Add mbstring functions

builtin-functions/_functions.txt

@@ -732,14 +732,6 @@ function setlocale ($category ::: int, $locale ::: string) ::: string | false;
 
 function iconv ($input_encoding ::: string, $output_encoding ::: string, $input_str ::: string) ::: string | false;
 
-function mb_check_encoding ($str ::: string, $encoding ::: string = "1251") ::: bool;
-function mb_strlen ($str ::: string, $encoding ::: string = "1251") ::: int;
-function mb_strpos ($haystack ::: string, $needle ::: string, $offset ::: int = 0, $encoding ::: string = "1251") ::: int | false;
-function mb_stripos ($haystack ::: string, $needle ::: string, $offset ::: int = 0, $encoding ::: string = "1251") ::: int | false;
-function mb_strtolower ($str ::: string, $encoding ::: string = "1251") ::: string;
-function mb_strtoupper ($str ::: string, $encoding ::: string = "1251") ::: string;
-function mb_substr ($str ::: string, $start ::: int, $length ::: mixed = PHP_INT_MAX, $encoding ::: string = "1251") ::: string;
-
 define('PHP_ROUND_HALF_UP', 123423141);
 define('PHP_ROUND_HALF_DOWN', 123423144);
 define('PHP_ROUND_HALF_EVEN', 123423145);
@@ -1624,3 +1616,63 @@ class DateTimeImmutable implements DateTimeInterface {
 }
 
 function getenv(string $varname = '', bool $local_only = false): mixed;
+
+function mb_check_encoding(array|string $value, ?string $encoding = null): bool;
+function mb_convert_encoding(array|string $string, string $to_encoding, array|string|null $from_encoding = null): array|string|false;
+function mb_strlen(string $string, ?string $encoding = null): int;
+function mb_strpos(string $haystack, string $needle, int $offset = 0, ?string $encoding = null): int|false;
+function mb_stripos(string $haystack, string $needle, int $offset = 0, ?string $encoding = null): int|false;
+function mb_strtolower(string $string, ?string $encoding = null): string;
+function mb_strtoupper(string $string, ?string $encoding = null): string;
+function mb_substr(string $string, int $start, ?int $length = null, ?string $encoding = null): string;
+function mb_chr(int $codepoint, ?string $encoding = null): string|false;
+function mb_convert_case(string $string, int $mode, ?string $encoding = null): string;
+function mb_convert_kana(string $string, string $mode = "KV", ?string $encoding = null): string;
+function mb_convert_variables(string $to_encoding, array|string $from_encoding, mixed &$vars): string|false; // ??? (change variable bytes + kwargs)
+function mb_decode_mimeheader(string $string): string;
+function mb_decode_numericentity(string $string, array $map, ?string $encoding = null): string;
+function mb_detect_encoding(string $string, array|string|null $encodings = null, bool $strict = false): string|false;
+function mb_detect_order(array|string|null $encoding = null): mixed; // return array|bool
+function mb_encode_mimeheader(string $string, ?string $charset = null, ?string $transfer_encoding = null, string $newline = "\r\n", int $indent = 0): string;
+function mb_encode_numericentity(string $string, array $map, ?string $encoding = null, bool $hex = false): string;
+function mb_encoding_aliases(string $encoding): array;
+function mb_ereg_match(string $pattern, string $string, ?string $options = null): bool;
+function mb_ereg_replace_callback(string $pattern, callable $callback, string $string, ?string $options = null): string|false|null;
+function mb_ereg_replace(string $pattern, string $replacement, string $string, ?string $options = null): string|false|null;
+function mb_ereg_search_getpos(): int;
+function mb_ereg_search_getregs(): array|false;
+function mb_ereg_search_init(string $string, ?string $pattern = null, ?string $options = null): bool;
+function mb_ereg_search_pos(?string $pattern = null, ?string $options = null): array|false;
+function mb_ereg_search_regs(?string $pattern = null, ?string $options = null): array|false;
+function mb_ereg_search_setpos(int $offset): bool;
+function mb_ereg_search(?string $pattern = null, ?string $options = null): bool;
+function mb_ereg(string $pattern, string $string, array &$matches = null): bool;
+function mb_eregi_replace(string $pattern, string $replacement, string $string, ?string $options = null): string|false|null;
+function mb_eregi(string $pattern, string $string, array &$matches = null): bool;
+function mb_get_info(string $type = "all"): array|string|int|false;
+function mb_http_input(?string $type = null): array|string|false;
+function mb_http_output(?string $encoding = null): string|false;
+function mb_internal_encoding(?string $encoding = null): string|false;
+function mb_language(?string $language = null): string|false;
+function mb_list_encodings(): array;
+function mb_ord(string $string, ?string $encoding = null): int|false;
+function mb_output_handler(string $string, int $status): string;
+function mb_parse_str(string $string, array &$result): bool;
+function mb_preferred_mime_name(string $encoding): string|false;
+function mb_regex_encoding(?string $encoding = null): string|false;
+function mb_regex_set_options(?string $options = null): string;
+function mb_scrub(string $string, ?string $encoding = null): string;
+function mb_send_mail(string $to, string $subject, string $message, array|string $additional_headers = [], ?string $additional_params = null): bool;
+function mb_split(string $pattern, string $string, int $limit = -1): array|false;
+function mb_str_split(string $string, int $length = 1, ?string $encoding = null): array;
+function mb_strcut(string $string, int $start, ?int $length = null, ?string $encoding = null): string;
+function mb_strimwidth(string $string, int $start, int $width, string $trim_marker = "", ?string $encoding = null): string;
+function mb_stristr(string $haystack, string $needle, bool $before_needle = false, ?string $encoding = null): string|false;
+function mb_strrchr(string $haystack, string $needle, bool $before_needle = false, ?string $encoding = null): string|false;
+function mb_strrichr(string $haystack, string $needle, bool $before_needle = false, ?string $encoding = null): string|false;
+function mb_strripos(string $haystack, string $needle, int $offset = 0, ?string $encoding = null): int|false;
+function mb_strrpos(string $haystack, string $needle, int $offset = 0, string $encoding = null): int|false;
+function mb_strstr(string $haystack, string $needle, bool $before_needle = false, ?string $encoding = null): string|false;
+function mb_strwidth(string $string, ?string $encoding = null): int;
+function mb_substitute_character(string|int|null $substitute_character = null): string|int|false;
+function mb_substr_count(string $haystack, string $needle, ?string $encoding = null): int;

cmake/external-libraries.cmake

@@ -1,5 +1,7 @@
 option(DOWNLOAD_MISSING_LIBRARIES "download and build missing libraries if needed" OFF)
+option(MBFL "download and build libmbfl" OFF)
 cmake_print_variables(DOWNLOAD_MISSING_LIBRARIES)
+cmake_print_variables(MBFL)
 function(handle_missing_library LIB_NAME)
     message(STATUS "------${LIB_NAME}---------")
     if(DOWNLOAD_MISSING_LIBRARIES)
@@ -9,6 +11,33 @@ function(handle_missing_library LIB_NAME)
     endif()
 endfunction()
 
+find_library(KPHP_TIMELIB kphp-timelib)
+if(KPHP_TIMELIB)
+    add_library(kphp-timelib STATIC IMPORTED ${KPHP_TIMELIB})
+else()
+    handle_missing_library("kphp-timelib")
+    FetchContent_Declare(kphp-timelib GIT_REPOSITORY https://github.com/VKCOM/timelib)
+    message(STATUS "---------------------")
+    FetchContent_MakeAvailable(kphp-timelib)
+    include_directories(${kphp-timelib_SOURCE_DIR}/include)
+    add_definitions(-DKPHP_TIMELIB_LIB_DIR="${kphp-timelib_SOURCE_DIR}/objs")
+    add_link_options(-L${kphp-timelib_SOURCE_DIR}/objs)
+endif()
+
+if(MBFL)
+    message(STATUS "MBFL=On, libmbfl will be downloaded and built")
+    add_compile_options(-DMBFL)
+    FetchContent_Declare(libmbfl GIT_REPOSITORY https://github.com/andreylzmw/libmbfl)
+    FetchContent_MakeAvailable(libmbfl)
+    include_directories(${libmbfl_SOURCE_DIR}/include)
+    add_definitions(-DLIBMBFL_LIB_DIR="${libmbfl_SOURCE_DIR}/objs")
+    add_link_options(-L${libmbfl_SOURCE_DIR}/objs)
+endif()
+
+# '-Wno-redundant-move' flag works for C++/ObjC++ but not for C, 
+# so build C libraries above
+add_compile_options(-Wno-redundant-move)
+
 find_package(fmt QUIET)
 if(NOT fmt_FOUND)
     handle_missing_library("fmtlib")
@@ -41,19 +70,6 @@ if(KPHP_TESTS)
     endif()
 endif()
 
-find_library(KPHP_TIMELIB kphp-timelib)
-if(KPHP_TIMELIB)
-    add_library(kphp-timelib STATIC IMPORTED ${KPHP_TIMELIB})
-else()
-    handle_missing_library("kphp-timelib")
-    FetchContent_Declare(kphp-timelib GIT_REPOSITORY https://github.com/VKCOM/timelib)
-    message(STATUS "---------------------")
-    FetchContent_MakeAvailable(kphp-timelib)
-    include_directories(${kphp-timelib_SOURCE_DIR}/include)
-    add_definitions(-DKPHP_TIMELIB_LIB_DIR="${kphp-timelib_SOURCE_DIR}/objs")
-    add_link_options(-L${kphp-timelib_SOURCE_DIR}/objs)
-endif()
-
 if(APPLE)
     if (DEFINED ENV{EPOLL_SHIM_REPO})
         FetchContent_Declare(

cmake/init-compilation-flags.cmake

@@ -76,6 +76,7 @@ if (PDO_DRIVER_PGSQL)
     add_definitions(-DPDO_DRIVER_PGSQL)
     add_compile_definitions(PDO_DRIVER_PGSQL_VERSION=${PostgreSQL_VERSION})
 endif()
+
 cmake_print_variables(PDO_DRIVER_PGSQL)
 
 option(KPHP_TESTS "Build the tests" ON)
@@ -118,7 +119,7 @@ elseif(CMAKE_SYSTEM_PROCESSOR STREQUAL "aarch64")
 endif()
 
 add_compile_options(-Werror -Wall -Wextra -Wunused-function -Wfloat-conversion -Wno-sign-compare
-                    -Wuninitialized -Wno-redundant-move -Wno-missing-field-initializers)
+                    -Wuninitialized -Wno-missing-field-initializers)
 
 if(NOT APPLE)
     check_cxx_compiler_flag(-gz=zlib DEBUG_COMPRESSION_IS_FOUND)

compiler/compiler-settings.cpp

@@ -283,6 +283,9 @@ void CompilerSettings::init() {
 
   remove_extra_spaces(extra_cxx_flags.value_);
   std::stringstream ss;
+  #ifdef MBFL
+  ss << " -DMBFL ";
+  #endif
   ss << extra_cxx_flags.get();
   ss << " -iquote" << kphp_src_path.get()
      << " -iquote " << kphp_src_path.get() << "objs/generated/auto/runtime";
@@ -331,6 +334,11 @@ void CompilerSettings::init() {
   ld_flags.value_ += " -L /usr/local/lib";
 #endif
 
+#ifdef LIBMBFL_LIB_DIR
+  external_static_libs.emplace_back("libmbfl");
+  ld_flags.value_ += " -L" LIBMBFL_LIB_DIR;
+#endif
+
 #if defined(__APPLE__) && defined(__arm64__)
   // for development under M1, manual installation of libucontext is needed
   // see the docs: https://vkcom.github.io/kphp/kphp-internals/developing-and-extending-kphp/compiling-kphp-from-sources.html

runtime/interface.cpp

@@ -2379,8 +2379,10 @@ static void free_runtime_libs() {
   free_kphp_backtrace();
 
   free_migration_php8();
-  free_detect_incorrect_encoding_names();
 
+  #ifndef MBFL
+  free_detect_incorrect_encoding_names();
+  #endif
   vk::singleton<JsonLogger>::get().reset_buffers();
 #ifdef PDO_DRIVER_MYSQL
   database_drivers::free_mysql_lib();