unicode_text_segmentation.hpp

// Copyright Take Vos 2020.
// Distributed under the Boost Software License, Version 1.0.
// (See accompanying file LICENSE_1_0.txt or copy at https://www.boost.org/LICENSE_1_0.txt)
 
#pragma once
 
#include "unicode_grapheme_cluster_break.hpp"
#include "unicode_description.hpp"
 
namespace hi::inline v1 {
 
struct grapheme_break_state {
    unicode_grapheme_cluster_break previous = unicode_grapheme_cluster_break::Other;
    int RI_count = 0;
    bool first_character = true;
    bool in_extended_pictograph = false;
 
    void reset() noexcept
    {
        previous = unicode_grapheme_cluster_break::Other;
        RI_count = 0;
        first_character = true;
        in_extended_pictograph = false;
    }
};
 
[[nodiscard]] bool breaks_grapheme(char32_t code_point, grapheme_break_state &state) noexcept;
 
void wrap_lines(auto first, auto last, float max_width, auto get_width, auto get_code_point, auto set_code_point) noexcept
{
    using enum unicode_general_category;
 
    auto it_at_last_space = last;
    float width_at_last_space = 0.0;
    float current_width = 0.0;
 
    for (auto it = first; it != last; ++it) {
        hilet code_point = get_code_point(*it);
        hilet description = unicode_description::find(code_point);
        hilet general_category = description->general_category();
 
        if (general_category == Zp || general_category == Zl) {
            // Reset the line on existing line and paragraph separator.
            it_at_last_space = last;
            width_at_last_space = 0.0f;
            current_width = 0.0;
            continue;
 
        } else if (general_category == Zs) {
            // Remember the length of the line at the end of the word.
            it_at_last_space = it;
            width_at_last_space = current_width;
        }
 
        current_width += get_width(*it);
        if (current_width >= max_width && it_at_last_space != last) {
            // The line is too long, replace the last space with a line separator.
            set_code_point(*it, U'\u2028');
            it_at_last_space = last;
            width_at_last_space = 0.0f;
            current_width = 0.0;
            continue;
        }
    }
}
 
} // namespace hi::inline v1