core/src/ArrayString.cpp

#include "core/utils/ArrayString.hpp"

#include "core/utils/Error.hpp"

using CharString = Core::CharString;
using Char32String = Core::Char32String;
using Error = Core::Error;
namespace ErrorCode = Core::ErrorCode;

template<typename T>
constexpr int stringLength(const T* c) {
    const T* i = c + 1;
    while(*(c++) != '\0') {}
    return static_cast<int>(c - i);
}

static c32 read(const char*& s) {
    if(*s == '\0') {
        return 0;
    }
    return static_cast<c32>(*(s++));
}

static Error readUnicode(c32& u, const char*& s) {
    u = read(s);
    if((u & 0x80) == 0) {
        return ErrorCode::NONE;
    }
    if((u & 0xE0) == 0xC0) {
        c32 u2 = read(s);
        if(u2 == 0) {
            return ErrorCode::INVALID_CHAR;
        }
        u = ((u & 0x1F) << 6) | (u2 & 0x3F);
        return ErrorCode::NONE;
    } else if((u & 0xF0) == 0xE0) {
        c32 u2 = read(s);
        c32 u3 = read(s);
        if(u2 == 0 || u3 == 0) {
            return ErrorCode::INVALID_CHAR;
        }
        u = ((u & 0xF) << 12) | ((u2 & 0x3F) << 6) | (u3 & 0x3F);
        return ErrorCode::NONE;
    } else if((u & 0xF8) == 0xF0) {
        c32 u2 = read(s);
        c32 u3 = read(s);
        c32 u4 = read(s);
        if(u2 == 0 || u3 == 0 || u4 == 0) {
            return ErrorCode::INVALID_CHAR;
        }
        u = ((u & 0x07) << 18) | ((u2 & 0x3F) << 12) | ((u3 & 0x3F) << 6) |
            (u4 & 0x3F);
        return ErrorCode::NONE;
    }
    return ErrorCode::INVALID_CHAR;
}

template<unsigned int L>
static void unicodeToChar(c32 c, char (&buffer)[L]) {
    static_assert(L >= 5, "to small char buffer");
    buffer[0] = '\0';
    if(c < (1 << 7)) {
        buffer[0] = static_cast<char>(((c >> 0) & 0x7F) | 0x0);
        buffer[1] = '\0';
    } else if(c < (1 << 11)) {
        buffer[0] = static_cast<char>(((c >> 6) & 0x1F) | 0xC0);
        buffer[1] = static_cast<char>(((c >> 0) & 0x3F) | 0x80);
        buffer[2] = '\0';
    } else if(c < (1 << 16)) {
        buffer[0] = static_cast<char>(((c >> 12) & 0x0F) | 0xE0);
        buffer[1] = static_cast<char>(((c >> 6) & 0x3F) | 0x80);
        buffer[2] = static_cast<char>(((c >> 0) & 0x3F) | 0x80);
        buffer[3] = '\0';
    } else if(c < (1 << 21)) {
        buffer[0] = static_cast<char>(((c >> 18) & 0x07) | 0xF0);
        buffer[1] = static_cast<char>(((c >> 12) & 0x3F) | 0x80);
        buffer[2] = static_cast<char>(((c >> 6) & 0x3F) | 0x80);
        buffer[3] = static_cast<char>(((c >> 0) & 0x3F) | 0x80);
        buffer[4] = '\0';
    }
}

static Error printChar(c32 u, u32 shift, u32 a, u32 o) {
    return Core::putChar(static_cast<int>(((u >> shift) & a) | o));
}

CharString::CharString(char* buffer, i32 bufferSize)
    : length(0), capacity(bufferSize), hash(0), data(buffer) {
    data[0] = '\0';
}

Error CharString::copyFrom(const CharString& s) {
    clear();
    return s.toString(*this);
}

bool CharString::operator==(const char* s) const {
    const char* p = data;
    while(*s == *p && *s != '\0') {
        s++;
        p++;
    }
    return *s == *p;
}

bool CharString::operator==(const CharString& other) const {
    if(length != other.getLength()) {
        return false;
    }
    for(int i = 0; i < length; i++) {
        if(data[i] != other[i]) {
            return false;
        }
    }
    return true;
}

bool CharString::operator!=(const char* s) const {
    return !((*this) == s);
}

bool CharString::operator!=(const CharString& other) const {
    return !((*this) == other);
}

char CharString::operator[](int index) const {
    return data[index];
}

int CharString::getLength() const {
    return length;
}

int CharString::getCapacity() const {
    return capacity - 1;
}

Error CharString::append(char c) {
    if(length >= capacity - 1) {
        return ErrorCode::CAPACITY_REACHED;
    }
    data[length++] = c;
    data[length] = '\0';
    addToHash(static_cast<c32>(c));
    return ErrorCode::NONE;
}

Error CharString::append(signed char c) {
    return append(static_cast<char>(c));
}

Error CharString::append(unsigned char c) {
    return append(static_cast<char>(c));
}

Error CharString::append(wchar_t c) {
    return append(static_cast<c32>(c));
}

Error CharString::append(c32 c) {
    char buffer[5];
    unicodeToChar(c, buffer);
    return append(static_cast<const char*>(buffer));
}

Error CharString::append(const char* s) {
    // stringLength as s could be some part of data
    for(int i = stringLength(s); i > 0; i--) {
        CORE_RETURN_ERROR(append(*(s++)));
    }
    return ErrorCode::NONE;
}

Error CharString::append(const c32* s) {
    // stringLength as s could be some part of data
    for(int i = stringLength(s); i > 0; i--) {
        CORE_RETURN_ERROR(append(*(s++)));
    }
    return ErrorCode::NONE;
}

Error CharString::append(const signed char* s) {
    return append(reinterpret_cast<const char*>(s));
}

Error CharString::append(const unsigned char* s) {
    return append(reinterpret_cast<const char*>(s));
}

Error CharString::append(bool b) {
    return b ? append("true") : append("false");
}

Error CharString::toString(CharString& s) const {
    int l = length; // length changes if &s == this
    for(int i = 0; i < l; i++) {
        CORE_RETURN_ERROR(s.append(data[i]));
    }
    return ErrorCode::NONE;
}

Error CharString::toString(Char32String& s) const {
    return s.append(static_cast<const char*>(data));
}

void CharString::clear() {
    length = 0;
    hash = 0;
    data[0] = '\0';
}

u32 CharString::hashCode() const {
    return hash;
}

Error CharString::print() const {
    for(int i = 0; i < length; i++) {
        CORE_RETURN_ERROR(Core::putChar(data[i]));
    }
    return ErrorCode::NONE;
}

Error CharString::printLine() const {
    CORE_RETURN_ERROR(print());
    CORE_RETURN_ERROR(Core::putChar('\n'));
    return ErrorCode::NONE;
}

bool CharString::startsWidth(const CharString& other, int from) const {
    if(from > length - other.getLength()) {
        return false;
    }
    for(int i = 0; i < other.getLength(); i++) {
        if(data[from + i] != other[i]) {
            return false;
        }
    }
    return true;
}

int CharString::search(const CharString& other, int from) const {
    for(int i = from; i < length; i++) {
        if(startsWidth(other, i)) {
            return i;
        }
    }
    return -1;
}

bool CharString::contains(const CharString& other, int from) const {
    return search(other, from) >= 0;
}

int CharString::search(char u, int from) const {
    for(int i = from; i < length; i++) {
        if(data[i] == u) {
            return i;
        }
    }
    return -1;
}

bool CharString::contains(char u, int from) const {
    return search(u, from) >= 0;
}

Error CharString::substring(CharString& s, int from, int to) const {
    s.clear();
    from = Math::max(from, 0);
    to = Math::min(to, length - 1);
    for(int i = from; i <= to; i++) {
        CORE_RETURN_ERROR(s.append(data[i]));
    }
    return ErrorCode::NONE;
}

Error CharString::substring(CharString& s, int from) const {
    return substring(s, from, length - 1);
}

Error CharString::replace(CharString& s, const CharString& search,
                          const CharString& replace) {
    int i = 0;
    while(i < length) {
        if(startsWidth(search, i)) {
            CORE_RETURN_ERROR(s.append(replace));
            i += search.getLength();
        } else {
            CORE_RETURN_ERROR(s.append(data[i]));
            i++;
        }
    }
    return copyFrom(s);
}

void CharString::replace(char search, char replace) {
    hash = 0;
    for(int i = 0; i < length; i++) {
        if(data[i] == search) {
            data[i] = replace;
        }
        addToHash(static_cast<c32>(data[i]));
    }
}

CharString::operator const char*() const {
    return data;
}

void CharString::addToHash(c32 u) {
    hash = static_cast<u32>(2120251889) * hash + static_cast<u32>(u);
}

Char32String::Char32String(c32* buffer, i32 bufferSize)
    : length(0), capacity(bufferSize), hash(0), data(buffer) {
    data[0] = '\0';
}

Error Char32String::copyFrom(const Char32String& s) {
    clear();
    return s.toString(*this);
}

bool Char32String::operator==(const c32* s) const {
    const c32* p = data;
    while(*s == *p && *s != '\0') {
        s++;
        p++;
    }
    return *s == *p;
}

bool Char32String::operator==(const Char32String& other) const {
    if(length != other.getLength()) {
        return false;
    }
    for(int i = 0; i < length; i++) {
        if(data[i] != other[i]) {
            return false;
        }
    }
    return true;
}

bool Char32String::operator!=(const c32* s) const {
    return !((*this) == s);
}

bool Char32String::operator!=(const Char32String& other) const {
    return !((*this) == other);
}

c32 Char32String::operator[](int index) const {
    return data[index];
}

int Char32String::getLength() const {
    return length;
}

int Char32String::getCapacity() const {
    return capacity - 1;
}

Error Char32String::append(char c) {
    return append(static_cast<c32>(c));
}

Error Char32String::append(signed char c) {
    return append(static_cast<char>(c));
}

Error Char32String::append(unsigned char c) {
    return append(static_cast<char>(c));
}

Error Char32String::append(wchar_t c) {
    return append(static_cast<c32>(c));
}

Error Char32String::append(c32 c) {
    if(length >= capacity - 1) {
        return ErrorCode::CAPACITY_REACHED;
    }
    data[length++] = c;
    data[length] = '\0';
    addToHash(static_cast<c32>(c));
    return ErrorCode::NONE;
}

Error Char32String::append(const char* s) {
    while(true) {
        c32 u = 0;
        CORE_RETURN_ERROR(readUnicode(u, s));
        if(u == 0) {
            return ErrorCode::NONE;
        }
        CORE_RETURN_ERROR(append(u));
    }
}

Error Char32String::append(const c32* s) {
    // stringLength as s could be some part of data
    for(int i = stringLength(s); i > 0; i--) {
        CORE_RETURN_ERROR(append(*(s++)));
    }
    return ErrorCode::NONE;
}

Error Char32String::append(const signed char* s) {
    return append(reinterpret_cast<const char*>(s));
}

Error Char32String::append(const unsigned char* s) {
    return append(reinterpret_cast<const char*>(s));
}

Error Char32String::append(bool b) {
    return b ? append("true") : append("false");
}

Error Char32String::toString(CharString& s) const {
    int l = length; // length changes if &s == this
    for(int i = 0; i < l; i++) {
        CORE_RETURN_ERROR(s.append(data[i]));
    }
    return ErrorCode::NONE;
}

Error Char32String::toString(Char32String& s) const {
    int l = length; // length changes if &s == this
    for(int i = 0; i < l; i++) {
        CORE_RETURN_ERROR(s.append(data[i]));
    }
    return ErrorCode::NONE;
}

void Char32String::clear() {
    length = 0;
    hash = 0;
    data[0] = '\0';
}

u32 Char32String::hashCode() const {
    return hash;
}

Error Char32String::print() const {
    for(int i = 0; i < length; i++) {
        c32 c = data[i];
        if(c < (1 << 7)) {
            CORE_RETURN_ERROR(printChar(c, 0, 0x7F, 0x0));
        } else if(c < (1 << 11)) {
            CORE_RETURN_ERROR(printChar(c, 6, 0x1F, 0xC0));
            CORE_RETURN_ERROR(printChar(c, 0, 0x3F, 0x80));
        } else if(c < (1 << 16)) {
            CORE_RETURN_ERROR(printChar(c, 12, 0x0F, 0xE0));
            CORE_RETURN_ERROR(printChar(c, 6, 0x3F, 0x80));
            CORE_RETURN_ERROR(printChar(c, 0, 0x3F, 0x80));
        } else if(c < (1 << 21)) {
            CORE_RETURN_ERROR(printChar(c, 18, 0x07, 0xF0));
            CORE_RETURN_ERROR(printChar(c, 12, 0x3F, 0x80));
            CORE_RETURN_ERROR(printChar(c, 6, 0x3F, 0x80));
            CORE_RETURN_ERROR(printChar(c, 0, 0x3F, 0x80));
        }
    }
    return ErrorCode::NONE;
}

Error Char32String::printLine() const {
    CORE_RETURN_ERROR(print());
    CORE_RETURN_ERROR(Core::putChar('\n'));
    return ErrorCode::NONE;
}

bool Char32String::startsWidth(const Char32String& other, int from) const {
    if(from > length - other.getLength()) {
        return false;
    }
    for(int i = 0; i < other.getLength(); i++) {
        if(data[from + i] != other[i]) {
            return false;
        }
    }
    return true;
}

int Char32String::search(const Char32String& other, int from) const {
    for(int i = from; i < length; i++) {
        if(startsWidth(other, i)) {
            return i;
        }
    }
    return -1;
}

bool Char32String::contains(const Char32String& other, int from) const {
    return search(other, from) >= 0;
}

int Char32String::search(c32 u, int from) const {
    for(int i = from; i < length; i++) {
        if(data[i] == u) {
            return i;
        }
    }
    return -1;
}

bool Char32String::contains(c32 u, int from) const {
    return search(u, from) >= 0;
}

Error Char32String::substring(Char32String& s, int from, int to) const {
    s.clear();
    from = Math::max(from, 0);
    to = Math::min(to, length - 1);
    for(int i = from; i <= to; i++) {
        CORE_RETURN_ERROR(s.append(data[i]));
    }
    return ErrorCode::NONE;
}

Error Char32String::substring(Char32String& s, int from) const {
    return substring(s, from, length - 1);
}

Error Char32String::replace(Char32String& s, const Char32String& search,
                            const Char32String& replace) {
    int i = 0;
    while(i < length) {
        if(startsWidth(search, i)) {
            CORE_RETURN_ERROR(s.append(replace));
            i += search.getLength();
        } else {
            CORE_RETURN_ERROR(s.append(data[i]));
            i++;
        }
    }
    return copyFrom(s);
}

void Char32String::replace(c32 search, c32 replace) {
    hash = 0;
    for(int i = 0; i < length; i++) {
        if(data[i] == search) {
            data[i] = replace;
        }
        addToHash(static_cast<c32>(data[i]));
    }
}

Char32String::operator const c32*() const {
    return data;
}

void Char32String::addToHash(c32 u) {
    hash = static_cast<u32>(2120251889) * hash + static_cast<u32>(u);
}