utf8.hh

#ifndef utf8_hh_INCLUDED
#define utf8_hh_INCLUDED

#include "assert.hh"
#include "unicode.hh"
#include "units.hh"
#include "optional.hh"

namespace Kakoune
{

namespace utf8
{

template<typename Iterator>
[[gnu::always_inline]]
inline char read(Iterator& it) noexcept { char c = *it; ++it; return c; }

// return true if it points to the first byte of a (either single or
// multibyte) character
[[gnu::always_inline]]
inline bool is_character_start(char c) noexcept
{
    return (c & 0xC0) != 0x80;
}

namespace InvalidPolicy
{

struct Assert
{
    Codepoint operator()(Codepoint cp) const { kak_assert(false); return cp; }
};

struct Pass
{
    Codepoint operator()(Codepoint cp) const noexcept { return cp; }
};

}

// returns the codepoint of the character whose first byte
// is pointed by it
template<typename InvalidPolicy = utf8::InvalidPolicy::Pass,
         typename Iterator, typename Sentinel>
Codepoint read_codepoint(Iterator& it, const Sentinel& end)
    noexcept(noexcept(InvalidPolicy{}(0)))
{
    if (it == end)
        return InvalidPolicy{}(-1);
    // According to rfc3629, UTF-8 allows only up to 4 bytes.
    // (21 bits codepoint)
    unsigned char byte = read(it);
    if ((byte & 0x80) == 0) // 0xxxxxxx
        return byte;

    if (it == end)
        return InvalidPolicy{}(byte);

    if ((byte & 0xE0) == 0xC0) // 110xxxxx
        return ((byte & 0x1F) << 6) | (read(it) & 0x3F);

    if ((byte & 0xF0) == 0xE0) // 1110xxxx
    {
        Codepoint cp = ((byte & 0x0F) << 12) | ((read(it) & 0x3F) << 6);
        if (it == end)
            return InvalidPolicy{}(cp);
        return cp | (read(it) & 0x3F);
    }

    if ((byte & 0xF8) == 0xF0) // 11110xxx
    {
        Codepoint cp = ((byte & 0x0F) << 18) | ((read(it) & 0x3F) << 12);
        if (it == end)
            return InvalidPolicy{}(cp);
        cp |= (read(it) & 0x3F) << 6;
        if (it == end)
            return InvalidPolicy{}(cp);
        return cp | (read(it) & 0x3F);
    }
    return InvalidPolicy{}(byte);
}

template<typename InvalidPolicy = utf8::InvalidPolicy::Pass,
         typename Iterator, typename Sentinel>
Codepoint codepoint(Iterator it, const Sentinel& end)
    noexcept(noexcept(read_codepoint<InvalidPolicy>(it, end)))
{
    return read_codepoint<InvalidPolicy>(it, end);
}

template<typename InvalidPolicy = utf8::InvalidPolicy::Pass>
ByteCount codepoint_size(char byte)
    noexcept(noexcept(InvalidPolicy{}(0)))
{
    if ((byte & 0x80) == 0) // 0xxxxxxx
        return 1;
    else if ((byte & 0xE0) == 0xC0) // 110xxxxx
        return 2;
    else if ((byte & 0xF0) == 0xE0) // 1110xxxx
        return 3;
    else if ((byte & 0xF8) == 0xF0) // 11110xxx
        return 4;
    else
    {
        InvalidPolicy{}(byte);
        return 1;
    }
}

template<typename InvalidPolicy = utf8::InvalidPolicy::Pass>
ByteCount codepoint_size(Codepoint cp)
    noexcept(noexcept(InvalidPolicy{}(0)))
{
    if (cp <= 0x7F)
        return 1;
    else if (cp <= 0x7FF)
        return 2;
    else if (cp <= 0xFFFF)
        return 3;
    else if (cp <= 0x10FFFF)
        return 4;
    else
    {
        InvalidPolicy{}(cp);
        return 0;
    }
}

template<typename Iterator, typename Sentinel>
void to_next(Iterator& it, const Sentinel& end) noexcept
{
    if (it != end)
        ++it;
    while (it != end and not is_character_start(*it))
        ++it;
}

// returns an iterator to next character first byte
template<typename Iterator, typename Sentinel>
Iterator next(Iterator it, const Sentinel& end) noexcept
{
    to_next(it, end);
    return it;
}

// returns it's parameter if it points to a character first byte,
// or else returns next character first byte
template<typename Iterator, typename Sentinel>
Iterator finish(Iterator it, const Sentinel& end) noexcept
{
    while (it != end and (*(it) & 0xC0) == 0x80)
        ++it;
    return it;
}

template<typename Iterator, typename Sentinel>
void to_previous(Iterator& it, const Sentinel& begin) noexcept
{
    if (it != begin)
        --it;
    while (it != begin and not is_character_start(*it))
        --it;
}
// returns an iterator to the previous character first byte
template<typename Iterator, typename Sentinel>
Iterator previous(Iterator it, const Sentinel& begin) noexcept
{
    to_previous(it, begin);
    return it;
}

// returns an iterator pointing to the first byte of the
// dth character after (or before if d < 0) the character
// pointed by it
template<typename Iterator, typename Sentinel>
Iterator advance(Iterator it, const Sentinel& end, CharCount d) noexcept
{
    if (it == end)
        return it;

    if (d < 0)
    {
        while (it != end and d++ != 0)
            to_previous(it, end);
    }
    else if (d > 0)
    {
        while (it != end and d-- != 0)
            to_next(it, end);
    }
    return it;
}

// returns an iterator pointing to the first byte of the
// character at the dth column after (or before if d < 0)
// the character pointed by it
template<typename Iterator, typename Sentinel>
Iterator advance(Iterator it, const Sentinel& end, ColumnCount d) noexcept
{
    if (it == end)
        return it;

    if (d < 0)
    {
        while (it != end and d < 0)
        {
            auto cur = it;
            to_previous(it, end);
            d += codepoint_width(codepoint(it, cur));
        }
    }
    else if (d > 0)
    {
        auto begin = it;
        while (it != end and d > 0)
        {
            d -= codepoint_width(read_codepoint(it, end));
            if (it != end and d < 0)
                to_previous(it, begin);
        }
    }
    return it;
}

// returns the character count between begin and end
template<typename Iterator, typename Sentinel>
CharCount distance(Iterator begin, const Sentinel& end) noexcept
{
    CharCount dist = 0;

    while (begin != end)
    {
        if (is_character_start(read(begin)))
            ++dist;
    }
    return dist;
}

// returns the column count between begin and end
template<typename Iterator, typename Sentinel>
ColumnCount column_distance(Iterator begin, const Sentinel& end) noexcept
{
    ColumnCount dist = 0;

    while (begin != end)
        dist += codepoint_width(read_codepoint(begin, end));
    return dist;
}

// returns an iterator to the first byte of the character it is into
template<typename Iterator, typename Sentinel>
Iterator character_start(Iterator it, const Sentinel& begin) noexcept
{
    while (it != begin and not is_character_start(*it))
        --it;
    return it;
}

// returns an optional iterator to the first byte of the previous character
// or no value if it is at begin
template<typename Iterator, typename Sentinel>
static Optional<Codepoint> prev_codepoint(Iterator it, const Sentinel& begin) noexcept
{
    if (it <= begin)
        return {};
    return codepoint(character_start(it -1, begin), it);
}


template<typename OutputIterator, typename InvalidPolicy = utf8::InvalidPolicy::Pass>
void dump(OutputIterator&& it, Codepoint cp)
{
    if (cp <= 0x7F)
        *it++ = cp;
    else if (cp <= 0x7FF)
    {
        *it++ = 0xC0 | (cp >> 6);
        *it++ = 0x80 | (cp & 0x3F);
    }
    else if (cp <= 0xFFFF)
    {
        *it++ = 0xE0 | (cp >> 12);
        *it++ = 0x80 | ((cp >> 6) & 0x3F);
        *it++ = 0x80 | (cp & 0x3F);
    }
    else if (cp <= 0x10FFFF)
    {
        *it++ = 0xF0 | (cp >> 18);
        *it++ = 0x80 | ((cp >> 12) & 0x3F);
        *it++ = 0x80 | ((cp >> 6)  & 0x3F);
        *it++ = 0x80 | (cp & 0x3F);
    }
    else
        InvalidPolicy{}(cp);
}

}

}

#endif // utf8_hh_INCLUDED