Isolate binary and Boost-dependent build components

CMakeLists.txt

@@ -15,12 +15,14 @@ project(kenlm)
 
 option(FORCE_STATIC "Build static executables" OFF)
 option(COMPILE_TESTS "Compile tests" OFF)
+option(BUILD_WITH_BOOST "Build binary tools (filter, builder, etc). Requires Boost" ON)
 option(ENABLE_PYTHON "Build Python bindings" OFF)
 option(BUILD_PYTHON_STANDALONE "Build standalone C++ lib with Python install" OFF)
 # Eigen3 less than 3.1.0 has a race condition: http://eigen.tuxfamily.org/bz/show_bug.cgi?id=466
 find_package(Eigen3 3.1.0 CONFIG)
 include(CMakeDependentOption)
-cmake_dependent_option(ENABLE_INTERPOLATE "Build interpolation program (depends on Eigen3)" ON "EIGEN3_FOUND AND NOT WIN32" OFF)
+cmake_dependent_option(ENABLE_INTERPOLATE "Build interpolation program (depends on Eigen3)" ON "BUILD_WITH_BOOST AND EIGEN3_FOUND AND NOT WIN32" OFF)
+cmake_dependent_option(BUILD_BENCHMARKS "Build benchmarks" ON "BUILD_WITH_BOOST" OFF)
 
 set(KENLM_MAX_ORDER 6 CACHE STRING "Maximum supported ngram order")
 
@@ -32,6 +34,9 @@ if (BUILD_PYTHON_STANDALONE)
   return()
 endif()
 
+if (NOT BUILD_WITH_BOOST AND COMPILE_TEST)
+  message(FATAL_ERROR "Cannot compile tests with BUILD_WITH_BOOST=OFF. Boost required for building tests")
+endif()
 
 if (FORCE_STATIC)
   #presumably overkill, is there a better way?
@@ -93,20 +98,30 @@ endif()
 # And our helper modules
 list(APPEND CMAKE_MODULE_PATH ${PROJECT_SOURCE_DIR}/cmake/modules)
 
-# We need boost
-find_package(Boost 1.41.0 REQUIRED COMPONENTS
-  program_options
-  system
-  thread
-  unit_test_framework
-)
+if (BUILD_WITH_BOOST)
+  set(BOOST_COMPONENTS
+    program_options
+    system
+    thread
+  )
+endif()
+if (BUILD_TESTING OR COMPILE_TESTS)
+  list(APPEND BOOST_COMPONENTS unit_test_framework)
+endif()
+
+if (BUILD_WITH_BOOST OR BUILD_TESTING OR COMPILE_TESTS)
+  find_package(Boost 1.41.0 REQUIRED COMPONENTS
+    ${BOOST_COMPONENTS}
+  )
+endif()
 
 # Define where include files live
 include_directories(${Boost_INCLUDE_DIRS})
 
 set(THREADS_PREFER_PTHREAD_FLAG ON)
 find_package(Threads REQUIRED)
 
+
 # Process subdirectories
 add_subdirectory(util)
 add_subdirectory(lm)
@@ -122,7 +137,12 @@ install(EXPORT kenlmTargets
   DESTINATION share/kenlm/cmake
 )
 
-foreach(SUBDIR IN ITEMS util util/double-conversion util/stream lm lm/builder lm/common lm/filter lm/interpolate)
+set(_headers_to_install lm util util/double-conversion )  
+if (BUILD_WITH_BOOST)
+  list(APPEND _headers_to_install lm/builder lm/filter lm/interpolate lm/common util/stream)
+endif()
+
+foreach(SUBDIR IN ITEMS ${_headers_to_install})
   file(GLOB HEADERS ${CMAKE_CURRENT_LIST_DIR}/${SUBDIR}/*.h ${CMAKE_CURRENT_LIST_DIR}/${SUBDIR}/*.hh)
   install(FILES ${HEADERS} DESTINATION include/kenlm/${SUBDIR} COMPONENT headers)
 endforeach(SUBDIR)

cmake/kenlmConfig.cmake.in

@@ -2,7 +2,9 @@
 
 include(CMakeFindDependencyMacro)
 
-find_dependency(Boost)
+if (@BUILD_WITH_BOOST@)
+  find_dependency(Boost)
+endif()
 find_dependency(Threads)
 
 # Compression libs

lm/CMakeLists.txt

@@ -1,3 +1,4 @@
+
 # Explicitly list the source files for this subdirectory
 #
 # If you add any source files to this subdirectory
@@ -22,34 +23,48 @@ set(KENLM_LM_SOURCE
 	vocab.cc
 )
 
-
 # Group these objects together for later use.
 #
 # Given add_library(foo OBJECT ${my_foo_sources}),
 # refer to these objects as $<TARGET_OBJECTS:foo>
 #
-add_subdirectory(common)
+if (BUILD_WITH_BOOST)
+  # only needed if building binaries
+  add_subdirectory(common)
+endif()
+
 
 add_library(kenlm ${KENLM_LM_SOURCE} ${KENLM_LM_COMMON_SOURCE})
 set_target_properties(kenlm PROPERTIES POSITION_INDEPENDENT_CODE ON)
 target_link_libraries(kenlm PUBLIC kenlm_util Threads::Threads)
 # Since headers are relative to `include/kenlm` at install time, not just `include`
-target_include_directories(kenlm PUBLIC $<INSTALL_INTERFACE:include/kenlm>)
+target_include_directories(kenlm PUBLIC
+  $<INSTALL_INTERFACE:include/kenlm>
+  $<INSTALL_INTERFACE:include>
+)
+
+if (BUILD_WITH_BOOST)
+  target_compile_definitions(kenlm PUBLIC -DBUILD_WITH_BOOST=$<BOOL:${BUILD_WITH_BOOST}>)
+endif()
 
 target_compile_definitions(kenlm PUBLIC -DKENLM_MAX_ORDER=${KENLM_MAX_ORDER})
 
 # This directory has children that need to be processed
-add_subdirectory(builder)
-add_subdirectory(filter)
-add_subdirectory(interpolate)
+if (BUILD_WITH_BOOST)
+  add_subdirectory(builder)
+  add_subdirectory(filter)
+endif()
+add_subdirectory(interpolate) # gated inside by ENABLE_INTERPOLATE
 
 # Explicitly list the executable files to be compiled
 set(EXE_LIST
   query
   fragment
   build_binary
-  kenlm_benchmark
-)
+  )
+if (BUILD_BENCHMARKS)
+  list(APPEND EXE_LIST kenlm_benchmark)
+endif()
 
 set(LM_LIBS kenlm kenlm_util Threads::Threads)
 
@@ -63,10 +78,9 @@ install(
 )
 
 AddExes(EXES ${EXE_LIST}
-        LIBRARIES ${LM_LIBS})
+  LIBRARIES ${LM_LIBS})
 
 if(BUILD_TESTING)
-
   set(KENLM_BOOST_TESTS_LIST left_test partial_test)
   AddTests(TESTS ${KENLM_BOOST_TESTS_LIST}
            LIBRARIES ${LM_LIBS}

util/CMakeLists.txt

@@ -18,7 +18,6 @@ set(KENLM_UTIL_SOURCE
 		integer_to_string.cc
 		mmap.cc
 		murmur_hash.cc
-		parallel_read.cc
 		pool.cc
 		read_compressed.cc
 		scoped.cc
@@ -27,13 +26,21 @@ set(KENLM_UTIL_SOURCE
 		usage.cc
 	)
 
+if (BUILD_WITH_BOOST AND Boost_FOUND)
+  list(APPEND KENLM_UTIL_SOURCE parallel_read.cc)
+endif()
+
 if (WIN32)
   set(KENLM_UTIL_SOURCE ${KENLM_UTIL_SOURCE} getopt.c)
 endif()
 
 # This directory has children that need to be processed
 add_subdirectory(double-conversion)
-add_subdirectory(stream)
+
+if (BUILD_WITH_BOOST)
+  # only required for building tools/binaries
+  add_subdirectory(stream)
+endif()
 
 add_library(kenlm_util ${KENLM_UTIL_DOUBLECONVERSION_SOURCE} ${KENLM_UTIL_STREAM_SOURCE} ${KENLM_UTIL_SOURCE})
 # Since headers are relative to `include/kenlm` at install time, not just `include`
@@ -83,13 +90,19 @@ endif()
 # Group these objects together for later use.
 set_target_properties(kenlm_util PROPERTIES POSITION_INDEPENDENT_CODE ON)
 target_link_libraries(kenlm_util
-  PUBLIC
-  # Boost is required for building binaries and tests
-  "$<BUILD_INTERFACE:${Boost_LIBRARIES}>"
   PRIVATE
   Threads::Threads
   ${RT})
 
+if (BUILD_TESTING OR COMPILE_TESTS OR BUILD_WITH_BOOST)
+  target_link_libraries(
+    kenlm_util
+    PUBLIC
+    # Boost is required for building binaries and tests
+    "$<BUILD_INTERFACE:${Boost_LIBRARIES}>"
+    )
+endif()
+
 install(
   TARGETS kenlm_util
   EXPORT kenlmTargets
@@ -99,7 +112,7 @@ install(
   INCLUDES DESTINATION include
 )
 
-if (NOT WIN32)
+if (NOT WIN32 AND BUILD_BENCHMARKS)
 AddExes(EXES probing_hash_table_benchmark
         LIBRARIES kenlm_util Threads::Threads)
 endif()

util/parallel_read.cc

@@ -2,7 +2,7 @@
 
 #include "file.hh"
 
-#ifdef WITH_THREADS
+#if BUILD_WITH_BOOST
 #include "thread_pool.hh"
 
 namespace util {
@@ -58,12 +58,12 @@ void ParallelRead(int fd, void *to, std::size_t amount, uint64_t offset) {
 
 } // namespace util
 
-#else // WITH_THREADS
+#else // defined(BUILD_WITH_BOOST)
 
 namespace util {
 void ParallelRead(int fd, void *to, std::size_t amount, uint64_t offset) {
  util::ErsatzPRead(fd, to, amount, offset);
 }
 } // namespace util
 
-#endif
+#endif // defined(BUILD_WITH_BOOST)