[R] Remove dependency on gendef for Visual Studio builds (fixes #5608) (#5764)

* [R-package] Remove dependency on gendef for Visual Studio builds (fixes #5608)

* clarify docs

* removed debugging print statement

* Make R CMake install more robust

* Fix doc format; add ToC

* Update build.rst

* Fix AppVeyor

Co-authored-by: Hyunsu Cho <chohyu01@cs.washington.edu>

.gitignore

@@ -93,6 +93,7 @@ metastore_db
 # files from R-package source install
 **/config.status
 R-package/src/Makevars
+*.lib
 
 # Visual Studio Code
 /.vscode/

CMakeLists.txt

@@ -226,11 +226,12 @@ add_dependencies(xgboost runxgboost)
 
 #-- Installing XGBoost
 if (R_LIB)
+  include(cmake/RPackageInstallTargetSetup.cmake)
   set_target_properties(xgboost PROPERTIES PREFIX "")
   if (APPLE)
     set_target_properties(xgboost PROPERTIES SUFFIX ".so")
   endif (APPLE)
-  setup_rpackage_install_target(xgboost ${CMAKE_CURRENT_BINARY_DIR})
+  setup_rpackage_install_target(xgboost "${CMAKE_CURRENT_BINARY_DIR}/R-package-install")
   set(CMAKE_INSTALL_PREFIX "${CMAKE_CURRENT_BINARY_DIR}/dummy_inst")
 endif (R_LIB)
 if (MINGW)

R-package/CMakeLists.txt

@@ -36,3 +36,6 @@ set(LINKED_LIBRARIES_PRIVATE ${LINKED_LIBRARIES_PRIVATE} ${LIBR_CORE_LIBRARY} PA
 if (USE_OPENMP)
   target_link_libraries(xgboost-r PRIVATE OpenMP::OpenMP_CXX)
 endif ()
+
+set(LIBR_HOME "${LIBR_HOME}" PARENT_SCOPE)
+set(LIBR_EXECUTABLE "${LIBR_EXECUTABLE}" PARENT_SCOPE)

R-package/inst/make-r-def.R

@@ -0,0 +1,96 @@
+# [description]
+#     Create a definition file (.def) from a .dll file, using objdump. This
+#     is used by FindLibR.cmake when building the R package with MSVC.
+#
+# [usage]
+#
+#     Rscript make-r-def.R something.dll something.def
+#
+# [references]
+#    * https://www.cs.colorado.edu/~main/cs1300/doc/mingwfaq.html
+
+args <- commandArgs(trailingOnly = TRUE)
+
+IN_DLL_FILE <- args[[1L]]
+OUT_DEF_FILE <- args[[2L]]
+DLL_BASE_NAME <- basename(IN_DLL_FILE)
+
+message(sprintf("Creating '%s' from '%s'", OUT_DEF_FILE, IN_DLL_FILE))
+
+# system() will not raise an R exception if the process called
+# fails. Wrapping it here to get that behavior.
+#
+# system() introduces a lot of overhead, at least on Windows,
+# so trying processx if it is available
+.pipe_shell_command_to_stdout <- function(command, args, out_file) {
+    has_processx <- suppressMessages({
+        suppressWarnings({
+            require("processx")  # nolint
+        })
+    })
+    if (has_processx) {
+        p <- processx::process$new(
+            command = command
+            , args = args
+            , stdout = out_file
+            , windows_verbatim_args = FALSE
+        )
+        invisible(p$wait())
+    } else {
+        message(paste0(
+            "Using system2() to run shell commands. Installing "
+            , "'processx' with install.packages('processx') might "
+            , "make this faster."
+        ))
+        exit_code <- system2(
+            command = command
+            , args = shQuote(args)
+            , stdout = out_file
+        )
+        if (exit_code != 0L) {
+            stop(paste0("Command failed with exit code: ", exit_code))
+        }
+    }
+    return(invisible(NULL))
+}
+
+# use objdump to dump all the symbols
+OBJDUMP_FILE <- "objdump-out.txt"
+.pipe_shell_command_to_stdout(
+    command = "objdump"
+    , args = c("-p", IN_DLL_FILE)
+    , out_file = OBJDUMP_FILE
+)
+
+objdump_results <- readLines(OBJDUMP_FILE)
+result <- file.remove(OBJDUMP_FILE)
+
+# Only one table in the objdump results matters for our purposes,
+# see https://www.cs.colorado.edu/~main/cs1300/doc/mingwfaq.html
+start_index <- which(
+    grepl(
+        pattern = "[Ordinal/Name Pointer] Table"
+        , x = objdump_results
+        , fixed = TRUE
+    )
+)
+empty_lines <- which(objdump_results == "")
+end_of_table <- empty_lines[empty_lines > start_index][1L]
+
+# Read the contents of the table
+exported_symbols <- objdump_results[(start_index + 1L):end_of_table]
+exported_symbols <- gsub("\t", "", exported_symbols)
+exported_symbols <- gsub(".*\\] ", "", exported_symbols)
+exported_symbols <- gsub(" ", "", exported_symbols)
+
+# Write R.def file
+writeLines(
+    text = c(
+        paste0("LIBRARY \"", DLL_BASE_NAME, "\"")
+        , "EXPORTS"
+        , exported_symbols
+    )
+    , con = OUT_DEF_FILE
+    , sep = "\n"
+)
+message(sprintf("Successfully created '%s'", OUT_DEF_FILE))

appveyor.yml

@@ -107,7 +107,7 @@ test_script:
       )
     # MSVC R package: run only the unit tests
     - if /i "%target%" == "rmsvc" (
-        cd build_rmsvc%ver%\R-package &&
+        cd R-package &&
         R.exe -q -e "library(testthat); setwd('tests'); source('testthat.R')"
       )
 

cmake/RPackageInstall.cmake.in

@@ -0,0 +1,34 @@
+# Commands to install the R package as a CMake install target
+
+function(check_call)
+  set(cmd COMMAND)
+  cmake_parse_arguments(
+    PARSE_ARGV 0
+    CALL_ARG "" "" "${cmd}"
+  )
+  string(REPLACE ";" " " commands "${CALL_ARG_COMMAND}")
+  message("Command: ${commands}")
+  execute_process(COMMAND ${CALL_ARG_COMMAND}
+                  OUTPUT_VARIABLE _out
+				  ERROR_VARIABLE _err
+				  RESULT_VARIABLE _res)
+  if(NOT "${_res}" EQUAL "0")
+    message(FATAL_ERROR "out: ${_out}, err: ${_err}, res: ${_res}")
+  endif()
+endfunction()
+
+# Important paths
+set(build_dir "@build_dir@")
+set(LIBR_EXECUTABLE "@LIBR_EXECUTABLE@")
+
+# Back up cmake_install.cmake
+file(WRITE "${build_dir}/R-package/src/Makevars" "all:")
+file(WRITE "${build_dir}/R-package/src/Makevars.win" "all:")
+
+# Install dependencies
+set(XGB_DEPS_SCRIPT
+    "deps = setdiff(c('data.table', 'magrittr', 'stringi'), rownames(installed.packages())); if(length(deps)>0) install.packages(deps, repo = 'https://cloud.r-project.org/')")
+check_call(COMMAND "${LIBR_EXECUTABLE}" -q -e "${XGB_DEPS_SCRIPT}")
+
+# Install the XGBoost R package
+check_call(COMMAND "${LIBR_EXECUTABLE}" CMD INSTALL --no-multiarch --build "${build_dir}/R-package")

cmake/RPackageInstallTargetSetup.cmake

@@ -0,0 +1,16 @@
+# Assembles the R-package files in build_dir;
+# if necessary, installs the main R package dependencies;
+# runs R CMD INSTALL.
+function(setup_rpackage_install_target rlib_target build_dir)
+  configure_file(${PROJECT_SOURCE_DIR}/cmake/RPackageInstall.cmake.in ${PROJECT_BINARY_DIR}/RPackageInstall.cmake @ONLY)
+  install(
+    DIRECTORY "${xgboost_SOURCE_DIR}/R-package"
+    DESTINATION "${build_dir}"
+    REGEX "src/*" EXCLUDE
+    REGEX "R-package/configure" EXCLUDE
+  )
+  install(TARGETS ${rlib_target}
+    LIBRARY DESTINATION "${build_dir}/R-package/src/"
+    RUNTIME DESTINATION "${build_dir}/R-package/src/")
+  install(SCRIPT ${PROJECT_BINARY_DIR}/RPackageInstall.cmake)
+endfunction()

cmake/Utils.cmake

@@ -110,38 +110,6 @@ function(format_gencode_flags flags out)
   set(${out} "${${out}}" PARENT_SCOPE)
 endfunction(format_gencode_flags flags)
 
-# Assembles the R-package files in build_dir;
-# if necessary, installs the main R package dependencies;
-# runs R CMD INSTALL.
-function(setup_rpackage_install_target rlib_target build_dir)
-  # backup cmake_install.cmake
-  install(CODE "file(COPY \"${build_dir}/R-package/cmake_install.cmake\"
-DESTINATION \"${build_dir}/bak\")")
-
-  install(CODE "file(REMOVE_RECURSE \"${build_dir}/R-package\")")
-  install(
-    DIRECTORY "${xgboost_SOURCE_DIR}/R-package"
-    DESTINATION "${build_dir}"
-    REGEX "src/*" EXCLUDE
-    REGEX "R-package/configure" EXCLUDE
-  )
-  install(TARGETS ${rlib_target}
-    LIBRARY DESTINATION "${build_dir}/R-package/src/"
-    RUNTIME DESTINATION "${build_dir}/R-package/src/")
-  install(CODE "file(WRITE \"${build_dir}/R-package/src/Makevars\" \"all:\")")
-  install(CODE "file(WRITE \"${build_dir}/R-package/src/Makevars.win\" \"all:\")")
-  set(XGB_DEPS_SCRIPT
-    "deps = setdiff(c('data.table', 'magrittr', 'stringi'), rownames(installed.packages()));\
-    if(length(deps)>0) install.packages(deps, repo = 'https://cloud.r-project.org/')")
-  install(CODE "execute_process(COMMAND \"${LIBR_EXECUTABLE}\" \"-q\" \"-e\" \"${XGB_DEPS_SCRIPT}\")")
-  install(CODE "execute_process(COMMAND \"${LIBR_EXECUTABLE}\" CMD INSTALL\
-    \"--no-multiarch\" \"--build\" \"${build_dir}/R-package\")")
-
-  # restore cmake_install.cmake
-  install(CODE "file(RENAME \"${build_dir}/bak/cmake_install.cmake\"
- \"${build_dir}/R-package/cmake_install.cmake\")")
-endfunction(setup_rpackage_install_target)
-
 macro(enable_nvtx target)
   find_package(NVTX REQUIRED)
   target_include_directories(${target} PRIVATE "${NVTX_INCLUDE_DIR}")

cmake/modules/FindLibR.cmake

@@ -23,7 +23,7 @@
 
 # Windows users might want to change this to their R version:
 if(NOT R_VERSION)
-  set(R_VERSION "3.4.1")
+  set(R_VERSION "4.0.0")
 endif()
 if(NOT R_ARCH)
   if("${CMAKE_SIZEOF_VOID_P}" STREQUAL "4")
@@ -43,16 +43,26 @@ function(create_rlib_for_msvc)
   if(NOT EXISTS "${LIBR_LIB_DIR}")
     message(FATAL_ERROR "LIBR_LIB_DIR was not set!")
   endif()
-  find_program(GENDEF_EXE gendef)
   find_program(DLLTOOL_EXE dlltool)
-  if(NOT GENDEF_EXE OR NOT DLLTOOL_EXE)
+  if(NOT DLLTOOL_EXE)
-    message(FATAL_ERROR "\nEither gendef.exe or dlltool.exe not found!\
+    message(FATAL_ERROR "\ndlltool.exe not found!\
       \nDo you have Rtools installed with its MinGW's bin/ in PATH?")
-  endif()  
+  endif()
+
   # extract symbols from R.dll into R.def and R.lib import library
-  execute_process(COMMAND ${GENDEF_EXE}
+  get_filename_component(
-    "-" "${LIBR_LIB_DIR}/R.dll"
+    LIBR_RSCRIPT_EXECUTABLE_DIR
-    OUTPUT_FILE "${CMAKE_CURRENT_BINARY_DIR}/R.def")
+    ${LIBR_EXECUTABLE}
+    DIRECTORY
+  )
+  set(LIBR_RSCRIPT_EXECUTABLE "${LIBR_RSCRIPT_EXECUTABLE_DIR}/Rscript")
+
+  execute_process(
+    COMMAND ${LIBR_RSCRIPT_EXECUTABLE}
+    "${CMAKE_CURRENT_BINARY_DIR}/../../R-package/inst/make-r-def.R"
+    "${LIBR_LIB_DIR}/R.dll" "${CMAKE_CURRENT_BINARY_DIR}/R.def"
+  )
+
   execute_process(COMMAND ${DLLTOOL_EXE}
     "--input-def" "${CMAKE_CURRENT_BINARY_DIR}/R.def"
     "--output-lib" "${CMAKE_CURRENT_BINARY_DIR}/R.lib")

doc/build.rst

@@ -49,19 +49,7 @@ Please refer to `Trouble Shooting`_ section first if you have any problem
 during installation. If the instructions do not work for you, please feel free
 to ask questions at `the user forum <https://discuss.xgboost.ai>`_.
 
-**Contents**
+.. contents:: Contents
-
-* `Building the Shared Library`_
-
-  - `Building on Linux Distributions`_
-  - `Building on OSX`_
-  - `Building on Windows`_
-  - `Building with GPU support`_
-
-* `Python Package Installation`_
-* `R Package Installation`_
-* `Trouble Shooting`_
-* `Building the documentation`_
 
 .. _build_shared_lib:
 
@@ -365,12 +353,11 @@ You can install XGBoost from CRAN just like any other R package:
 
    and then run ``install.packages("xgboost")``. Without OpenMP, XGBoost will only use a single CPU core, leading to suboptimal training speed.
 
-Installing the development version
+Installing the development version (Linux / Mac OSX)
-----------------------------------
+----------------------------------------------------
 
 Make sure you have installed git and a recent C++ compiler supporting C++11 (See above
-sections for requirements of building C++ core).  On Windows, Rtools must be installed,
+sections for requirements of building C++ core).
-and its bin directory has to be added to ``PATH`` during the installation.
 
 Due to the use of git-submodules, ``devtools::install_github`` can no longer be used to install the latest version of R package.
 Thus, one has to run git to check out the code first:
@@ -390,6 +377,37 @@ Thus, one has to run git to check out the code first:
 If all fails, try `Building the shared library`_ to see whether a problem is specific to R
 package or not.  Notice that the R package is installed by CMake directly.
 
+Installing the development version with Visual Studio
+-----------------------------------------------------
+
+On Windows, CMake with Visual C++ Build Tools (or Visual Studio) can be used to build the R package.
+
+While not required, this build can be faster if you install the R package ``processx`` with ``install.packages("processx")``.
+
+.. note:: Setting correct PATH environment variable on Windows
+
+  If you are using Windows, make sure to include the right directories in the PATH environment variable.
+
+  * If you are using R 4.x with RTools 4.0:
+    - ``C:\rtools40\usr\bin``
+    - ``C:\rtools40\mingw64\bin``
+
+  * If you are using R 3.x with RTools 3.x:
+
+    - ``C:\Rtools\bin``
+    - ``C:\Rtools\mingw_64\bin``
+
+Open the Command Prompt and navigate to the XGBoost directory, and then run the following commands. Make sure to specify the correct R version.
+
+.. code-block:: bash
+
+  cd C:\path\to\xgboost
+  mkdir build
+  cd build
+  cmake .. -G"Visual Studio 16 2019" -A x64 -DR_LIB=ON -DR_VERSION=4.0.0
+  cmake --build . --target install --config Release
+
+
 .. _r_gpu_support:
 
 Installing R package with GPU support
@@ -409,19 +427,32 @@ On Linux, starting from the XGBoost directory type:
 When default target is used, an R package shared library would be built in the ``build`` area.
 The ``install`` target, in addition, assembles the package files with this shared library under ``build/R-package`` and runs ``R CMD INSTALL``.
 
-On Windows, CMake with Visual C++ Build Tools (or Visual Studio) has to be used to build an R package with GPU support. Rtools must also be installed (perhaps, some other MinGW distributions with ``gendef.exe`` and ``dlltool.exe`` would work, but that was not tested).
+On Windows, CMake with Visual Studio has to be used to build an R package with GPU support. Rtools must also be installed.
+
+.. note:: Setting correct PATH environment variable on Windows
+
+  If you are using Windows, make sure to include the right directories in the PATH environment variable.
+
+  * If you are using R 4.x with RTools 4.0:
+
+    - ``C:\rtools40\usr\bin``
+    - ``C:\rtools40\mingw64\bin``
+  * If you are using R 3.x with RTools 3.x:
+
+    - ``C:\Rtools\bin``
+    - ``C:\Rtools\mingw_64\bin``
+
+Open the Command Prompt and navigate to the XGBoost directory, and then run the following commands. Make sure to specify the correct R version.
 
 .. code-block:: bash
 
+  cd C:\path\to\xgboost
   mkdir build
   cd build
-  cmake .. -G"Visual Studio 14 2015 Win64" -DUSE_CUDA=ON -DR_LIB=ON
+  cmake .. -G"Visual Studio 16 2019" -A x64 -DUSE_CUDA=ON -DR_LIB=ON -DR_VERSION=4.0.0
   cmake --build . --target install --config Release
 
-When ``--target xgboost`` is used, an R package DLL would be built under ``build/Release``.
+If CMake can't find your R during the configuration step, you might provide the location of R to CMake like this: ``-DLIBR_HOME="C:\Program Files\R\R-4.0.0"``.
-The ``--target install``, in addition, assembles the package files with this dll under ``build/R-package`` and runs ``R CMD INSTALL``.
-
-If cmake can't find your R during the configuration step, you might provide the location of its executable to cmake like this: ``-DLIBR_EXECUTABLE="C:/Program Files/R/R-3.4.1/bin/x64/R.exe"``.
 
 If on Windows you get a "permission denied" error when trying to write to ...Program Files/R/... during the package installation, create a ``.Rprofile`` file in your personal home directory (if you don't already have one in there), and add a line to it which specifies the location of your R packages user library, like the following: