diff options
Diffstat (limited to 'src/common/string_util.cpp')
| -rw-r--r-- | src/common/string_util.cpp | 204 |
1 files changed, 130 insertions, 74 deletions
diff --git a/src/common/string_util.cpp b/src/common/string_util.cpp index c1f22bda3..54943d306 100644 --- a/src/common/string_util.cpp +++ b/src/common/string_util.cpp @@ -3,34 +3,29 @@ // Refer to the license.txt file included. #include <algorithm> -#include <cstdlib> -#include <cstdio> #include "common/common.h" -#include "common/common_paths.h" #include "common/string_util.h" #ifdef _WIN32 #include <Windows.h> + #include <codecvt> #else - #include <cerrno> #include <iconv.h> #endif +namespace Common { + /// Make a string lowercase -void LowerStr(char* str) { - for (int i = 0; str[i]; i++) { - str[i] = tolower(str[ i ]); - } +std::string ToLower(std::string str) { + std::transform(str.begin(), str.end(), str.begin(), ::tolower); + return str; } /// Make a string uppercase -void UpperStr(char* str) { - for (int i=0; i < strlen(str); i++) { - if(str[i] >= 'a' && str[i] <= 'z') { - str[i] &= 0xDF; - } - } +std::string ToUpper(std::string str) { + std::transform(str.begin(), str.end(), str.begin(), ::toupper); + return str; } // faster than sscanf @@ -192,9 +187,9 @@ bool TryParse(const std::string &str, u32 *const output) bool TryParse(const std::string &str, bool *const output) { - if ("1" == str || !strcasecmp("true", str.c_str())) + if ("1" == str || "true" == ToLower(str)) *output = true; - else if ("0" == str || !strcasecmp("false", str.c_str())) + else if ("0" == str || "false" == ToLower(str)) *output = false; else return false; @@ -202,13 +197,6 @@ bool TryParse(const std::string &str, bool *const output) return true; } -std::string StringFromInt(int value) -{ - char temp[16]; - sprintf(temp, "%i", value); - return temp; -} - std::string StringFromBool(bool value) { return value ? "True" : "False"; @@ -283,12 +271,17 @@ std::string TabsToSpaces(int tab_size, const std::string &in) std::string ReplaceAll(std::string result, const std::string& src, const std::string& dest) { - while(1) + size_t pos = 0; + + if (src == dest) + return result; + + while ((pos = result.find(src, pos)) != std::string::npos) { - size_t pos = result.find(src); - if (pos == std::string::npos) break; result.replace(pos, src.size(), dest); + pos += dest.length(); } + return result; } @@ -419,7 +412,19 @@ std::string UriEncode(const std::string & sSrc) #ifdef _WIN32 -std::string UTF16ToUTF8(const std::wstring& input) +std::string UTF16ToUTF8(const std::u16string& input) +{ + std::wstring_convert<std::codecvt_utf8_utf16<char16_t>, char16_t> convert; + return convert.to_bytes(input); +} + +std::u16string UTF8ToUTF16(const std::string& input) +{ + std::wstring_convert<std::codecvt_utf8_utf16<char16_t>, char16_t> convert; + return convert.from_bytes(input); +} + +static std::string UTF16ToUTF8(const std::wstring& input) { auto const size = WideCharToMultiByte(CP_UTF8, 0, input.data(), input.size(), nullptr, 0, nullptr, nullptr); @@ -432,7 +437,7 @@ std::string UTF16ToUTF8(const std::wstring& input) return output; } -std::wstring CPToUTF16(u32 code_page, const std::string& input) +static std::wstring CPToUTF16(u32 code_page, const std::string& input) { auto const size = MultiByteToWideChar(code_page, 0, input.data(), input.size(), nullptr, 0); @@ -445,7 +450,7 @@ std::wstring CPToUTF16(u32 code_page, const std::string& input) return output; } -std::wstring UTF8ToUTF16(const std::string& input) +std::wstring UTF8ToUTF16W(const std::string &input) { return CPToUTF16(CP_UTF8, input); } @@ -463,61 +468,123 @@ std::string CP1252ToUTF8(const std::string& input) #else template <typename T> -std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>& input) +static std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>& input) { std::string result; iconv_t const conv_desc = iconv_open("UTF-8", fromcode); - if ((iconv_t)-1 == conv_desc) + if ((iconv_t)(-1) == conv_desc) { ERROR_LOG(COMMON, "Iconv initialization failure [%s]: %s", fromcode, strerror(errno)); + iconv_close(conv_desc); + return {}; } - else - { - size_t const in_bytes = sizeof(T) * input.size(); - size_t const out_buffer_size = 4 * in_bytes; - std::string out_buffer; - out_buffer.resize(out_buffer_size); + const size_t in_bytes = sizeof(T) * input.size(); + // Multiply by 4, which is the max number of bytes to encode a codepoint + const size_t out_buffer_size = 4 * in_bytes; - auto src_buffer = &input[0]; - size_t src_bytes = in_bytes; - auto dst_buffer = &out_buffer[0]; - size_t dst_bytes = out_buffer.size(); + std::string out_buffer; + out_buffer.resize(out_buffer_size); - while (src_bytes != 0) - { - size_t const iconv_result = iconv(conv_desc, (char**)(&src_buffer), &src_bytes, - &dst_buffer, &dst_bytes); + auto src_buffer = &input[0]; + size_t src_bytes = in_bytes; + auto dst_buffer = &out_buffer[0]; + size_t dst_bytes = out_buffer.size(); + + while (0 != src_bytes) + { + size_t const iconv_result = iconv(conv_desc, (char**)(&src_buffer), &src_bytes, + &dst_buffer, &dst_bytes); - if ((size_t)-1 == iconv_result) + if (static_cast<size_t>(-1) == iconv_result) + { + if (EILSEQ == errno || EINVAL == errno) { - if (EILSEQ == errno || EINVAL == errno) + // Try to skip the bad character + if (0 != src_bytes) { - // Try to skip the bad character - if (src_bytes != 0) - { - --src_bytes; - ++src_buffer; - } - } - else - { - ERROR_LOG(COMMON, "iconv failure [%s]: %s", fromcode, strerror(errno)); - break; + --src_bytes; + ++src_buffer; } } + else + { + ERROR_LOG(COMMON, "iconv failure [%s]: %s", fromcode, strerror(errno)); + break; + } } + } - out_buffer.resize(out_buffer_size - dst_bytes); - out_buffer.swap(result); - + out_buffer.resize(out_buffer_size - dst_bytes); + out_buffer.swap(result); + + iconv_close(conv_desc); + + return result; +} + +std::u16string UTF8ToUTF16(const std::string& input) +{ + std::u16string result; + + iconv_t const conv_desc = iconv_open("UTF-16", "UTF-8"); + if ((iconv_t)(-1) == conv_desc) + { + ERROR_LOG(COMMON, "Iconv initialization failure [UTF-8]: %s", strerror(errno)); iconv_close(conv_desc); + return {}; + } + + const size_t in_bytes = sizeof(char) * input.size(); + // Multiply by 4, which is the max number of bytes to encode a codepoint + const size_t out_buffer_size = 4 * sizeof(char16_t) * in_bytes; + + std::u16string out_buffer; + out_buffer.resize(out_buffer_size); + + char* src_buffer = const_cast<char*>(&input[0]); + size_t src_bytes = in_bytes; + char* dst_buffer = (char*)(&out_buffer[0]); + size_t dst_bytes = out_buffer.size(); + + while (0 != src_bytes) + { + size_t const iconv_result = iconv(conv_desc, &src_buffer, &src_bytes, + &dst_buffer, &dst_bytes); + + if (static_cast<size_t>(-1) == iconv_result) + { + if (EILSEQ == errno || EINVAL == errno) + { + // Try to skip the bad character + if (0 != src_bytes) + { + --src_bytes; + ++src_buffer; + } + } + else + { + ERROR_LOG(COMMON, "iconv failure [UTF-8]: %s", strerror(errno)); + break; + } + } } + + out_buffer.resize(out_buffer_size - dst_bytes); + out_buffer.swap(result); + + iconv_close(conv_desc); return result; } +std::string UTF16ToUTF8(const std::u16string& input) +{ + return CodeToUTF8("UTF-16", input); +} + std::string CP1252ToUTF8(const std::string& input) { //return CodeToUTF8("CP1252//TRANSLIT", input); @@ -531,17 +598,6 @@ std::string SHIFTJISToUTF8(const std::string& input) return CodeToUTF8("SJIS", input); } -std::string UTF16ToUTF8(const std::wstring& input) -{ - std::string result = - // CodeToUTF8("UCS-2", input); - // CodeToUTF8("UCS-2LE", input); - // CodeToUTF8("UTF-16", input); - CodeToUTF8("UTF-16LE", input); - - // TODO: why is this needed? - result.erase(std::remove(result.begin(), result.end(), 0x00), result.end()); - return result; -} - #endif + +} |