diff --git a/src/common/string_util.cpp b/src/common/string_util.cpp index 3dd73ac7e..88e636805 100644 --- a/src/common/string_util.cpp +++ b/src/common/string_util.cpp @@ -8,9 +8,7 @@ #include "windows_headers.h" #endif -namespace StringUtil { - -std::string StdStringFromFormat(const char* format, ...) +std::string StringUtil::StdStringFromFormat(const char* format, ...) { std::va_list ap; va_start(ap, format); @@ -19,7 +17,7 @@ std::string StdStringFromFormat(const char* format, ...) return ret; } -std::string StdStringFromFormatV(const char* format, std::va_list ap) +std::string StringUtil::StdStringFromFormatV(const char* format, std::va_list ap) { std::va_list ap_copy; va_copy(ap_copy, ap); @@ -32,12 +30,18 @@ std::string StdStringFromFormatV(const char* format, std::va_list ap) va_end(ap_copy); std::string ret; - ret.resize(len); - std::vsnprintf(ret.data(), ret.size() + 1, format, ap); + + // If an encoding error occurs, len is -1. Which we definitely don't want to resize to. + if (len > 0) + { + ret.resize(len); + std::vsnprintf(ret.data(), ret.size() + 1, format, ap); + } + return ret; } -bool WildcardMatch(const char* subject, const char* mask, bool case_sensitive /*= true*/) +bool StringUtil::WildcardMatch(const char* subject, const char* mask, bool case_sensitive /*= true*/) { if (case_sensitive) { @@ -133,7 +137,7 @@ bool WildcardMatch(const char* subject, const char* mask, bool case_sensitive /* } } -std::size_t Strlcpy(char* dst, const char* src, std::size_t size) +std::size_t StringUtil::Strlcpy(char* dst, const char* src, std::size_t size) { std::size_t len = std::strlen(src); if (len < size) @@ -148,7 +152,7 @@ std::size_t Strlcpy(char* dst, const char* src, std::size_t size) return len; } -std::size_t Strlcpy(char* dst, const std::string_view& src, std::size_t size) +std::size_t StringUtil::Strlcpy(char* dst, const std::string_view& src, std::size_t size) { std::size_t len = src.length(); if (len < size) @@ -164,7 +168,7 @@ std::size_t Strlcpy(char* dst, const std::string_view& src, std::size_t size) return len; } -std::optional> DecodeHex(const std::string_view& in) +std::optional> StringUtil::DecodeHex(const std::string_view& in) { std::vector data; data.reserve(in.size() / 2); @@ -181,18 +185,138 @@ std::optional> DecodeHex(const std::string_view& in) return {data}; } -std::string EncodeHex(const u8* data, int length) +std::string StringUtil::EncodeHex(const u8* data, int length) { std::stringstream ss; for (int i = 0; i < length; i++) ss << std::hex << std::setfill('0') << std::setw(2) << static_cast(data[i]); - + return ss.str(); } +std::string_view StringUtil::StripWhitespace(const std::string_view& str) +{ + std::string_view::size_type start = 0; + while (start < str.size() && std::isspace(str[start])) + start++; + if (start == str.size()) + return {}; + + std::string_view::size_type end = str.size() - 1; + while (end > start && std::isspace(str[end])) + end--; + + return str.substr(start, end - start + 1); +} + +void StringUtil::StripWhitespace(std::string* str) +{ + { + const char* cstr = str->c_str(); + std::string_view::size_type start = 0; + while (start < str->size() && std::isspace(cstr[start])) + start++; + if (start != 0) + str->erase(0, start); + } + + { + const char* cstr = str->c_str(); + std::string_view::size_type start = str->size(); + while (start > 0 && std::isspace(cstr[start - 1])) + start--; + if (start != str->size()) + str->erase(start); + } +} + +std::vector StringUtil::SplitString(const std::string_view& str, char delimiter, + bool skip_empty /*= true*/) +{ + std::vector res; + std::string_view::size_type last_pos = 0; + std::string_view::size_type pos; + while (last_pos < str.size() && (pos = str.find(delimiter, last_pos)) != std::string_view::npos) + { + std::string_view part(StripWhitespace(str.substr(last_pos, pos - last_pos))); + if (!skip_empty || !part.empty()) + res.push_back(std::move(part)); + + last_pos = pos + 1; + } + + if (last_pos < str.size()) + { + std::string_view part(StripWhitespace(str.substr(last_pos))); + if (!skip_empty || !part.empty()) + res.push_back(std::move(part)); + } + + return res; +} + +std::string StringUtil::ReplaceAll(const std::string_view& subject, const std::string_view& search, + const std::string_view& replacement) +{ + std::string ret(subject); + ReplaceAll(&ret, search, replacement); + return ret; +} + +void StringUtil::ReplaceAll(std::string* subject, const std::string_view& search, const std::string_view& replacement) +{ + if (!subject->empty()) + { + std::string::size_type start_pos = 0; + while ((start_pos = subject->find(search, start_pos)) != std::string::npos) + { + subject->replace(start_pos, search.length(), replacement); + start_pos += replacement.length(); + } + } +} + +bool StringUtil::ParseAssignmentString(const std::string_view& str, std::string_view* key, std::string_view* value) +{ + const std::string_view::size_type pos = str.find('='); + if (pos == std::string_view::npos) + { + *key = std::string_view(); + *value = std::string_view(); + return false; + } + + *key = StripWhitespace(str.substr(0, pos)); + if (pos != (str.size() - 1)) + *value = StripWhitespace(str.substr(pos + 1)); + else + *value = std::string_view(); + + return true; +} + +void StringUtil::AppendUTF16CharacterToUTF8(std::string& s, u16 ch) +{ + if (ch & 0xf800) + { + s.push_back(static_cast(static_cast(0xe0 | static_cast(ch >> 12)))); + s.push_back(static_cast(static_cast(0x80 | static_cast(((ch >> 6) & 0x3f))))); + s.push_back(static_cast(static_cast(0x80 | static_cast((ch & 0x3f))))); + } + else if (ch & 0xff80) + { + s.push_back(static_cast(static_cast(0xc0 | static_cast((ch >> 6))))); + s.push_back(static_cast(static_cast(0x80 | static_cast((ch & 0x3f))))); + } + else + { + s.push_back(static_cast(static_cast(ch))); + } +} + #ifdef _WIN32 -std::wstring UTF8StringToWideString(const std::string_view& str) +std::wstring StringUtil::UTF8StringToWideString(const std::string_view& str) { std::wstring ret; if (!UTF8StringToWideString(ret, str)) @@ -201,7 +325,7 @@ std::wstring UTF8StringToWideString(const std::string_view& str) return ret; } -bool UTF8StringToWideString(std::wstring& dest, const std::string_view& str) +bool StringUtil::UTF8StringToWideString(std::wstring& dest, const std::string_view& str) { int wlen = MultiByteToWideChar(CP_UTF8, 0, str.data(), static_cast(str.length()), nullptr, 0); if (wlen < 0) @@ -214,7 +338,7 @@ bool UTF8StringToWideString(std::wstring& dest, const std::string_view& str) return true; } -std::string WideStringToUTF8String(const std::wstring_view& str) +std::string StringUtil::WideStringToUTF8String(const std::wstring_view& str) { std::string ret; if (!WideStringToUTF8String(ret, str)) @@ -223,7 +347,7 @@ std::string WideStringToUTF8String(const std::wstring_view& str) return ret; } -bool WideStringToUTF8String(std::string& dest, const std::wstring_view& str) +bool StringUtil::WideStringToUTF8String(std::string& dest, const std::wstring_view& str) { int mblen = WideCharToMultiByte(CP_UTF8, 0, str.data(), static_cast(str.length()), nullptr, 0, nullptr, nullptr); if (mblen < 0) @@ -240,5 +364,3 @@ bool WideStringToUTF8String(std::string& dest, const std::wstring_view& str) } #endif - -} // namespace StringUtil diff --git a/src/common/string_util.h b/src/common/string_util.h index 5d96eca86..0ad086778 100644 --- a/src/common/string_util.h +++ b/src/common/string_util.h @@ -100,13 +100,15 @@ template<> inline std::optional FromChars(const std::string_view& str, int base) { if (Strncasecmp("true", str.data(), str.length()) == 0 || Strncasecmp("yes", str.data(), str.length()) == 0 || - Strncasecmp("on", str.data(), str.length()) == 0 || Strncasecmp("1", str.data(), str.length()) == 0) + Strncasecmp("on", str.data(), str.length()) == 0 || Strncasecmp("1", str.data(), str.length()) == 0 || + Strncasecmp("enabled", str.data(), str.length()) == 0 || Strncasecmp("1", str.data(), str.length()) == 0) { return true; } if (Strncasecmp("false", str.data(), str.length()) == 0 || Strncasecmp("no", str.data(), str.length()) == 0 || - Strncasecmp("off", str.data(), str.length()) == 0 || Strncasecmp("0", str.data(), str.length()) == 0) + Strncasecmp("off", str.data(), str.length()) == 0 || Strncasecmp("0", str.data(), str.length()) == 0 || + Strncasecmp("disabled", str.data(), str.length()) == 0 || Strncasecmp("0", str.data(), str.length()) == 0) { return false; } @@ -119,20 +121,82 @@ std::optional> DecodeHex(const std::string_view& str); std::string EncodeHex(const u8* data, int length); /// starts_with from C++20 -ALWAYS_INLINE static bool StartsWith(const std::string_view& str, const char* prefix) +ALWAYS_INLINE static bool StartsWith(const std::string_view& str, const std::string_view& prefix) { - return (str.compare(0, std::strlen(prefix), prefix) == 0); + return (str.compare(0, prefix.length(), prefix) == 0); } -ALWAYS_INLINE static bool EndsWith(const std::string_view& str, const char* suffix) +ALWAYS_INLINE static bool EndsWith(const std::string_view& str, const std::string_view& suffix) { - const std::size_t suffix_length = std::strlen(suffix); + const std::size_t suffix_length = suffix.length(); return (str.length() >= suffix_length && str.compare(str.length() - suffix_length, suffix_length, suffix) == 0); } +/// StartsWith/EndsWith variants which aren't case sensitive. +ALWAYS_INLINE static bool StartsWithNoCase(const std::string_view& str, const std::string_view& prefix) +{ + return (!str.empty() && Strncasecmp(str.data(), prefix.data(), prefix.length()) == 0); +} +ALWAYS_INLINE static bool EndsWithNoCase(const std::string_view& str, const std::string_view& suffix) +{ + const std::size_t suffix_length = suffix.length(); + return (str.length() >= suffix_length && + Strncasecmp(str.data() + (str.length() - suffix_length), suffix.data(), suffix_length) == 0); +} + +/// Strip whitespace from the start/end of the string. +std::string_view StripWhitespace(const std::string_view& str); +void StripWhitespace(std::string* str); + +/// Splits a string based on a single character delimiter. +std::vector SplitString(const std::string_view& str, char delimiter, bool skip_empty = true); + +/// Joins a string together using the specified delimiter. +template +static inline std::string JoinString(const T& start, const T& end, char delimiter) +{ + std::string ret; + for (auto it = start; it != end; ++it) + { + if (it != start) + ret += delimiter; + ret.append(*it); + } + return ret; +} +template +static inline std::string JoinString(const T& start, const T& end, const std::string_view& delimiter) +{ + std::string ret; + for (auto it = start; it != end; ++it) + { + if (it != start) + ret.append(delimiter); + ret.append(*it); + } + return ret; +} + +/// Replaces all instances of search in subject with replacement. +std::string ReplaceAll(const std::string_view& subject, const std::string_view& search, + const std::string_view& replacement); +void ReplaceAll(std::string* subject, const std::string_view& search, const std::string_view& replacement); + +/// Parses an assignment string (Key = Value) into its two components. +bool ParseAssignmentString(const std::string_view& str, std::string_view* key, std::string_view* value); + +/// Appends a UTF-16/UTF-32 codepoint to a UTF-8 string. +void AppendUTF16CharacterToUTF8(std::string& s, u16 ch); + /// Strided memcpy/memcmp. ALWAYS_INLINE static void StrideMemCpy(void* dst, std::size_t dst_stride, const void* src, std::size_t src_stride, std::size_t copy_size, std::size_t count) { + if (src_stride == dst_stride && src_stride == copy_size) + { + std::memcpy(dst, src, src_stride * count); + return; + } + const u8* src_ptr = static_cast(src); u8* dst_ptr = static_cast(dst); for (std::size_t i = 0; i < count; i++) @@ -146,6 +210,9 @@ ALWAYS_INLINE static void StrideMemCpy(void* dst, std::size_t dst_stride, const ALWAYS_INLINE static int StrideMemCmp(const void* p1, std::size_t p1_stride, const void* p2, std::size_t p2_stride, std::size_t copy_size, std::size_t count) { + if (p1_stride == p2_stride && p1_stride == copy_size) + return std::memcmp(p1, p2, p1_stride * count); + const u8* p1_ptr = static_cast(p1); const u8* p2_ptr = static_cast(p2); for (std::size_t i = 0; i < count; i++)