diff options
Diffstat (limited to 'lib/reline/unicode.rb')
| -rw-r--r-- | lib/reline/unicode.rb | 149 |
1 files changed, 89 insertions, 60 deletions
diff --git a/lib/reline/unicode.rb b/lib/reline/unicode.rb index 6000c9f82a..82c9ec427c 100644 --- a/lib/reline/unicode.rb +++ b/lib/reline/unicode.rb @@ -38,33 +38,8 @@ class Reline::Unicode NON_PRINTING_START = "\1" NON_PRINTING_END = "\2" CSI_REGEXP = /\e\[[\d;]*[ABCDEFGHJKSTfminsuhl]/ - OSC_REGEXP = /\e\]\d+(?:;[^;]+)*\a/ + OSC_REGEXP = /\e\]\d+(?:;[^;\a\e]+)*(?:\a|\e\\)/ WIDTH_SCANNER = /\G(?:(#{NON_PRINTING_START})|(#{NON_PRINTING_END})|(#{CSI_REGEXP})|(#{OSC_REGEXP})|(\X))/o - NON_PRINTING_START_INDEX = 0 - NON_PRINTING_END_INDEX = 1 - CSI_REGEXP_INDEX = 2 - OSC_REGEXP_INDEX = 3 - GRAPHEME_CLUSTER_INDEX = 4 - - 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 def self.escape_for_print(str) str.chars.map! { |gr| @@ -132,15 +107,14 @@ class Reline::Unicode width = 0 rest = str.encode(Encoding::UTF_8) in_zero_width = false - rest.scan(WIDTH_SCANNER) do |gc| + rest.scan(WIDTH_SCANNER) do |non_printing_start, non_printing_end, csi, osc, gc| case - when gc[NON_PRINTING_START_INDEX] + when non_printing_start in_zero_width = true - when gc[NON_PRINTING_END_INDEX] + when non_printing_end in_zero_width = false - when gc[CSI_REGEXP_INDEX], gc[OSC_REGEXP_INDEX] - when gc[GRAPHEME_CLUSTER_INDEX] - gc = gc[GRAPHEME_CLUSTER_INDEX] + when csi, osc + when gc unless in_zero_width width += get_mbchar_width(gc) end @@ -154,30 +128,40 @@ class Reline::Unicode end end - def self.split_by_width(str, max_width, encoding = str.encoding) + def self.split_by_width(str, max_width, encoding = str.encoding, offset: 0) lines = [String.new(encoding: encoding)] height = 1 - width = 0 + width = offset rest = str.encode(Encoding::UTF_8) in_zero_width = false - rest.scan(WIDTH_SCANNER) do |gc| + seq = String.new(encoding: encoding) + rest.scan(WIDTH_SCANNER) do |non_printing_start, non_printing_end, csi, osc, gc| case - when gc[NON_PRINTING_START_INDEX] + when non_printing_start in_zero_width = true - when gc[NON_PRINTING_END_INDEX] + lines.last << NON_PRINTING_START + when non_printing_end 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] + lines.last << NON_PRINTING_END + when csi + lines.last << csi + unless in_zero_width + if csi == -"\e[m" || csi == -"\e[0m" + seq.clear + else + seq << csi + end + end + when osc + lines.last << osc + seq << osc + when gc 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) + lines << seq.dup height += 1 end end @@ -194,34 +178,79 @@ class Reline::Unicode end # Take a chunk of a String cut by width with escape sequences. - def self.take_range(str, start_col, max_width, encoding = str.encoding) - chunk = String.new(encoding: encoding) + def self.take_range(str, start_col, max_width) + take_mbchar_range(str, start_col, max_width).first + end + + def self.take_mbchar_range(str, start_col, width, cover_begin: false, cover_end: false, padding: false) + chunk = String.new(encoding: str.encoding) + + end_col = start_col + width total_width = 0 rest = str.encode(Encoding::UTF_8) in_zero_width = false - rest.scan(WIDTH_SCANNER) do |gc| + chunk_start_col = nil + chunk_end_col = nil + has_csi = false + rest.scan(WIDTH_SCANNER) do |non_printing_start, non_printing_end, csi, osc, gc| case - when gc[NON_PRINTING_START_INDEX] + when non_printing_start in_zero_width = true - when gc[NON_PRINTING_END_INDEX] + chunk << NON_PRINTING_START + when non_printing_end 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] + chunk << NON_PRINTING_END + when csi + has_csi = true + chunk << csi + when osc + chunk << osc + when gc if in_zero_width chunk << gc + next + end + + mbchar_width = get_mbchar_width(gc) + prev_width = total_width + total_width += mbchar_width + + if (cover_begin || padding ? total_width <= start_col : prev_width < start_col) + # Current character haven't reached start_col yet + next + elsif padding && !cover_begin && prev_width < start_col && start_col < total_width + # Add preceding padding. This padding might have background color. + chunk << ' ' + chunk_start_col ||= start_col + chunk_end_col = total_width + next + elsif (cover_end ? prev_width < end_col : total_width <= end_col) + # Current character is in the range + chunk << gc + chunk_start_col ||= prev_width + chunk_end_col = total_width + break if total_width >= end_col 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 + # Current character exceeds end_col + if padding && end_col < total_width + # Add succeeding padding. This padding might have background color. + chunk << ' ' + chunk_start_col ||= prev_width + chunk_end_col = end_col + end + break end end end - chunk + chunk_start_col ||= start_col + chunk_end_col ||= start_col + if padding && chunk_end_col < end_col + # Append padding. This padding should not include background color. + chunk << "\e[0m" if has_csi + chunk << ' ' * (end_col - chunk_end_col) + chunk_end_col = end_col + end + [chunk, chunk_start_col, chunk_end_col - chunk_start_col] end def self.get_next_mbchar_size(line, byte_pointer) |