles for more information: https://lwn.net/Articles/584225/ https://outflux.net/blog/archives/2014/01/27/fstack-protector-strong/
2015-08-26 15:00:33 -07:00
if(MINGW)
Mingw: Use POSIX compatible stdio #3232 - There are some differences between stdio (*printf) functions in POSIX and the MS runtime, this commit enables Mingw compatibility for these functions
2014-07-03 06:32:09 -07:00
# Use POSIX compatible stdio in Mingw
add_definitions(-D__USE_MINGW_ANSI_STDIO)
endif()
Turn off -Werror by default. It inteferes with development activities by breaking your build in the middle of a refactor. Instead, let's enable -Werror on the Travis CI builds via a TRAVIS_CI_BUILD option.
2014-05-14 16:08:41 -07:00
cmake: Build with -fstack-protector-strong if available #2597 If not available, fall back to -fstack-protector + --param=ssp-buffer-size=4 If that isn't available, do nothing. See the following articles for more information: https://lwn.net/Articles/584225/ https://outflux.net/blog/archives/2014/01/27/fstack-protector-strong/
2015-08-26 15:00:33 -07:00
check_c_compiler_flag(-fstack-protector-strong HAS_FSTACK_PROTECTOR_STRONG_FLAG)
check_c_compiler_flag(-fstack-protector HAS_FSTACK_PROTECTOR_FLAG)
if(HAS_FSTACK_PROTECTOR_STRONG_FLAG)
add_definitions(-fstack-protector-strong)
elseif(HAS_FSTACK_PROTECTOR_FLAG)
add_definitions(-fstack-protector --param ssp-buffer-size=4)
endif()
cmake: Enable -fdiagnostics-color=auto gcc flag. #3137
2015-08-27 00:56:46 -07:00
check_c_compiler_flag(-fdiagnostics-color=auto HAS_DIAG_COLOR_FLAG)
if(HAS_DIAG_COLOR_FLAG)
add_definitions(-fdiagnostics-color=auto)
endif()
Turn off -Werror by default. It inteferes with development activities by breaking your build in the middle of a refactor. Instead, let's enable -Werror on the Travis CI builds via a TRAVIS_CI_BUILD option.
2014-05-14 16:08:41 -07:00
option(
TRAVIS_CI_BUILD "Travis CI build. Extra compilation flags will be set." OFF)
if(TRAVIS_CI_BUILD)
message(STATUS "Travis CI build enabled.")
add_definitions(-Werror)
endif()
Use CMAKE_C_FLAGS variable to set compiler flags. If the compiler is some GNU-alike variant, set the compiler flags to use the gnu99 dialect of C and enable all warnings. Non-GNU compilers may have to have their own magic added to set dialect and enable warnings. Closes #179.
2014-02-27 05:27:20 -07:00
build: only attempt to use -Og if it's supported Commit a1d411f9c991c03488c74c6266eb37cc2bab970e just assumes that gcc will support the `-Og` option, but gcc that comes with Ubuntu 12.04 does not. Let's check to see if the flag is supported, and then decide whether to enable it or not.
2014-11-09 04:58:17 -07:00
if(CMAKE_COMPILER_IS_GNUCC)
check_c_compiler_flag(-Og HAS_OG_FLAG)
else()
set(HAS_OG_FLAG 0)
endif()
CMake: Set custom RelWithDebInfo compile flags.
2014-11-08 04:49:04 -07:00
# Set custom build flags for RelWithDebInfo.
# -DNDEBUG purposely omitted because we want assertions.
build: only attempt to use -Og if it's supported Commit a1d411f9c991c03488c74c6266eb37cc2bab970e just assumes that gcc will support the `-Og` option, but gcc that comes with Ubuntu 12.04 does not. Let's check to see if the flag is supported, and then decide whether to enable it or not.
2014-11-09 04:58:17 -07:00
if(HAS_OG_FLAG)
CMake: Set custom RelWithDebInfo compile flags.
2014-11-08 04:49:04 -07:00
set(CMAKE_C_FLAGS_RELWITHDEBINFO "-Og -g"
CACHE STRING "Flags used by the compiler during release builds with debug info." FORCE)
MSVC: Don't use GCC -O/-g flags. #3267
2015-05-17 05:59:31 -07:00
elseif(NOT MSVC)
CMake: Set custom RelWithDebInfo compile flags.
2014-11-08 04:49:04 -07:00
set(CMAKE_C_FLAGS_RELWITHDEBINFO "-O2 -g"
CACHE STRING "Flags used by the compiler during release builds with debug info." FORCE)
endif()
Add automatic generation of headers - The 'stripdecls.py' script replaces declarations in all headers by includes to generated headers. `ag '#\s*if(?!ndef NEOVIM_).*((?!#\s*endif).*\n)*#ifdef INCLUDE_GENERATED'` was used for this. - Add and integrate gendeclarations.lua into the build system to generate the required includes. - Add -Wno-unused-function - Made a bunch of old-style definitions ANSI This adds a requirement: all type and structure definitions must be present before INCLUDE_GENERATED_DECLARATIONS-protected include. Warning: mch_expandpath (path.h.generated.h) was moved manually. So far it is the only exception.
2014-05-10 06:24:13 -07:00
Import vim from changeset v5628:c9cad40b4181 - Cleanup source tree, leaving only files necessary for compilation/testing - Process files through unifdef to remove tons of FEAT_* macros - Process files through uncrustify to normalize source code formatting. - Port the build system to cmake
2014-01-31 06:39:15 -07:00
if(CMAKE_BUILD_TYPE MATCHES Debug)
set(DEBUG 1)
else()
set(DEBUG 0)
endif()
CMake: Set custom RelWithDebInfo compile flags.
2014-11-08 04:49:04 -07:00
add_definitions(-DINCLUDE_GENERATED_DECLARATIONS)
add_definitions(-DHAVE_CONFIG_H)
Don't allow undefined references under Linux. Many other systems expect this already, but on Linux the default is to allow them. Let's turn that off.
2014-05-14 18:14:28 -07:00
if(CMAKE_COMPILER_IS_GNUCXX AND CMAKE_SYSTEM_NAME STREQUAL "Linux")
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -Wl,--no-undefined")
set(CMAKE_SHARED_LINKER_FLAGS "${CMAKE_SHARED_LINKER_FLAGS} -Wl,--no-undefined")
set(CMAKE_MODULE_LINKER_FLAGS "${CMAKE_MODULE_LINKER_FLAGS} -Wl,--no-undefined")
endif()
illumos requires the use of limits.h for things like INT_MAX #2049
2015-02-23 13:45:11 -07:00
if(CMAKE_COMPILER_IS_GNUCXX AND CMAKE_SYSTEM_NAME STREQUAL "SunOS")
set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -Wl,--no-undefined -lsocket")
endif()
Add missing linker flags required for LuaJIT on 64-bit Mac OS X. This was brought up in #1294.
2014-10-17 03:50:13 -07:00
if(CMAKE_SYSTEM_NAME STREQUAL "Darwin" AND CMAKE_SIZEOF_VOID_P EQUAL 8)
# Required for luajit.
set(CMAKE_EXE_LINKER_FLAGS
"${CMAKE_EXE_LINKER_FLAGS} -pagezero_size 10000 -image_base 100000000")
set(CMAKE_SHARED_LINKER_FLAGS
"${CMAKE_SHARED_LINKER_FLAGS} -image_base 100000000")
set(CMAKE_MODULE_LINKER_FLAGS
"${CMAKE_MODULE_LINKER_FLAGS} -image_base 100000000")
endif()
Fix #451: Build broken on OSX. Problem: Build breaks in OSX if macports ncurses present, due to header conflict involving 'term.h'. Solution: Modify include search order.
2014-04-03 12:00:43 -07:00
include_directories("${PROJECT_BINARY_DIR}/config")
include_directories("${PROJECT_SOURCE_DIR}/src")
link to rt if it provides clock_gettime As noted in #128, if clock_gettime is provided by librt then it does not end up being linked into the static libuv.a binary. This might be considered a bug in libuv but we can address it here. Detect if librt provides the clock_gettime symbol and, if so, append it to the list of libraries linked into nvim. On non-librt systems the behaviour should be as before.
2014-02-26 03:06:59 -07:00
# Modules used by platform auto-detection
include(CheckLibraryExists)
Fix #451: Build broken on OSX. Problem: Build breaks in OSX if macports ncurses present, due to header conflict involving 'term.h'. Solution: Modify include search order.
2014-04-03 12:00:43 -07:00
find_package(LibUV REQUIRED)
Tell cmake to use `-isystem` third-party includes This adds the `SYSTEM` parameter to `include_directories`, which will tell cmake to use `-isystem` instead of `-I` for specifying include directories. One advantage is that compilers won't emit warnings for included files that belong to dependencies.
2014-05-15 04:21:13 -07:00
include_directories(SYSTEM ${LIBUV_INCLUDE_DIRS})
Fix #451: Build broken on OSX. Problem: Build breaks in OSX if macports ncurses present, due to header conflict involving 'term.h'. Solution: Modify include search order.
2014-04-03 12:00:43 -07:00
Add cmake module for finding msgpack
2014-04-10 11:39:34 -07:00
find_package(Msgpack REQUIRED)
Tell cmake to use `-isystem` third-party includes This adds the `SYSTEM` parameter to `include_directories`, which will tell cmake to use `-isystem` instead of `-I` for specifying include directories. One advantage is that compilers won't emit warnings for included files that belong to dependencies.
2014-05-15 04:21:13 -07:00
include_directories(SYSTEM ${MSGPACK_INCLUDE_DIRS})
Add cmake module for finding msgpack
2014-04-10 11:39:34 -07:00
Fix #451: Build broken on OSX. Problem: Build breaks in OSX if macports ncurses present, due to header conflict involving 'term.h'. Solution: Modify include search order.
2014-04-03 12:00:43 -07:00
find_package(LuaJit REQUIRED)
Tell cmake to use `-isystem` third-party includes This adds the `SYSTEM` parameter to `include_directories`, which will tell cmake to use `-isystem` instead of `-I` for specifying include directories. One advantage is that compilers won't emit warnings for included files that belong to dependencies.
2014-05-15 04:21:13 -07:00
include_directories(SYSTEM ${LUAJIT_INCLUDE_DIRS})
Fix #451: Build broken on OSX. Problem: Build breaks in OSX if macports ncurses present, due to header conflict involving 'term.h'. Solution: Modify include search order.
2014-04-03 12:00:43 -07:00
CMake: Refer to Unibilium instead of LibUnibilium.
2015-04-19 12:51:31 -07:00
find_package(Unibilium REQUIRED)
include_directories(SYSTEM ${UNIBILIUM_INCLUDE_DIRS})
ui: Rewrite the builtin terminal UI Now all terminal-handling code was moved to src/nvim/tui, which implements a new terminal UI based on libtermkey and unibilium
2015-02-13 08:06:05 -07:00
find_package(LibTermkey REQUIRED)
Fixed compilation when using custom libtermkey (not installed in base system).
2015-03-14 23:07:06 -07:00
include_directories(SYSTEM ${LIBTERMKEY_INCLUDE_DIRS})
ui: Rewrite the builtin terminal UI Now all terminal-handling code was moved to src/nvim/tui, which implements a new terminal UI based on libtermkey and unibilium
2015-02-13 08:06:05 -07:00
deps: Add libvterm to the project dependencies libvterm is a terminal emulation library with abstract display. It will be used to implement builtin terminal emulation into Nvim.
2015-02-28 06:41:53 -07:00
find_package(LibVterm REQUIRED)
include_directories(SYSTEM ${LIBVTERM_INCLUDE_DIRS})
CMake: Add option for Clang TSan, rename option ASAN -> ASAN_UBSAN.
2015-06-09 01:43:22 -07:00
option(CLANG_ASAN_UBSAN "Enable Clang address & undefined behavior sanitizer for nvim binary." OFF)
CMake: Add option for Clang MSan.
2015-06-08 09:49:23 -07:00
option(CLANG_MSAN "Enable Clang memory sanitizer for nvim binary." OFF)
CMake: Add option for Clang TSan, rename option ASAN -> ASAN_UBSAN.
2015-06-09 01:43:22 -07:00
option(CLANG_TSAN "Enable Clang thread sanitizer for nvim binary." OFF)
if((CLANG_ASAN_UBSAN AND CLANG_MSAN)
OR (CLANG_ASAN_UBSAN AND CLANG_TSAN)
OR (CLANG_MSAN AND CLANG_TSAN))
message(FATAL_ERROR "Sanitizers cannot be enabled simultaneously.")
CMake: Add option for Clang MSan.
2015-06-08 09:49:23 -07:00
endif()
CMake: Add option for Clang TSan, rename option ASAN -> ASAN_UBSAN.
2015-06-09 01:43:22 -07:00
if((CLANG_ASAN_UBSAN OR CLANG_MSAN OR CLANG_TSAN) AND NOT CMAKE_C_COMPILER_ID MATCHES "Clang")
CMake: Add option for Clang MSan.
2015-06-08 09:49:23 -07:00
message(FATAL_ERROR "Sanitizers are only supported for Clang.")
deps: Add jemalloc as an optional dependency Jemalloc will be used if the cmake option `USE_JEMALLOC` is enabled(which is the default). To avoid trouble with clang's ASAN, it is disabled by default if the `SANITIZE` option is enabled. Since jemalloc has thread cache for small objects, it fills the gap created by removing klib memory pools. The `xstrdup` funciton(memory.c) had to be reimplemented on top of `xmalloc` to make it work with a custom allocator.
2015-04-12 07:40:08 -07:00
endif()
CMake: Add option for Clang TSan, rename option ASAN -> ASAN_UBSAN.
2015-06-09 01:43:22 -07:00
if(CLANG_ASAN_UBSAN OR CLANG_MSAN OR CLANG_TSAN)
CMake: Add option for Clang MSan.
2015-06-08 09:49:23 -07:00
message(STATUS "Sanitizers have been enabled; don't use jemalloc.")
else()
deps: Add jemalloc as an optional dependency Jemalloc will be used if the cmake option `USE_JEMALLOC` is enabled(which is the default). To avoid trouble with clang's ASAN, it is disabled by default if the `SANITIZE` option is enabled. Since jemalloc has thread cache for small objects, it fills the gap created by removing klib memory pools. The `xstrdup` funciton(memory.c) had to be reimplemented on top of `xmalloc` to make it work with a custom allocator.
2015-04-12 07:40:08 -07:00
find_package(JeMalloc)
if(JEMALLOC_FOUND)
include_directories(SYSTEM ${JEMALLOC_INCLUDE_DIRS})
endif()
endif()
Fix #451: Build broken on OSX. Problem: Build breaks in OSX if macports ncurses present, due to header conflict involving 'term.h'. Solution: Modify include search order.
2014-04-03 12:00:43 -07:00
find_package(LibIntl)
if(LibIntl_FOUND)
build: use the proper libintl include variable What was there worked, but it wasn't meant to be the variable that you use for inclusion.
2014-11-08 14:09:47 -07:00
include_directories(SYSTEM ${LibIntl_INCLUDE_DIRS})
Fix #451: Build broken on OSX. Problem: Build breaks in OSX if macports ncurses present, due to header conflict involving 'term.h'. Solution: Modify include search order.
2014-04-03 12:00:43 -07:00
endif()
build: pull iconv detection into its own FindIconv.cmake file This will provide better control for those who may want to alter which one gets used.
2014-11-08 14:18:04 -07:00
find_package(Iconv)
if(Iconv_FOUND)
include_directories(SYSTEM ${Iconv_INCLUDE_DIRS})
Fix #451: Build broken on OSX. Problem: Build breaks in OSX if macports ncurses present, due to header conflict involving 'term.h'. Solution: Modify include search order.
2014-04-03 12:00:43 -07:00
endif()
use CMake's built in pthread detection CMake ships with a standard FindThreads module which can be used to a) test for a threading library and b) confirm that it is pthread. It also allows the hard-coding of the threading library name to be removed from ``src/CMakeLists.txt``. Make it an error not to have a pthread library installed and indicate to CMake that we strongly prefer pthread to any other platform threading library.
2014-02-24 11:52:12 -07:00
# Determine platform's threading library. Set CMAKE_THREAD_PREFER_PTHREAD
Remove pthread fatal check from CMakeLists.txt #891
2014-06-25 14:47:27 -07:00
# explicitly to indicate a strong preference for pthread.
use CMake's built in pthread detection CMake ships with a standard FindThreads module which can be used to a) test for a threading library and b) confirm that it is pthread. It also allows the hard-coding of the threading library name to be removed from ``src/CMakeLists.txt``. Make it an error not to have a pthread library installed and indicate to CMake that we strongly prefer pthread to any other platform threading library.
2014-02-24 11:52:12 -07:00
set(CMAKE_THREAD_PREFER_PTHREAD ON)
find_package(Threads REQUIRED)
Revamp the build system. This achieves several goals: * Less reliance on scripts so we have better portability to Windows (though we still have a ways to go for proper Windows support). Luajit, luarocks, moonscript, and busted are all installed via CMake now. * Trying to make use of pkg-config to get the correct libraries. The latest libuv is still broken in this regard, but we'll at least be in a position to use it. * Allow the use of Ninja or make. The former runs faster in many environments, and automatically makes use of parallel builds. This also allows for system installed dependencies--though not through the Makefile just yet--and adds support for FreeBSD. This also make us build libuv and luajit as static libraries only, since we're only concerned about having static libraries for our bundled dependencies.
2014-03-03 08:09:06 -07:00
set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR}/bin)
Allow building as a static -fPIC library
2014-11-30 19:22:46 -07:00
set(CMAKE_LIBRARY_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR}/lib)
set(CMAKE_ARCHIVE_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR}/lib)
Changed binary output directory from src/ to bin/
2014-02-22 08:30:50 -07:00
Check dependencies for each lua interpreter. - When USE_BUNDLED=OFF and the system has multiple lua interpreters (luajit AND lua) it may occur that only the later has the needed dependencies (lua-lpeg, lua-cmsgpack). If we pick luajit then finding the dependencies FAILS. - This commit groups detection of the lua interpreter with the lua dependencies it tries to find them for both the interpreters and choses the first one that has them. Helped-by: John Szakmeister <john@szakmeister.net>
2014-07-11 04:12:18 -07:00
# Find Lua interpreter
Refactor the lua dependency checking. This is in preparation for the next step, which is to find a suitable lua interpreter, rather than just erroring when an interpreter is found but doesn't have the necessary dependencies. Helped-by: John Szakmeister <john@szakmeister.net>
2014-07-11 04:12:10 -07:00
include(LuaHelpers)
deps: Switch lua msgpack library to one that supports 2.0
2014-09-16 10:12:58 -07:00
set(LUA_DEPENDENCIES lpeg MessagePack bit)
Check dependencies for each lua interpreter. - When USE_BUNDLED=OFF and the system has multiple lua interpreters (luajit AND lua) it may occur that only the later has the needed dependencies (lua-lpeg, lua-cmsgpack). If we pick luajit then finding the dependencies FAILS. - This commit groups detection of the lua interpreter with the lua dependencies it tries to find them for both the interpreters and choses the first one that has them. Helped-by: John Szakmeister <john@szakmeister.net>
2014-07-11 04:12:18 -07:00
if(NOT LUA_PRG)
foreach(CURRENT_LUA_PRG luajit lua)
# If LUA_PRG is set find_program() will not search
unset(LUA_PRG CACHE)
unset(LUA_PRG_WORKS)
find_program(LUA_PRG ${CURRENT_LUA_PRG})
if(LUA_PRG)
check_lua_deps(${LUA_PRG} "${LUA_DEPENDENCIES}" LUA_PRG_WORKS)
if(LUA_PRG_WORKS)
break()
endif()
endif()
endforeach()
else()
check_lua_deps(${LUA_PRG} "${LUA_DEPENDENCIES}" LUA_PRG_WORKS)
endif()
Add checks for lpeg/cmsgpack lua packages These checks will stop the build process with meaningful error messages if any of those packages are not installed.
2014-04-13 03:06:35 -07:00
Refactor the lua dependency checking. This is in preparation for the next step, which is to find a suitable lua interpreter, rather than just erroring when an interpreter is found but doesn't have the necessary dependencies. Helped-by: John Szakmeister <john@szakmeister.net>
2014-07-11 04:12:10 -07:00
if(NOT LUA_PRG_WORKS)
message(FATAL_ERROR "A suitable Lua interpreter was not found")
Add checks for lpeg/cmsgpack lua packages These checks will stop the build process with meaningful error messages if any of those packages are not installed.
2014-04-13 03:06:35 -07:00
endif()
Refactor the lua dependency checking. This is in preparation for the next step, which is to find a suitable lua interpreter, rather than just erroring when an interpreter is found but doesn't have the necessary dependencies. Helped-by: John Szakmeister <john@szakmeister.net>
2014-07-11 04:12:10 -07:00
message(STATUS "Using the Lua interpreter ${LUA_PRG}")
Add checks for lpeg/cmsgpack lua packages These checks will stop the build process with meaningful error messages if any of those packages are not installed.
2014-04-13 03:06:35 -07:00
build: install with the correct permissions The install() command will create the parent directories, but it does so with the user's umask. We want to do our best to make sure the correct permissions are being set, without clobbering existing permissions. To do this, this commit introduces an install_helper(), which is similar in signature to the install() command, to help ensure that directories are created ahead of the actual install() command. This will attempt to use 0644 permissions for files and 0755 permissions for directories by default--though they can be overridden. To make this work correctly, without trying to introduce some mechanism with setting the umask, it meant that there's a small portion that makes use of an "internal" version of the file() command. It has been tested on CMake 2.8.11, 2.8.12, and 3.0.2, and works correctly on all versions. This fixes #1201 and #1086.
2014-09-19 04:37:22 -07:00
# Setup busted.
Revamp the build system. This achieves several goals: * Less reliance on scripts so we have better portability to Windows (though we still have a ways to go for proper Windows support). Luajit, luarocks, moonscript, and busted are all installed via CMake now. * Trying to make use of pkg-config to get the correct libraries. The latest libuv is still broken in this regard, but we'll at least be in a position to use it. * Allow the use of Ninja or make. The former runs faster in many environments, and automatically makes use of parallel builds. This also allows for system installed dependencies--though not through the Makefile just yet--and adds support for FreeBSD. This also make us build libuv and luajit as static libraries only, since we're only concerned about having static libraries for our bundled dependencies.
2014-03-03 08:09:06 -07:00
find_program(BUSTED_PRG busted)
build: install with the correct permissions The install() command will create the parent directories, but it does so with the user's umask. We want to do our best to make sure the correct permissions are being set, without clobbering existing permissions. To do this, this commit introduces an install_helper(), which is similar in signature to the install() command, to help ensure that directories are created ahead of the actual install() command. This will attempt to use 0644 permissions for files and 0755 permissions for directories by default--though they can be overridden. To make this work correctly, without trying to introduce some mechanism with setting the umask, it meant that there's a small portion that makes use of an "internal" version of the file() command. It has been tested on CMake 2.8.11, 2.8.12, and 3.0.2, and works correctly on all versions. This fixes #1201 and #1086.
2014-09-19 04:37:22 -07:00
if(NOT BUSTED_OUTPUT_TYPE)
set(BUSTED_OUTPUT_TYPE "utfTerminal")
endif()
Revamp the build system. This achieves several goals: * Less reliance on scripts so we have better portability to Windows (though we still have a ways to go for proper Windows support). Luajit, luarocks, moonscript, and busted are all installed via CMake now. * Trying to make use of pkg-config to get the correct libraries. The latest libuv is still broken in this regard, but we'll at least be in a position to use it. * Allow the use of Ninja or make. The former runs faster in many environments, and automatically makes use of parallel builds. This also allows for system installed dependencies--though not through the Makefile just yet--and adds support for FreeBSD. This also make us build libuv and luajit as static libraries only, since we're only concerned about having static libraries for our bundled dependencies.
2014-03-03 08:09:06 -07:00
build: install with the correct permissions The install() command will create the parent directories, but it does so with the user's umask. We want to do our best to make sure the correct permissions are being set, without clobbering existing permissions. To do this, this commit introduces an install_helper(), which is similar in signature to the install() command, to help ensure that directories are created ahead of the actual install() command. This will attempt to use 0644 permissions for files and 0755 permissions for directories by default--though they can be overridden. To make this work correctly, without trying to introduce some mechanism with setting the umask, it meant that there's a small portion that makes use of an "internal" version of the file() command. It has been tested on CMake 2.8.11, 2.8.12, and 3.0.2, and works correctly on all versions. This fixes #1201 and #1086.
2014-09-19 04:37:22 -07:00
include(InstallHelpers)
Build: Use GNUInstallDirs, install man pages #2649 For now, only install man pages matching "nvim*.1": we don't want to install xxd.1 as it might conflict with that of a user's Vim installation. closes #1826 Reviewed-by: Florian Walch <florian@fwalch.com> Helped-by: John Szakmeister <john@szakmeister.net>
2015-05-17 11:45:30 -07:00
file(GLOB MANPAGES
RELATIVE ${CMAKE_CURRENT_SOURCE_DIR}
Remove outdated and unused manuals #2891 `nvim-[lang].1`: The non-english manuals are completely outdated and still written in roff, as opposed to mdoc, which is used for `nvim.1`. Given that, they're nearly useless at the moment, and when/if they are updated, they should probably be rewritten from scratch using `nvim.1` as a reference. `xxd*.1`: xxd hasn't been in the source tree for a long time, so the manual is of little use. `nvimtutor*.1`: The vimtutor script hasn't ever shipped with nvim, and the consensus seems to be that it won't, at least in the form of an executable installed alongside `$(PREFIX)/bin/nvim` (see #2700). In `nvim.1`, the argument to the `.Os` macro was removed. This was done because its only purpose was to signify that nvim and nvimtutor were part of the "Neovim" distribution, i.e., one and the same, which isn't applicable anymore because `nvimtutor.1` is being removed. From the `.Os` documentation in `man mdoc`: Os Operating system version for display in the page footer. This is the mandatory third macro of any mdoc file. Its syntax is as follows: .Os [system [version]] The optional system parameter specifies the relevant operating system or environment. It is suggested to leave it unspecified, in which case mandoc(1) uses its -Ios argument or, if that isn't specified either, sysname and release as returned by uname(3). Examples: .Os .Os KTH/CSC/TCS .Os BSD 4.3 See also Dd and Dt. Reviewed-by: Felipe Morales <hel.sheep@gmail.com> Reviewed-by: Florian Walch <florian@fwalch.com> Reviewed-by: Justin M. Keyes <justinkz@gmail.com> [ci skip]
2015-06-24 11:12:51 -07:00
man/nvim.1)
Build: Use GNUInstallDirs, install man pages #2649 For now, only install man pages matching "nvim*.1": we don't want to install xxd.1 as it might conflict with that of a user's Vim installation. closes #1826 Reviewed-by: Florian Walch <florian@fwalch.com> Helped-by: John Szakmeister <john@szakmeister.net>
2015-05-17 11:45:30 -07:00
install_helper(
FILES ${MANPAGES}
DESTINATION ${CMAKE_INSTALL_MANDIR}/man1)
build: install with the correct permissions The install() command will create the parent directories, but it does so with the user's umask. We want to do our best to make sure the correct permissions are being set, without clobbering existing permissions. To do this, this commit introduces an install_helper(), which is similar in signature to the install() command, to help ensure that directories are created ahead of the actual install() command. This will attempt to use 0644 permissions for files and 0755 permissions for directories by default--though they can be overridden. To make this work correctly, without trying to introduce some mechanism with setting the umask, it meant that there's a small portion that makes use of an "internal" version of the file() command. It has been tested on CMake 2.8.11, 2.8.12, and 3.0.2, and works correctly on all versions. This fixes #1201 and #1086.
2014-09-19 04:37:22 -07:00
# Go down the tree.
Revamp the build system. This achieves several goals: * Less reliance on scripts so we have better portability to Windows (though we still have a ways to go for proper Windows support). Luajit, luarocks, moonscript, and busted are all installed via CMake now. * Trying to make use of pkg-config to get the correct libraries. The latest libuv is still broken in this regard, but we'll at least be in a position to use it. * Allow the use of Ninja or make. The former runs faster in many environments, and automatically makes use of parallel builds. This also allows for system installed dependencies--though not through the Makefile just yet--and adds support for FreeBSD. This also make us build libuv and luajit as static libraries only, since we're only concerned about having static libraries for our bundled dependencies.
2014-03-03 08:09:06 -07:00
build: install with the correct permissions The install() command will create the parent directories, but it does so with the user's umask. We want to do our best to make sure the correct permissions are being set, without clobbering existing permissions. To do this, this commit introduces an install_helper(), which is similar in signature to the install() command, to help ensure that directories are created ahead of the actual install() command. This will attempt to use 0644 permissions for files and 0755 permissions for directories by default--though they can be overridden. To make this work correctly, without trying to introduce some mechanism with setting the umask, it meant that there's a small portion that makes use of an "internal" version of the file() command. It has been tested on CMake 2.8.11, 2.8.12, and 3.0.2, and works correctly on all versions. This fixes #1201 and #1086.
2014-09-19 04:37:22 -07:00
add_subdirectory(src/nvim)
version: Add compilation info.
2014-11-09 05:52:15 -07:00
# Read compilation flags from src/nvim,
# used in config subdirectory below.
include(GetCompileFlags)
get_compile_flags(NVIM_VERSION_CFLAGS)
build: install with the correct permissions The install() command will create the parent directories, but it does so with the user's umask. We want to do our best to make sure the correct permissions are being set, without clobbering existing permissions. To do this, this commit introduces an install_helper(), which is similar in signature to the install() command, to help ensure that directories are created ahead of the actual install() command. This will attempt to use 0644 permissions for files and 0755 permissions for directories by default--though they can be overridden. To make this work correctly, without trying to introduce some mechanism with setting the umask, it meant that there's a small portion that makes use of an "internal" version of the file() command. It has been tested on CMake 2.8.11, 2.8.12, and 3.0.2, and works correctly on all versions. This fixes #1201 and #1086.
2014-09-19 04:37:22 -07:00
add_subdirectory(test/includes)
version: Add compilation info.
2014-11-09 05:52:15 -07:00
add_subdirectory(config)
term: use an argument vector for termopen(). Old behaviour: termopen('cmd') would run `&shell &shcf "cmd"`, which caused the functional tests to fail on some systems due to the process not "owning" the terminal. Also, it is inconsistent with jobstart(). Modify termopen() so that &shell is not invoked, but maintain the old behaviour with :terminal. Factor the common code for building the argument vector from jobstart() and modify the functional tests to call termopen() instead of :terminal (fixes #2354). Also: * Add a 'name' option for termopen() so that `:terminal {cmd}` produces a buffer named "term//{cwd}/{cmd}" and termopen() users can customize the name. * Update the documentation. * Add functional tests for `:terminal` sinse its behaviour now differs from termopen(). Add "test/functional/fixtures/shell-test.c" and move "test/functional/job/tty-test.c" there, too. Helped-by: Justin M. Keyes <@justinmk>
2015-04-13 20:53:16 -07:00
add_subdirectory(test/functional/fixtures) # compile pty/shell test programs
cmake: Generate tags and some vim syntax elements
2015-04-02 09:45:34 -07:00
add_subdirectory(runtime)
job: Allow spawning jobs connected to pseudo terminals
2015-02-23 08:34:20 -07:00
build: install with the correct permissions The install() command will create the parent directories, but it does so with the user's umask. We want to do our best to make sure the correct permissions are being set, without clobbering existing permissions. To do this, this commit introduces an install_helper(), which is similar in signature to the install() command, to help ensure that directories are created ahead of the actual install() command. This will attempt to use 0644 permissions for files and 0755 permissions for directories by default--though they can be overridden. To make this work correctly, without trying to introduce some mechanism with setting the umask, it meant that there's a small portion that makes use of an "internal" version of the file() command. It has been tested on CMake 2.8.11, 2.8.12, and 3.0.2, and works correctly on all versions. This fixes #1201 and #1086.
2014-09-19 04:37:22 -07:00
# Setup some test-related bits. We do this after going down the tree because we
# need some of the targets.
Don't require busted. Only provide the unittest target if busted was found. And only build nvim-test if the unittest target exists by excluding nvim-test from all. Note: this means nvim-test won't be built by default, but it will be built when you try to run unittests.
2014-03-22 04:14:55 -07:00
if(BUSTED_PRG)
Generate a lua module to help pass build-related settings. This allows us to avoid hard-coding paths and using environment variables to communicate key information to unit tests, which fits with the overall goal of making sure that folks driving CMake directly can continue to do out-of-tree builds.
2014-04-30 02:10:37 -07:00
get_property(TEST_INCLUDE_DIRS DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}
PROPERTY INCLUDE_DIRECTORIES)
Set the policy to 'OLD' to avoid warning CMP0026 in CMake 3.0.
2014-07-15 14:35:08 -07:00
# Set policy CMP0026 to OLD so we avoid CMake warnings on newer
# versions of cmake.
if(POLICY CMP0026)
cmake_policy(SET CMP0026 OLD)
endif()
Generate a lua module to help pass build-related settings. This allows us to avoid hard-coding paths and using environment variables to communicate key information to unit tests, which fits with the overall goal of making sure that folks driving CMake directly can continue to do out-of-tree builds.
2014-04-30 02:10:37 -07:00
get_target_property(TEST_LIBNVIM_PATH nvim-test LOCATION)
configure_file(
test/config/paths.lua.in
${CMAKE_BINARY_DIR}/test/config/paths.lua)
build: create prereq targets for test dependencies This will help make sure that we build all the right prereqs before manually running tests under QuickBuild. Notice that shell-test has been added as a prereq for the functional tests, since it's a requirement for testing the terminal features.
2015-05-08 03:07:56 -07:00
set(UNITTEST_PREREQS nvim-test unittest-headers)
set(FUNCTIONALTEST_PREREQS nvim tty-test shell-test)
set(BENCHMARK_PREREQS nvim tty-test)
# Useful for automated build systems, if they want to manually run the tests.
add_custom_target(unittest-prereqs
DEPENDS ${UNITTEST_PREREQS})
add_custom_target(functionaltest-prereqs
DEPENDS ${FUNCTIONALTEST_PREREQS})
add_custom_target(benchmark-prereqs
DEPENDS ${BENCHMARK_PREREQS})
Don't require busted. Only provide the unittest target if busted was found. And only build nvim-test if the unittest target exists by excluding nvim-test from all. Note: this means nvim-test won't be built by default, but it will be built when you try to run unittests.
2014-03-22 04:14:55 -07:00
add_custom_target(unittest
COMMAND ${CMAKE_COMMAND}
-DBUSTED_PRG=${BUSTED_PRG}
Fix unit tests Apparently busted 1.11.0 is broken(https://github.com/Olivine-Labs/busted/issues/236) in a way that is causing the unit tests to fail. This pins the version to 1.10.0 and also fixes a wrong variable set when msgpack was added as a dependency
2014-04-14 12:46:16 -07:00
-DLUA_PRG=${LUA_PRG}
Don't require busted. Only provide the unittest target if busted was found. And only build nvim-test if the unittest target exists by excluding nvim-test from all. Note: this means nvim-test won't be built by default, but it will be built when you try to run unittests.
2014-03-22 04:14:55 -07:00
-DWORKING_DIR=${CMAKE_CURRENT_SOURCE_DIR}
-DBUSTED_OUTPUT_TYPE=${BUSTED_OUTPUT_TYPE}
-DTEST_DIR=${CMAKE_CURRENT_SOURCE_DIR}/test
Generate a lua module to help pass build-related settings. This allows us to avoid hard-coding paths and using environment variables to communicate key information to unit tests, which fits with the overall goal of making sure that folks driving CMake directly can continue to do out-of-tree builds.
2014-04-30 02:10:37 -07:00
-DBUILD_DIR=${CMAKE_BINARY_DIR}
test: Replace vroom by lua/busted for functional tests The 'lupa' python package provides a simple way to seamless integrate lua and python code. This commit replaces vroom by a python script that exposes the 'neovim' package to a lua state, and invokes busted to run functional tests. This is a temporary solution that will enable writing functional tests using lua/bused while a lua client library is not available. The reason for dropping vroom is flexibility: Lua/busted has a nice DSL-style syntax while also providing the customization power of a full programming language. Another reason is to use a single framework for unit/functional tests. Two other changes were performed in this commit: - Instead of "gcc-unittest/gcc-ia32", the travis builds for gcc are now identified by "gcc/gcc-32". They will run unit/functional tests for both 64 and 32 bits. - Old integration tests(in src/nvim/testdir) are now ran by the 'oldtest' target
2014-09-29 05:43:52 -07:00
-DTEST_TYPE=unit
build: fix CMAKE_MODULE_PATH usage Fixes #1447. `CMAKE_MODULE_PATH` is meant to be a list of directories, and as such, is not the proper way to launch our scripts. Let's use `${PROJECT_SOURCE_DIR}/cmake` instead. Also, let's not outright set `CMAKE_MODULE_PATH`, but instead append our location to the list.
2014-11-10 17:26:01 -07:00
-P ${PROJECT_SOURCE_DIR}/cmake/RunTests.cmake
build: create prereq targets for test dependencies This will help make sure that we build all the right prereqs before manually running tests under QuickBuild. Notice that shell-test has been added as a prereq for the functional tests, since it's a requirement for testing the terminal features.
2015-05-08 03:07:56 -07:00
DEPENDS ${UNITTEST_PREREQS})
test: Replace vroom by lua/busted for functional tests The 'lupa' python package provides a simple way to seamless integrate lua and python code. This commit replaces vroom by a python script that exposes the 'neovim' package to a lua state, and invokes busted to run functional tests. This is a temporary solution that will enable writing functional tests using lua/bused while a lua client library is not available. The reason for dropping vroom is flexibility: Lua/busted has a nice DSL-style syntax while also providing the customization power of a full programming language. Another reason is to use a single framework for unit/functional tests. Two other changes were performed in this commit: - Instead of "gcc-unittest/gcc-ia32", the travis builds for gcc are now identified by "gcc/gcc-32". They will run unit/functional tests for both 64 and 32 bits. - Old integration tests(in src/nvim/testdir) are now ran by the 'oldtest' target
2014-09-29 05:43:52 -07:00
CMake: Rename target "test" to "functionaltest". Remove build warning: The target name "test" is reserved or not valid for certain CMake features, such as generator expressions, and may result in undefined behavior.
2014-11-05 03:54:20 -07:00
add_custom_target(functionaltest
test: Replace vroom by lua/busted for functional tests The 'lupa' python package provides a simple way to seamless integrate lua and python code. This commit replaces vroom by a python script that exposes the 'neovim' package to a lua state, and invokes busted to run functional tests. This is a temporary solution that will enable writing functional tests using lua/bused while a lua client library is not available. The reason for dropping vroom is flexibility: Lua/busted has a nice DSL-style syntax while also providing the customization power of a full programming language. Another reason is to use a single framework for unit/functional tests. Two other changes were performed in this commit: - Instead of "gcc-unittest/gcc-ia32", the travis builds for gcc are now identified by "gcc/gcc-32". They will run unit/functional tests for both 64 and 32 bits. - Old integration tests(in src/nvim/testdir) are now ran by the 'oldtest' target
2014-09-29 05:43:52 -07:00
COMMAND ${CMAKE_COMMAND}
test: Remove run-functional-tests.py Now that the lua client is available, python/lupa are no longer necessary to run the functional tests. The helper functions previously defined in run-functional-tests.py were adapted to test/functional/helpers.lua.
2014-10-08 08:56:28 -07:00
-DBUSTED_PRG=${BUSTED_PRG}
build: fix running of functional tests directly with CMake If you aren't just building everything into build/, then the functional tests fail because they can't find the nvim executable. Let's pass in the location of the nvim executable, and set NVIM_PRG environment variable accordingly.
2014-11-05 05:26:35 -07:00
-DNVIM_PRG=$<TARGET_FILE:nvim>
test: Replace vroom by lua/busted for functional tests The 'lupa' python package provides a simple way to seamless integrate lua and python code. This commit replaces vroom by a python script that exposes the 'neovim' package to a lua state, and invokes busted to run functional tests. This is a temporary solution that will enable writing functional tests using lua/bused while a lua client library is not available. The reason for dropping vroom is flexibility: Lua/busted has a nice DSL-style syntax while also providing the customization power of a full programming language. Another reason is to use a single framework for unit/functional tests. Two other changes were performed in this commit: - Instead of "gcc-unittest/gcc-ia32", the travis builds for gcc are now identified by "gcc/gcc-32". They will run unit/functional tests for both 64 and 32 bits. - Old integration tests(in src/nvim/testdir) are now ran by the 'oldtest' target
2014-09-29 05:43:52 -07:00
-DWORKING_DIR=${CMAKE_CURRENT_SOURCE_DIR}
-DBUSTED_OUTPUT_TYPE=${BUSTED_OUTPUT_TYPE}
-DTEST_DIR=${CMAKE_CURRENT_SOURCE_DIR}/test
-DBUILD_DIR=${CMAKE_BINARY_DIR}
-DTEST_TYPE=functional
build: fix CMAKE_MODULE_PATH usage Fixes #1447. `CMAKE_MODULE_PATH` is meant to be a list of directories, and as such, is not the proper way to launch our scripts. Let's use `${PROJECT_SOURCE_DIR}/cmake` instead. Also, let's not outright set `CMAKE_MODULE_PATH`, but instead append our location to the list.
2014-11-10 17:26:01 -07:00
-P ${PROJECT_SOURCE_DIR}/cmake/RunTests.cmake
build: create prereq targets for test dependencies This will help make sure that we build all the right prereqs before manually running tests under QuickBuild. Notice that shell-test has been added as a prereq for the functional tests, since it's a requirement for testing the terminal features.
2015-05-08 03:07:56 -07:00
DEPENDS ${FUNCTIONALTEST_PREREQS})
vim-patch:7.4.497 #2295 Problem: With some regexp patterns the NFA engine uses many states and becomes very slow. To the user it looks like Vim freezes. Solution: When the number of states reaches a limit fall back to the old engine. (Christian Brabandt) https://github.com/vim/vim/releases/tag/v7-4-497 Helped-by: David Bürgin <676c7473@gmail.com> Helped-by: Justin M. Keyes <justinkz@gmail.com> Helped-by: Scott Prager <splinterofchaos@gmail.com>
2015-03-28 14:28:37 -07:00
add_custom_target(benchmark
COMMAND ${CMAKE_COMMAND}
-DBUSTED_PRG=${BUSTED_PRG}
-DNVIM_PRG=$<TARGET_FILE:nvim>
-DWORKING_DIR=${CMAKE_CURRENT_SOURCE_DIR}
-DBUSTED_OUTPUT_TYPE=${BUSTED_OUTPUT_TYPE}
-DTEST_DIR=${CMAKE_CURRENT_SOURCE_DIR}/test
-DBUILD_DIR=${CMAKE_BINARY_DIR}
-DTEST_TYPE=benchmark
-P ${PROJECT_SOURCE_DIR}/cmake/RunTests.cmake
build: create prereq targets for test dependencies This will help make sure that we build all the right prereqs before manually running tests under QuickBuild. Notice that shell-test has been added as a prereq for the functional tests, since it's a requirement for testing the terminal features.
2015-05-08 03:07:56 -07:00
DEPENDS ${BENCHMARK_PREREQS})
Don't require busted. Only provide the unittest target if busted was found. And only build nvim-test if the unittest target exists by excluding nvim-test from all. Note: this means nvim-test won't be built by default, but it will be built when you try to run unittests.
2014-03-22 04:14:55 -07:00
endif()
Reference in New Issue Copy Permalink