class Reline::Unicode

Constants

CSI_REGEXP
CSI_REGEXP_INDEX
EscapedChars
EscapedPairs
GRAPHEME_CLUSTER_INDEX
HalfwidthDakutenHandakuten
MBCharWidthRE
NON_PRINTING_END
NON_PRINTING_END_INDEX
NON_PRINTING_START
NON_PRINTING_START_INDEX
OSC_REGEXP
OSC_REGEXP_INDEX
WIDTH_SCANNER

Public Class Methods

calculate_width(str, allow_escape_code = false) click to toggle source
# File lib/reline/unicode.rb, line 130
def self.calculate_width(str, allow_escape_code = false)
  if allow_escape_code
    width = 0
    rest = str.encode(Encoding::UTF_8)
    in_zero_width = false
    rest.scan(WIDTH_SCANNER) do |gc|
      case
      when gc[NON_PRINTING_START_INDEX]
        in_zero_width = true
      when gc[NON_PRINTING_END_INDEX]
        in_zero_width = false
      when gc[CSI_REGEXP_INDEX], gc[OSC_REGEXP_INDEX]
      when gc[GRAPHEME_CLUSTER_INDEX]
        gc = gc[GRAPHEME_CLUSTER_INDEX]
        unless in_zero_width
          width += get_mbchar_width(gc)
        end
      end
    end
    width
  else
    str.encode(Encoding::UTF_8).grapheme_clusters.inject(0) { |w, gc|
      w + get_mbchar_width(gc)
    }
  end
end
ed_transpose_words(line, byte_pointer) click to toggle source
# File lib/reline/unicode.rb, line 330
def self.ed_transpose_words(line, byte_pointer)
  right_word_start = nil
  size = get_next_mbchar_size(line, byte_pointer)
  mbchar = line.byteslice(byte_pointer, size)
  if size.zero?
    # ' aaa bbb [cursor]'
    byte_size = 0
    while 0 < (byte_pointer + byte_size)
      size = get_prev_mbchar_size(line, byte_pointer + byte_size)
      mbchar = line.byteslice(byte_pointer + byte_size - size, size)
      break if mbchar.encode(Encoding::UTF_8) =~ /\p{Word}/
      byte_size -= size
    end
    while 0 < (byte_pointer + byte_size)
      size = get_prev_mbchar_size(line, byte_pointer + byte_size)
      mbchar = line.byteslice(byte_pointer + byte_size - size, size)
      break if mbchar.encode(Encoding::UTF_8) =~ /\P{Word}/
      byte_size -= size
    end
    right_word_start = byte_pointer + byte_size
    byte_size = 0
    while line.bytesize > (byte_pointer + byte_size)
      size = get_next_mbchar_size(line, byte_pointer + byte_size)
      mbchar = line.byteslice(byte_pointer + byte_size, size)
      break if mbchar.encode(Encoding::UTF_8) =~ /\P{Word}/
      byte_size += size
    end
    after_start = byte_pointer + byte_size
  elsif mbchar.encode(Encoding::UTF_8) =~ /\p{Word}/
    # ' aaa bb[cursor]b'
    byte_size = 0
    while 0 < (byte_pointer + byte_size)
      size = get_prev_mbchar_size(line, byte_pointer + byte_size)
      mbchar = line.byteslice(byte_pointer + byte_size - size, size)
      break if mbchar.encode(Encoding::UTF_8) =~ /\P{Word}/
      byte_size -= size
    end
    right_word_start = byte_pointer + byte_size
    byte_size = 0
    while line.bytesize > (byte_pointer + byte_size)
      size = get_next_mbchar_size(line, byte_pointer + byte_size)
      mbchar = line.byteslice(byte_pointer + byte_size, size)
      break if mbchar.encode(Encoding::UTF_8) =~ /\P{Word}/
      byte_size += size
    end
    after_start = byte_pointer + byte_size
  else
    byte_size = 0
    while (line.bytesize - 1) > (byte_pointer + byte_size)
      size = get_next_mbchar_size(line, byte_pointer + byte_size)
      mbchar = line.byteslice(byte_pointer + byte_size, size)
      break if mbchar.encode(Encoding::UTF_8) =~ /\p{Word}/
      byte_size += size
    end
    if (byte_pointer + byte_size) == (line.bytesize - 1)
      # ' aaa bbb [cursor] '
      after_start = line.bytesize
      while 0 < (byte_pointer + byte_size)
        size = get_prev_mbchar_size(line, byte_pointer + byte_size)
        mbchar = line.byteslice(byte_pointer + byte_size - size, size)
        break if mbchar.encode(Encoding::UTF_8) =~ /\p{Word}/
        byte_size -= size
      end
      while 0 < (byte_pointer + byte_size)
        size = get_prev_mbchar_size(line, byte_pointer + byte_size)
        mbchar = line.byteslice(byte_pointer + byte_size - size, size)
        break if mbchar.encode(Encoding::UTF_8) =~ /\P{Word}/
        byte_size -= size
      end
      right_word_start = byte_pointer + byte_size
    else
      # ' aaa [cursor] bbb '
      right_word_start = byte_pointer + byte_size
      while line.bytesize > (byte_pointer + byte_size)
        size = get_next_mbchar_size(line, byte_pointer + byte_size)
        mbchar = line.byteslice(byte_pointer + byte_size, size)
        break if mbchar.encode(Encoding::UTF_8) =~ /\P{Word}/
        byte_size += size
      end
      after_start = byte_pointer + byte_size
    end
  end
  byte_size = right_word_start - byte_pointer
  while 0 < (byte_pointer + byte_size)
    size = get_prev_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size - size, size)
    break if mbchar.encode(Encoding::UTF_8) =~ /\p{Word}/
    byte_size -= size
  end
  middle_start = byte_pointer + byte_size
  byte_size = middle_start - byte_pointer
  while 0 < (byte_pointer + byte_size)
    size = get_prev_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size - size, size)
    break if mbchar.encode(Encoding::UTF_8) =~ /\P{Word}/
    byte_size -= size
  end
  left_word_start = byte_pointer + byte_size
  [left_word_start, middle_start, right_word_start, after_start]
end
em_backward_word(line, byte_pointer) click to toggle source
# File lib/reline/unicode.rb, line 290
def self.em_backward_word(line, byte_pointer)
  width = 0
  byte_size = 0
  while 0 < (byte_pointer - byte_size)
    size = get_prev_mbchar_size(line, byte_pointer - byte_size)
    mbchar = line.byteslice(byte_pointer - byte_size - size, size)
    break if mbchar.encode(Encoding::UTF_8) =~ /\p{Word}/
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  while 0 < (byte_pointer - byte_size)
    size = get_prev_mbchar_size(line, byte_pointer - byte_size)
    mbchar = line.byteslice(byte_pointer - byte_size - size, size)
    break if mbchar.encode(Encoding::UTF_8) =~ /\P{Word}/
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  [byte_size, width]
end
em_big_backward_word(line, byte_pointer) click to toggle source
# File lib/reline/unicode.rb, line 310
def self.em_big_backward_word(line, byte_pointer)
  width = 0
  byte_size = 0
  while 0 < (byte_pointer - byte_size)
    size = get_prev_mbchar_size(line, byte_pointer - byte_size)
    mbchar = line.byteslice(byte_pointer - byte_size - size, size)
    break if mbchar =~ /\S/
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  while 0 < (byte_pointer - byte_size)
    size = get_prev_mbchar_size(line, byte_pointer - byte_size)
    mbchar = line.byteslice(byte_pointer - byte_size - size, size)
    break if mbchar =~ /\s/
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  [byte_size, width]
end
em_forward_word(line, byte_pointer) click to toggle source
# File lib/reline/unicode.rb, line 241
def self.em_forward_word(line, byte_pointer)
  width = 0
  byte_size = 0
  while line.bytesize > (byte_pointer + byte_size)
    size = get_next_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size, size)
    break if mbchar.encode(Encoding::UTF_8) =~ /\p{Word}/
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  while line.bytesize > (byte_pointer + byte_size)
    size = get_next_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size, size)
    break if mbchar.encode(Encoding::UTF_8) =~ /\P{Word}/
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  [byte_size, width]
end
em_forward_word_with_capitalization(line, byte_pointer) click to toggle source
# File lib/reline/unicode.rb, line 261
def self.em_forward_word_with_capitalization(line, byte_pointer)
  width = 0
  byte_size = 0
  new_str = String.new
  while line.bytesize > (byte_pointer + byte_size)
    size = get_next_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size, size)
    break if mbchar.encode(Encoding::UTF_8) =~ /\p{Word}/
    new_str += mbchar
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  first = true
  while line.bytesize > (byte_pointer + byte_size)
    size = get_next_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size, size)
    break if mbchar.encode(Encoding::UTF_8) =~ /\P{Word}/
    if first
      new_str += mbchar.upcase
      first = false
    else
      new_str += mbchar.downcase
    end
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  [byte_size, width, new_str]
end
escape_for_print(str) click to toggle source
# File lib/reline/unicode.rb, line 69
def self.escape_for_print(str)
  str.chars.map! { |gr|
    escaped = EscapedPairs[gr.ord]
    if escaped && gr != -"\n" && gr != -"\t"
      escaped
    else
      gr
    end
  }.join
end
get_mbchar_byte_size_by_first_char(c) click to toggle source
# File lib/reline/unicode.rb, line 49
def self.get_mbchar_byte_size_by_first_char(c)
  # Checks UTF-8 character byte size
  case c.ord
  # 0b0xxxxxxx
  when ->(code) { (code ^ 0b10000000).allbits?(0b10000000) } then 1
  # 0b110xxxxx
  when ->(code) { (code ^ 0b00100000).allbits?(0b11100000) } then 2
  # 0b1110xxxx
  when ->(code) { (code ^ 0b00010000).allbits?(0b11110000) } then 3
  # 0b11110xxx
  when ->(code) { (code ^ 0b00001000).allbits?(0b11111000) } then 4
  # 0b111110xx
  when ->(code) { (code ^ 0b00000100).allbits?(0b11111100) } then 5
  # 0b1111110x
  when ->(code) { (code ^ 0b00000010).allbits?(0b11111110) } then 6
  # successor of mbchar
  else 0
  end
end
get_mbchar_width(mbchar) click to toggle source
# File lib/reline/unicode.rb, line 110
def self.get_mbchar_width(mbchar)
  ord = mbchar.ord
  if (0x00 <= ord and ord <= 0x1F) # in EscapedPairs
    return 2
  elsif (0x20 <= ord and ord <= 0x7E) # printable ASCII chars
    return 1
  end
  m = mbchar.encode(Encoding::UTF_8).match(MBCharWidthRE)
  case
  when m.nil? then 1 # TODO should be U+FFFD � REPLACEMENT CHARACTER
  when m[:width_2_1], m[:width_2_2], m[:width_2_3] then 2
  when m[:width_3] then 3
  when m[:width_0] then 0
  when m[:width_1] then 1
  when m[:ambiguous_width] then Reline.ambiguous_width
  else
    nil
  end
end
get_next_mbchar_size(line, byte_pointer) click to toggle source
# File lib/reline/unicode.rb, line 227
def self.get_next_mbchar_size(line, byte_pointer)
  grapheme = line.byteslice(byte_pointer..-1).grapheme_clusters.first
  grapheme ? grapheme.bytesize : 0
end
get_prev_mbchar_size(line, byte_pointer) click to toggle source
# File lib/reline/unicode.rb, line 232
def self.get_prev_mbchar_size(line, byte_pointer)
  if byte_pointer.zero?
    0
  else
    grapheme = line.byteslice(0..(byte_pointer - 1)).grapheme_clusters.last
    grapheme ? grapheme.bytesize : 0
  end
end
split_by_width(str, max_width, encoding = str.encoding) click to toggle source
# File lib/reline/unicode.rb, line 157
def self.split_by_width(str, max_width, encoding = str.encoding)
  lines = [String.new(encoding: encoding)]
  height = 1
  width = 0
  rest = str.encode(Encoding::UTF_8)
  in_zero_width = false
  rest.scan(WIDTH_SCANNER) do |gc|
    case
    when gc[NON_PRINTING_START_INDEX]
      in_zero_width = true
    when gc[NON_PRINTING_END_INDEX]
      in_zero_width = false
    when gc[CSI_REGEXP_INDEX]
      lines.last << gc[CSI_REGEXP_INDEX]
    when gc[OSC_REGEXP_INDEX]
      lines.last << gc[OSC_REGEXP_INDEX]
    when gc[GRAPHEME_CLUSTER_INDEX]
      gc = gc[GRAPHEME_CLUSTER_INDEX]
      unless in_zero_width
        mbchar_width = get_mbchar_width(gc)
        if (width += mbchar_width) > max_width
          width = mbchar_width
          lines << nil
          lines << String.new(encoding: encoding)
          height += 1
        end
      end
      lines.last << gc
    end
  end
  # The cursor moves to next line in first
  if width == max_width
    lines << nil
    lines << String.new(encoding: encoding)
    height += 1
  end
  [lines, height]
end
take_range(str, start_col, max_width, encoding = str.encoding) click to toggle source

Take a chunk of a String cut by width with escape sequences.

# File lib/reline/unicode.rb, line 197
def self.take_range(str, start_col, max_width, encoding = str.encoding)
  chunk = String.new(encoding: encoding)
  total_width = 0
  rest = str.encode(Encoding::UTF_8)
  in_zero_width = false
  rest.scan(WIDTH_SCANNER) do |gc|
    case
    when gc[NON_PRINTING_START_INDEX]
      in_zero_width = true
    when gc[NON_PRINTING_END_INDEX]
      in_zero_width = false
    when gc[CSI_REGEXP_INDEX]
      chunk << gc[CSI_REGEXP_INDEX]
    when gc[OSC_REGEXP_INDEX]
      chunk << gc[OSC_REGEXP_INDEX]
    when gc[GRAPHEME_CLUSTER_INDEX]
      gc = gc[GRAPHEME_CLUSTER_INDEX]
      if in_zero_width
        chunk << gc
      else
        mbchar_width = get_mbchar_width(gc)
        total_width += mbchar_width
        break if (start_col + max_width) < total_width
        chunk << gc if start_col < total_width
      end
    end
  end
  chunk
end
vi_backward_word(line, byte_pointer) click to toggle source
# File lib/reline/unicode.rb, line 619
def self.vi_backward_word(line, byte_pointer)
  width = 0
  byte_size = 0
  while 0 < (byte_pointer - byte_size)
    size = get_prev_mbchar_size(line, byte_pointer - byte_size)
    mbchar = line.byteslice(byte_pointer - byte_size - size, size)
    if mbchar =~ /\S/
      if mbchar =~ /\w/
        started_by = :word
      else
        started_by = :non_word_printable
      end
      break
    end
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  while 0 < (byte_pointer - byte_size)
    size = get_prev_mbchar_size(line, byte_pointer - byte_size)
    mbchar = line.byteslice(byte_pointer - byte_size - size, size)
    case started_by
    when :word
      break if mbchar =~ /\W/
    when :non_word_printable
      break if mbchar =~ /[\w\s]/
    end
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  [byte_size, width]
end
vi_big_backward_word(line, byte_pointer) click to toggle source
# File lib/reline/unicode.rb, line 481
def self.vi_big_backward_word(line, byte_pointer)
  width = 0
  byte_size = 0
  while 0 < (byte_pointer - byte_size)
    size = get_prev_mbchar_size(line, byte_pointer - byte_size)
    mbchar = line.byteslice(byte_pointer - byte_size - size, size)
    break if mbchar =~ /\S/
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  while 0 < (byte_pointer - byte_size)
    size = get_prev_mbchar_size(line, byte_pointer - byte_size)
    mbchar = line.byteslice(byte_pointer - byte_size - size, size)
    break if mbchar =~ /\s/
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  [byte_size, width]
end
vi_big_forward_end_word(line, byte_pointer) click to toggle source
# File lib/reline/unicode.rb, line 451
def self.vi_big_forward_end_word(line, byte_pointer)
  if (line.bytesize - 1) > byte_pointer
    size = get_next_mbchar_size(line, byte_pointer)
    mbchar = line.byteslice(byte_pointer, size)
    width = get_mbchar_width(mbchar)
    byte_size = size
  else
    return [0, 0]
  end
  while (line.bytesize - 1) > (byte_pointer + byte_size)
    size = get_next_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size, size)
    break if mbchar =~ /\S/
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  prev_width = width
  prev_byte_size = byte_size
  while line.bytesize > (byte_pointer + byte_size)
    size = get_next_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size, size)
    break if mbchar =~ /\s/
    prev_width = width
    prev_byte_size = byte_size
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  [prev_byte_size, prev_width]
end
vi_big_forward_word(line, byte_pointer) click to toggle source
# File lib/reline/unicode.rb, line 431
def self.vi_big_forward_word(line, byte_pointer)
  width = 0
  byte_size = 0
  while (line.bytesize - 1) > (byte_pointer + byte_size)
    size = get_next_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size, size)
    break if mbchar =~ /\s/
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  while (line.bytesize - 1) > (byte_pointer + byte_size)
    size = get_next_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size, size)
    break if mbchar =~ /\S/
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  [byte_size, width]
end
vi_first_print(line) click to toggle source
# File lib/reline/unicode.rb, line 651
def self.vi_first_print(line)
  width = 0
  byte_size = 0
  while (line.bytesize - 1) > byte_size
    size = get_next_mbchar_size(line, byte_size)
    mbchar = line.byteslice(byte_size, size)
    if mbchar =~ /\S/
      break
    end
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  [byte_size, width]
end
vi_forward_end_word(line, byte_pointer) click to toggle source
# File lib/reline/unicode.rb, line 542
def self.vi_forward_end_word(line, byte_pointer)
  if (line.bytesize - 1) > byte_pointer
    size = get_next_mbchar_size(line, byte_pointer)
    mbchar = line.byteslice(byte_pointer, size)
    if mbchar =~ /\w/
      started_by = :word
    elsif mbchar =~ /\s/
      started_by = :space
    else
      started_by = :non_word_printable
    end
    width = get_mbchar_width(mbchar)
    byte_size = size
  else
    return [0, 0]
  end
  if (line.bytesize - 1) > (byte_pointer + byte_size)
    size = get_next_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size, size)
    if mbchar =~ /\w/
      second = :word
    elsif mbchar =~ /\s/
      second = :space
    else
      second = :non_word_printable
    end
    second_width = get_mbchar_width(mbchar)
    second_byte_size = size
  else
    return [byte_size, width]
  end
  if second == :space
    width += second_width
    byte_size += second_byte_size
    while (line.bytesize - 1) > (byte_pointer + byte_size)
      size = get_next_mbchar_size(line, byte_pointer + byte_size)
      mbchar = line.byteslice(byte_pointer + byte_size, size)
      if mbchar =~ /\S/
        if mbchar =~ /\w/
          started_by = :word
        else
          started_by = :non_word_printable
        end
        break
      end
      width += get_mbchar_width(mbchar)
      byte_size += size
    end
  else
    case [started_by, second]
    when [:word, :non_word_printable], [:non_word_printable, :word]
      started_by = second
    else
      width += second_width
      byte_size += second_byte_size
      started_by = second
    end
  end
  prev_width = width
  prev_byte_size = byte_size
  while line.bytesize > (byte_pointer + byte_size)
    size = get_next_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size, size)
    case started_by
    when :word
      break if mbchar =~ /\W/
    when :non_word_printable
      break if mbchar =~ /[\w\s]/
    end
    prev_width = width
    prev_byte_size = byte_size
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  [prev_byte_size, prev_width]
end
vi_forward_word(line, byte_pointer, drop_terminate_spaces = false) click to toggle source
# File lib/reline/unicode.rb, line 501
def self.vi_forward_word(line, byte_pointer, drop_terminate_spaces = false)
  if line.bytesize > byte_pointer
    size = get_next_mbchar_size(line, byte_pointer)
    mbchar = line.byteslice(byte_pointer, size)
    if mbchar =~ /\w/
      started_by = :word
    elsif mbchar =~ /\s/
      started_by = :space
    else
      started_by = :non_word_printable
    end
    width = get_mbchar_width(mbchar)
    byte_size = size
  else
    return [0, 0]
  end
  while line.bytesize > (byte_pointer + byte_size)
    size = get_next_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size, size)
    case started_by
    when :word
      break if mbchar =~ /\W/
    when :space
      break if mbchar =~ /\S/
    when :non_word_printable
      break if mbchar =~ /\w|\s/
    end
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  return [byte_size, width] if drop_terminate_spaces
  while line.bytesize > (byte_pointer + byte_size)
    size = get_next_mbchar_size(line, byte_pointer + byte_size)
    mbchar = line.byteslice(byte_pointer + byte_size, size)
    break if mbchar =~ /\S/
    width += get_mbchar_width(mbchar)
    byte_size += size
  end
  [byte_size, width]
end