string.h

Go to the documentation of this file.

// Luanti
// SPDX-License-Identifier: LGPL-2.1-or-later
// Copyright (C) 2010-2013 celeron55, Perttu Ahola <celeron55@gmail.com>
 
#pragma once
 
#include "irrlichttypes_bloated.h"
#include "config.h" // IS_CLIENT_BUILD
#if IS_CLIENT_BUILD
#include "irrString.h"
#endif
#include <cstdlib>
#include <string>
#include <string_view>
#include <cstring>
#include <vector>
#include <limits>
#include <sstream>
#include <iomanip>
#include <cctype>
#include <cwctype>
#include <unordered_map>
#include <optional>
 
class Translations;
 
#define STRINGIFY(x) #x
#define TOSTRING(x) STRINGIFY(x)
 
// Checks whether a value is an ASCII printable character
#define IS_ASCII_PRINTABLE_CHAR(x)   \
        (((unsigned int)(x) >= 0x20) &&  \
        ( (unsigned int)(x) <= 0x7e))
 
// Checks whether a value is in a Unicode private use area
#define IS_PRIVATE_USE_CHAR16(x)    \
        ((wchar_t)(x) >= 0xE000 &&   \
          (wchar_t)(x) <= 0xF8FF)
#define IS_PRIVATE_USE_CHAR32(x)    \
        (((wchar_t)(x) >= 0xF0000 &&  \
          (wchar_t)(x) <= 0xFFFFD) || \
         ((wchar_t)(x) >= 0x100000 && \
          (wchar_t)(x) <= 0x10FFFD))
#if WCHAR_MAX > 0xFFFF
#define IS_PRIVATE_USE_CHAR(x) (IS_PRIVATE_USE_CHAR16(x) || IS_PRIVATE_USE_CHAR32(x))
#else
#define IS_PRIVATE_USE_CHAR(x) IS_PRIVATE_USE_CHAR16(x)
#endif
 
// Checks whether a byte is an inner byte for an utf-8 multibyte sequence
#define IS_UTF8_MULTB_INNER(x)       \
        (((unsigned char)(x) >= 0x80) && \
        ( (unsigned char)(x) <= 0xbf))
 
// Checks whether a byte is a start byte for an utf-8 multibyte sequence
#define IS_UTF8_MULTB_START(x)       \
        (((unsigned char)(x) >= 0xc2) && \
        ( (unsigned char)(x) <= 0xf4))
 
// Given a start byte x for an utf-8 multibyte sequence
// it gives the length of the whole sequence in bytes.
#define UTF8_MULTB_START_LEN(x)            \
        (((unsigned char)(x) < 0xe0) ? 2 :     \
        (((unsigned char)(x) < 0xf0) ? 3 : 4))
 
typedef std::unordered_map<std::string, std::string> StringMap;
 
struct FlagDesc {
        const char *name;
        u32 flag;
};
 
// Try to avoid converting between wide and UTF-8 unless you need to
// input/output stuff via Irrlicht
std::wstring utf8_to_wide(std::string_view input);
std::string wide_to_utf8(std::wstring_view input);
 
void wide_add_codepoint(std::wstring &result, char32_t codepoint);
 
std::string urlencode(std::string_view str);
std::string urldecode(std::string_view str);
 
u32 readFlagString(std::string str, const FlagDesc *flagdesc, u32 *flagmask);
std::string writeFlagString(u32 flags, const FlagDesc *flagdesc, u32 flagmask);
 
size_t mystrlcpy(char *dst, const char *src, size_t size) noexcept;
char *mystrtok_r(char *s, const char *sep, char **lasts) noexcept;
 
u64 read_seed(const char *str);
bool parseColorString(const std::string &value, video::SColor &color, bool quiet,
                unsigned char default_alpha = 0xff);
std::string encodeHexColorString(video::SColor color);
 
inline char my_tolower(char c)
{
        // By design this function cannot handle any Unicode (codepoints don't fit into char),
        // but make sure to pass it through unchanged.
        // tolower() can mangle it if the POSIX locale is not UTF-8.
        if (static_cast<unsigned char>(c) > 0x7f)
                return c;
        // toupper(3): "If the argument c is of type char, it must be cast to unsigned char"
        return tolower(static_cast<unsigned char>(c));
}
 
inline std::string padStringRight(std::string str, size_t len)
{
        if (len > str.size())
                str.insert(str.end(), len - str.size(), ' ');
 
        return str;
}
 
inline std::string_view removeStringEnd(std::string_view str,
                const char *ends[])
{
        const char **p = ends;
 
        for (; *p && (*p)[0] != '\0'; p++) {
                std::string_view end(*p);
                if (str.size() < end.size())
                        continue;
                if (str.compare(str.size() - end.size(), end.size(), end) == 0)
                        return str.substr(0, str.size() - end.size());
        }
 
        return std::string_view();
}
 
 
#define MAKE_VARIANT(_name, _t0, _t1) \
        template <typename T, typename... Args> \
        inline auto _name(_t0 arg1, _t1 arg2, Args&&... args) \
        { \
                return (_name)(std::basic_string_view<T>(arg1), std::basic_string_view<T>(arg2), \
                        std::forward<Args>(args)...); \
        }
 
 
template <typename T>
inline bool str_equal(std::basic_string_view<T> s1,
                std::basic_string_view<T> s2,
                bool case_insensitive = false)
{
        if (!case_insensitive)
                return s1 == s2;
 
        if (s1.size() != s2.size())
                return false;
 
        for (size_t i = 0; i < s1.size(); ++i)
                if (my_tolower(s1[i]) != my_tolower(s2[i]))
                        return false;
 
        return true;
}
 
// For some reason an std::string will not implicitly get converted
// to an std::basic_string_view<char> in the template case above, so we need
// these three wrappers. It works if you take out the template parameters.
// see also <https://stackoverflow.com/questions/68380141/>
MAKE_VARIANT(str_equal, const std::basic_string<T> &, const std::basic_string<T> &)
 
MAKE_VARIANT(str_equal, std::basic_string_view<T>, const std::basic_string<T> &)
 
MAKE_VARIANT(str_equal, const std::basic_string<T> &, std::basic_string_view<T>)
 
 
 
template <typename T>
inline bool str_starts_with(std::basic_string_view<T> str,
                std::basic_string_view<T> prefix,
                bool case_insensitive = false)
{
        if (str.size() < prefix.size())
                return false;
 
        if (!case_insensitive)
                return str.compare(0, prefix.size(), prefix) == 0;
 
        for (size_t i = 0; i < prefix.size(); ++i)
                if (my_tolower(str[i]) != my_tolower(prefix[i]))
                        return false;
        return true;
}
 
// (same conversion issue here)
MAKE_VARIANT(str_starts_with, const std::basic_string<T> &, const std::basic_string<T> &)
 
MAKE_VARIANT(str_starts_with, std::basic_string_view<T>, const std::basic_string<T> &)
 
MAKE_VARIANT(str_starts_with, const std::basic_string<T> &, std::basic_string_view<T>)
 
// (the same but with char pointers, only for the prefix argument)
MAKE_VARIANT(str_starts_with, const std::basic_string<T> &, const T*)
 
MAKE_VARIANT(str_starts_with, std::basic_string_view<T>, const T*)
 
 
 
template <typename T>
inline bool str_ends_with(std::basic_string_view<T> str,
                std::basic_string_view<T> suffix,
                bool case_insensitive = false)
{
        if (str.size() < suffix.size())
                return false;
 
        size_t start = str.size() - suffix.size();
        if (!case_insensitive)
                return str.compare(start, suffix.size(), suffix) == 0;
 
        for (size_t i = 0; i < suffix.size(); ++i)
                if (my_tolower(str[start + i]) != my_tolower(suffix[i]))
                        return false;
        return true;
}
 
// (same conversion issue here)
MAKE_VARIANT(str_ends_with, const std::basic_string<T> &, const std::basic_string<T> &)
 
MAKE_VARIANT(str_ends_with, std::basic_string_view<T>, const std::basic_string<T> &)
 
MAKE_VARIANT(str_ends_with, const std::basic_string<T> &, std::basic_string_view<T>)
 
// (the same but with char pointers, only for the suffix argument)
MAKE_VARIANT(str_ends_with, const std::basic_string<T> &, const T*)
 
MAKE_VARIANT(str_ends_with, std::basic_string_view<T>, const T*)
 
 
#undef MAKE_VARIANT
 
 
template <typename T>
inline std::vector<std::basic_string<T> > str_split(
                const std::basic_string<T> &str,
                T delimiter)
{
        std::vector<std::basic_string<T> > parts;
        std::basic_stringstream<T> sstr(str);
        std::basic_string<T> part;
 
        while (std::getline(sstr, part, delimiter))
                parts.push_back(part);
 
        return parts;
}
 
 
inline std::string lowercase(std::string_view str)
{
        std::string s2;
        s2.resize(str.size());
        for (size_t i = 0; i < str.size(); i++)
                s2[i] = my_tolower(str[i]);
        return s2;
}
 
 
inline bool my_isspace(const char c)
{
        return std::isspace(c);
}
 
inline bool my_isspace(const wchar_t c)
{
        return std::iswspace(c);
}
 
template<typename T>
inline std::basic_string_view<T> trim(const std::basic_string_view<T> &str)
{
        size_t front = 0;
        size_t back = str.size();
 
        while (front < back && my_isspace(str[front]))
                ++front;
 
        while (back > front && my_isspace(str[back - 1]))
                --back;
 
        return str.substr(front, back - front);
}
 
// If input was a temporary string keep it one to make sure patterns like
// trim(func_that_returns_str()) are predictable regarding memory allocation
// and don't lead to UAF. ↓ ↓ ↓
 
template<typename T>
inline std::basic_string<T> trim(std::basic_string<T> &&str)
{
        std::basic_string<T> ret(trim(std::basic_string_view<T>(str)));
        return ret;
}
 
template<typename T>
inline std::basic_string_view<T> trim(const std::basic_string<T> &str)
{
        return trim(std::basic_string_view<T>(str));
}
 
// The above declaration causes ambiguity with char pointers so we have to fix that:
template<typename T>
inline std::basic_string_view<T> trim(const T *str)
{
        return trim(std::basic_string_view<T>(str));
}
 
 
inline bool is_yes(std::string_view str)
{
        std::string s2 = lowercase(trim(str));
 
        return s2 == "y" || s2 == "yes" || s2 == "true" || atoi(s2.c_str()) != 0;
}
 
 
inline s32 mystoi(const std::string &str, s32 min, s32 max)
{
        s32 i = atoi(str.c_str());
 
        if (i < min)
                i = min;
        if (i > max)
                i = max;
 
        return i;
}
 
inline s32 mystoi(const std::string &str)
{
        return atoi(str.c_str());
}
 
inline float mystof(const std::string &str)
{
        return atof(str.c_str());
}
 
#define stoi mystoi
#define stof mystof
 
template <typename T>
inline T from_string(const std::string &str)
{
        std::istringstream tmp(str);
        T t;
        tmp >> t;
        return t;
}
 
inline s64 stoi64(const std::string &str) { return from_string<s64>(str); }
 
inline std::string itos(s32 i) { return std::to_string(i); }
inline std::string i64tos(s64 i) { return std::to_string(i); }
 
inline std::string ftos(float f)
{
        std::ostringstream oss;
        oss << std::setprecision(std::numeric_limits<float>::max_digits10) << f;
        return oss.str();
}
 
 
inline void str_replace(std::string &str, std::string_view pattern,
                std::string_view replacement)
{
        std::string::size_type start = str.find(pattern, 0);
        while (start != str.npos) {
                str.replace(start, pattern.size(), replacement);
                start = str.find(pattern, start + replacement.size());
        }
}
 
inline void str_formspec_escape(std::string &str)
{
        str_replace(str, "\\", "\\\\");
        str_replace(str, "]", "\\]");
        str_replace(str, "[", "\\[");
        str_replace(str, ";", "\\;");
        str_replace(str, ",", "\\,");
        str_replace(str, "$", "\\$");
}
 
void str_replace(std::string &str, char from, char to);
 
 
inline bool string_allowed(std::string_view str, std::string_view allowed_chars)
{
        return str.find_first_not_of(allowed_chars) == str.npos;
}
 
 
inline bool string_allowed_blacklist(std::string_view str,
                std::string_view blacklisted_chars)
{
        return str.find_first_of(blacklisted_chars) == str.npos;
}
 
 
std::string wrap_rows(std::string_view from, unsigned row_len, bool has_color_codes = false);
 
 
template <typename T>
inline std::basic_string<T> unescape_string(const std::basic_string<T> &s)
{
        std::basic_string<T> res;
        res.reserve(s.size());
 
        for (size_t i = 0; i < s.length(); i++) {
                if (s[i] == '\\') {
                        i++;
                        if (i >= s.length())
                                break;
                }
                res += s[i];
        }
 
        return res;
}
 
template <typename T>
std::basic_string<T> unescape_enriched(const std::basic_string<T> &s)
{
        std::basic_string<T> output;
        output.reserve(s.size());
        size_t i = 0;
        while (i < s.length()) {
                if (s[i] == '\x1b') {
                        ++i;
                        if (i == s.length()) continue;
                        if (s[i] == '(') {
                                ++i;
                                while (i < s.length() && s[i] != ')') {
                                        if (s[i] == '\\') {
                                                ++i;
                                        }
                                        ++i;
                                }
                                ++i;
                        } else {
                                ++i;
                        }
                        continue;
                }
                output += s[i];
                ++i;
        }
        return output;
}
 
template <typename T>
std::vector<std::basic_string<T> > split(const std::basic_string<T> &s, T delim)
{
        std::vector<std::basic_string<T> > tokens;
 
        std::basic_string<T> current;
        bool last_was_escape = false;
        for (size_t i = 0; i < s.length(); i++) {
                T si = s[i];
                if (last_was_escape) {
                        current += '\\';
                        current += si;
                        last_was_escape = false;
                } else {
                        if (si == delim) {
                                tokens.push_back(current);
                                current.clear();
                                last_was_escape = false;
                        } else if (si == '\\') {
                                last_was_escape = true;
                        } else {
                                current += si;
                                last_was_escape = false;
                        }
                }
        }
        //push last element
        tokens.push_back(current);
 
        return tokens;
}
 
std::wstring translate_string(std::wstring_view s, Translations *translations);
 
std::wstring translate_string(std::wstring_view s);
 
inline std::wstring unescape_translate(std::wstring_view s)
{
        return unescape_enriched(translate_string(s));
}
 
inline bool is_number(std::string_view to_check)
{
        for (char i : to_check)
                if (!std::isdigit(i))
                        return false;
 
        return !to_check.empty();
}
 
 
inline const char *bool_to_cstr(bool val)
{
        return val ? "true" : "false";
}
 
inline const std::string duration_to_string(int sec)
{
        std::ostringstream ss;
        const char *neg = "";
        if (sec < 0) {
                sec = -sec;
                neg = "-";
        }
        int total_sec = sec;
        int min = sec / 60;
        sec %= 60;
        int hour = min / 60;
        min %= 60;
        int day = hour / 24;
        hour %= 24;
 
        if (day > 0) {
                ss << neg << day << "d";
                if (hour > 0 || min > 0 || sec > 0)
                        ss << " ";
        }
 
        if (hour > 0) {
                ss << neg << hour << "h";
                if (min > 0 || sec > 0)
                        ss << " ";
        }
 
        if (min > 0) {
                ss << neg << min << "min";
                if (sec > 0)
                        ss << " ";
        }
 
        if (sec > 0 || total_sec == 0) {
                ss << neg << sec << "s";
        }
 
        return ss.str();
}
 
inline std::string str_join(const std::vector<std::string> &list,
                std::string_view delimiter)
{
        std::ostringstream oss;
        bool first = true;
        for (const auto &part : list) {
                if (!first)
                        oss << delimiter;
                oss << part;
                first = false;
        }
        return oss.str();
}
 
#if IS_CLIENT_BUILD
inline std::string stringw_to_utf8(const irr::core::stringw &input)
{
        std::wstring_view sv(input.c_str(), input.size());
        return wide_to_utf8(sv);
}
 
inline irr::core::stringw utf8_to_stringw(std::string_view input)
{
        std::wstring str = utf8_to_wide(input);
        return irr::core::stringw(str.c_str(), str.size());
}
#endif
 
std::string sanitizeDirName(std::string_view str, std::string_view optional_prefix);
 
[[nodiscard]]
std::string sanitize_untrusted(std::string_view str, bool keep_escapes = true);
 
void safe_print_string(std::ostream &os, std::string_view str);
 
std::optional<v3f> str_to_v3f(std::string_view str);