diff options
Diffstat (limited to 'tool/enc-unicode.rb')
| -rwxr-xr-x | tool/enc-unicode.rb | 26 |
1 files changed, 22 insertions, 4 deletions
diff --git a/tool/enc-unicode.rb b/tool/enc-unicode.rb index 9d49f427bb..a89390ad8f 100755 --- a/tool/enc-unicode.rb +++ b/tool/enc-unicode.rb @@ -12,6 +12,9 @@ # You can get source file for gperf. After this, simply make ruby. # Or directly run: # tool/enc-unicode.rb --header data_dir emoji_data_dir > enc/unicode/<VERSION>/name2ctype.h +# +# There are Makefile rules that automate steps above: `make update-unicode` and +# `make enc/unicode/<VERSION>/name2ctype.h`. while arg = ARGV.shift case arg @@ -143,7 +146,8 @@ def define_posix_props(data) data['Space'] = data['White_Space'] data['Blank'] = data['Space_Separator'] + [0x0009] data['Cntrl'] = data['Cc'] - data['Word'] = data['Alpha'] + data['Mark'] + data['Digit'] + data['Connector_Punctuation'] + data['Word'] = data['Alpha'] + data['Mark'] + data['Digit'] + + data['Connector_Punctuation'] + data['Join_Control'] data['Graph'] = data['Any'] - data['Space'] - data['Cntrl'] - data['Surrogate'] - data['Unassigned'] data['Print'] = data['Graph'] + data['Space_Separator'] @@ -161,14 +165,24 @@ def parse_scripts(data, categories) names = {} files.each do |file| data_foreach(file[:fn]) do |line| + # Parse Unicode data files and store code points and properties. if /^# Total (?:code points|elements): / =~ line data[current] = cps categories[current] = file[:title] (names[file[:title]] ||= []) << current cps = [] - elsif /^(\h+)(?:\.\.(\h+))?\s*;\s*(\w+)/ =~ line - current = $3 + elsif /^(\h+)(?:\.\.(\h+))?\s*;\s*(\w(?:[\w\s;]*\w)?)/ =~ line + # $1: The first hexadecimal code point or the start of a range. + # $2: The end code point of the range, if present. + # If there's no range (just a single code point), $2 is nil. + # $3: The property or other info. + # Example: + # line = "0915..0939 ; InCB; Consonant # Lo [37] DEVANAGARI LETTER KA..DEVANAGARI LETTER HA" + # $1 = "0915" + # $2 = "0939" + # $3 = "InCB; Consonant" $2 ? cps.concat(($1.to_i(16)..$2.to_i(16)).to_a) : cps.push($1.to_i(16)) + current = $3.gsub(/\W+/, '_') end end end @@ -486,7 +500,11 @@ end output.ifdef :USE_UNICODE_PROPERTIES do props.each do |name| i += 1 - name = normalize_propname(name) + name = if name.start_with?('InCB') + name.downcase.gsub(/_/, '=') + else + normalize_propname(name) + end name_to_index[name] = i puts "%-40s %3d" % [name + ',', i] end |