# ---- CMake Policies ----

cmake_minimum_required(VERSION 3.26)

if(POLICY CMP0167)
    # deprecated by definition, but we use old behaviour for find_package(Boost) to load CMake's FindBoost module
    # TODO: port project to "Modern CMake" https://gist.github.com/mbinna/c61dbb39bca0e4fb7d1f73b0d66a4fd1
    cmake_policy(SET CMP0167 OLD)
endif()
if(POLICY CMP0144)
    cmake_policy(SET CMP0144 NEW)
endif()

if(POLICY CMP0175)
    # Don't allow invalid arguments of `add_custom_command()`
    cmake_policy(SET CMP0175 NEW)
endif()

# ---- Project ----

project(
    Desbordante
    LANGUAGES CXX
    DESCRIPTION "Open-source data profiling tool"
)

# ---- Build Environment Validation ----

# In-source build validation
if(PROJECT_SOURCE_DIR STREQUAL PROJECT_BINARY_DIR)
    message(
        FATAL_ERROR
            "In-source builds not allowed. Please make a new directory (called a build directory) and run CMake from there."
    )
endif()

# Untested compiler notification
if(NOT CMAKE_CXX_COMPILER_ID MATCHES "^(GNU|AppleClang|Clang)$")
    message(WARNING "This project has primarily been tested with GCC, LLVM Clang, and Apple Clang."
                    "Other compilers may not be fully supported."
    )
endif()

# --- Build Configuration Options ---

option(DESBORDANTE_BUILD_BENCHMARKS "Build benchmarks" OFF)
option(DESBORDANTE_BUILD_NATIVE "Build for host machine" ON)
option(DESBORDANTE_BUILD_TESTS "Build tests" ON)
# This option only takes effect if DESBORDANTE_BUILD_TESTS or DESBORDANTE_BUILD_BENCHMARKS is ON
option(DESBORDANTE_FETCH_DATASETS "Fetch datasets for tests or benchmarks" ON)
option(DESBORDANTE_GDB_SYMBOLS "Include debug information for use by GDB" OFF)
# ---- Vertical Hashing Optimization ----
# Performance/functionality trade-off:
# - Disabled by default for maximum performance
# - Enabled for wide datasets (>32 columns) with potential performance penalty
option(DESBORDANTE_SAFE_VERTICAL_HASHING
        "Enable safe vertical hashing mode. Allows processing wide datasets (>32 columns) \
    at the cost of reduced performance. Recommended for exploratory data analysis." OFF
)
option(DESBORDANTE_UNPACK_DATASETS "Unpack datasets" ON)
option(DESBORDANTE_USE_LTO "Build using interprocedural optimization" OFF)

set(DESBORDANTE_SANITIZER
    ""
    CACHE STRING "Build with sanitizer, possible values: ADDRESS, UB"
)
set(DESBORDANTE_LOG_LEVEL
    ""
    CACHE STRING "Set log level, possible values: TRACE, DEBUG, INFO, WARN, ERROR, CRITICAL"
)

set(DESBORDANTE_BINDINGS
    OFF
    CACHE STRING "Build Python bindings"
)
set_property(CACHE DESBORDANTE_BINDINGS PROPERTY STRINGS OFF BUILD INSTALL)

# By default select Debug build
if(NOT CMAKE_BUILD_TYPE)
    set(CMAKE_BUILD_TYPE Debug)
endif()

if(DESBORDANTE_SANITIZER)
    if(DESBORDANTE_SANITIZER STREQUAL "ADDRESS")
        set(ASAN ON)
    elseif(DESBORDANTE_SANITIZER STREQUAL "UB")
        set(UBSAN ON)
    else()
        message(FATAL_ERROR "Unknown sanitizer '${DESBORDANTE_SANITIZER}', try cmake -LH")
    endif()
endif()

if(ASAN AND DESBORDANTE_BINDINGS)
    message(
        WARNING
            "ASan runtime must be linked with Python or manually preloaded with LD_PRELOAD=/usr/lib/libasan.so"
    )
endif()

# --- Setting CMake build variables ---

# Enable colored Ninja and compilers diagnostics
set(CMAKE_COLOR_DIAGNOSTICS ON)

set(CMAKE_CXX_STANDARD 20)
set(CMAKE_RUNTIME_OUTPUT_DIRECTORY "${CMAKE_BINARY_DIR}/target")

if(DESBORDANTE_BUILD_NATIVE)
    string(JOIN ";" BUILD_OPTS "${BUILD_OPTS}" "-march=native")
endif()

# RELEASE build options
string(JOIN ";" RELEASE_BUILD_OPTS "${BUILD_OPTS}" "-O3")

set(DEBUG_BUILD_OPTS "${BUILD_OPTS}")

# Set common DEBUG build options
string(
    JOIN
    ";"
    DEBUG_BUILD_OPTS
    "${DEBUG_BUILD_OPTS}"
    "-g"
    "-Wall"
    "-Wextra"
    "-Werror"
    "-fno-omit-frame-pointer"
    "-fno-optimize-sibling-calls"
)

# Workaround for Clang bug in versions 18 and above:
# See issue https://github.com/llvm/llvm-project/issues/76515
if(CMAKE_CXX_COMPILER_ID STREQUAL "Clang" AND CMAKE_CXX_COMPILER_VERSION VERSION_GREATER_EQUAL "18")
    message(WARNING "C++ compiler is Clang++-${CMAKE_CXX_COMPILER_VERSION}. "
                    "Suppressing deprecated declaration warnings. "
                    "Consider using another version of Clang."
    )
    list(APPEND DEBUG_BUILD_OPTS "-Wno-deprecated-declarations")
endif()

if(ASAN)
    # Set DEBUG build options specific for build with ASAN
    set(ASAN_OPTS "-fsanitize=address")

    execute_process(COMMAND grep -q "Ubuntu" /etc/os-release RESULT_VARIABLE IS_UBUNTU)
    if(IS_UBUNTU EQUAL 0 AND CMAKE_CXX_FLAGS MATCHES "-stdlib=libc\\+\\+")
        # alloc-dealloc-mismatch generates false positives on boost exceptions
        # This applies only to Ubuntu package:
        # https://github.com/llvm/llvm-project/issues/59432
        message(
            WARNING
                "Running on Ubuntu with libc++. ASAN is broken in Ubuntu package with libc++15-19. If you are using \
                these versions of libc++, set ASAN_OPTIONS=alloc_dealloc_mismatch=0 to suppress alloc-dealloc-mismatch \
                check or consider using another distro or higher libc++ version for full ASAN coverage"
        )
    endif()

    string(
        JOIN
        ";"
        DEBUG_BUILD_OPTS
        "${DEBUG_BUILD_OPTS}"
        "-O1"
        "-Wno-error" # Use of -Werror is discouraged with sanitizers
        # See https://gcc.gnu.org/onlinedocs/gcc/Instrumentation-Options.html#index-fsanitize_003dbuiltin
        "${ASAN_OPTS}"
    )
    set(DEBUG_LINK_OPTS "${ASAN_OPTS}")
elseif(UBSAN)
    # Set DEBUG build options specific for build with UBSAN
    string(
        JOIN
        ";"
        UBSAN_OPTS
        "-fsanitize=undefined"
        "-fsanitize=float-divide-by-zero"
        "-Wno-error" # Use of -Werror is discouraged with sanitizers
        # See https://gcc.gnu.org/onlinedocs/gcc/Instrumentation-Options.html#index-fsanitize_003dbuiltin
        "-fno-sanitize=signed-integer-overflow" # Remove this when CustomRandom gets fixed
        "-fno-sanitize=shift" # Remove this when CustomRandom gets fixed
        "-fno-sanitize-recover=all"
    ) # For tests to fail if UBSan finds an error
    if(CMAKE_CXX_COMPILER_ID STREQUAL "Clang" AND CMAKE_HOST_APPLE)
        # Limit some UB sanitizer checks to "src" directory on macOS when building with Clang,
        # because libraries (STL, googletest, boost, etc.) are somehow broken
        string(JOIN ";" UBSAN_OPTS ${UBSAN_OPTS}
               "-fsanitize-ignorelist=${CMAKE_SOURCE_DIR}/ub_sanitizer_ignore_list.txt"
        )
    endif()
    string(JOIN ";" DEBUG_BUILD_OPTS "${DEBUG_BUILD_OPTS}" "-O1" "${UBSAN_OPTS}")
    set(DEBUG_LINK_OPTS "${UBSAN_OPTS}")
else()
    # No sanitizer, just debug build
    string(JOIN ";" DEBUG_BUILD_OPTS "${DEBUG_BUILD_OPTS}" "-O0")
endif()

if(DESBORDANTE_GDB_SYMBOLS)
    add_compile_options(-ggdb3)
endif()

add_compile_options("$<$<CONFIG:Debug>:${DEBUG_BUILD_OPTS}>")
add_link_options("$<$<CONFIG:Debug>:${DEBUG_LINK_OPTS}>")

add_compile_options("$<$<CONFIG:Release>:${RELEASE_BUILD_OPTS}>")

if(DESBORDANTE_USE_LTO)
    set(CMAKE_INTERPROCEDURAL_OPTIMIZATION TRUE)
endif()

if(DESBORDANTE_BINDINGS)
    set(CMAKE_POSITION_INDEPENDENT_CODE ON)
endif()

# --- Dependencies installation ---

include(cmake/desbordante_deps.cmake)

# --- Core Project Structure and Module Setup ---

include_directories("src")

add_subdirectory("src/core")

# --- Optional Modules Setup ---

# Configure tests
if(DESBORDANTE_BUILD_TESTS)
    enable_testing()
endif()

if (DESBORDANTE_BUILD_TESTS OR DESBORDANTE_BUILD_BENCHMARKS)
    # Configure the datasets required for tests and benchmarks (downloaded from desbordante-data)
    add_subdirectory("datasets")
    add_subdirectory("src/tests")
endif()

# Configure Python bindings
if(DESBORDANTE_BINDINGS)
    add_subdirectory("src/python_bindings")
endif()