multibyte 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- data/.git/COMMIT_EDITMSG +57 -0
- data/.git/HEAD +1 -0
- data/.git/config +8 -0
- data/.git/description +1 -0
- data/.git/hooks/applypatch-msg +15 -0
- data/.git/hooks/commit-msg +21 -0
- data/.git/hooks/post-commit +8 -0
- data/.git/hooks/post-receive +16 -0
- data/.git/hooks/post-update +8 -0
- data/.git/hooks/pre-applypatch +14 -0
- data/.git/hooks/pre-commit +70 -0
- data/.git/hooks/pre-rebase +150 -0
- data/.git/hooks/update +107 -0
- data/.git/index +0 -0
- data/.git/info/exclude +6 -0
- data/.git/logs/HEAD +4 -0
- data/.git/logs/refs/heads/master +4 -0
- data/.git/logs/refs/remotes/origin/master +3 -0
- data/.git/objects/02/e9a792fa31b00f0f06012e0b5e3ef327c40f64 +0 -0
- data/.git/objects/10/0b93820ade4c16225673b4ca62bb3ade63c313 +0 -0
- data/.git/objects/10/aab6490174970a2349d1050882c644f8d07006 +7 -0
- data/.git/objects/11/cda2eaac1ae103a13bf835456de05f53b0c8ad +0 -0
- data/.git/objects/1d/61c71cd5cf0f4dae6305b007a6177e1cecd7ea +0 -0
- data/.git/objects/2c/84cd00c156ba3b46e733c7e34dd9aea0c4fa75 +0 -0
- data/.git/objects/2f/3a16e4a756ec7a2d32c4bd8386b855f41cae61 +0 -0
- data/.git/objects/2f/437425b2c085bb529379e409d6b4bc7f9f9fba +0 -0
- data/.git/objects/3f/9afe5eaf00ad4d2ffa2e9315b96f530b0e85d4 +2 -0
- data/.git/objects/42/4a5f37c6fe3a7cac54b0f85688c1cce7da9cdf +0 -0
- data/.git/objects/47/26f03dc598a372cb959e3cbe33334de70802fd +0 -0
- data/.git/objects/55/37e6b2a449fe6d6c800803d0f72ffbcf44c297 +0 -0
- data/.git/objects/64/334b80d5fcead16f13cb8b7a45120ea460747b +0 -0
- data/.git/objects/69/1ed3b65603a0d0dabfb66720e7163664be78f5 +0 -0
- data/.git/objects/72/141584a5836168fe912ebce5b4537bd0eafc09 +1 -0
- data/.git/objects/7c/e872bc55ff766e542d2fbbe00e9dde70baaa92 +2 -0
- data/.git/objects/93/e03faa6906892434fa9f2ad230c4ececc64055 +0 -0
- data/.git/objects/9d/c9a787700bafb70e1833a760b73dec26a6a56e +0 -0
- data/.git/objects/9f/672edccfaaa00b6584f599c3e68560e6a5bedf +0 -0
- data/.git/objects/ab/1cd78b02ec268ce7d2faf4b1692f4bed9273fe +0 -0
- data/.git/objects/ad/8866bd33ae1afe9278cf5ee833d71977af561c +0 -0
- data/.git/objects/af/c3ea327fa220ee42a5ebecfcc7d218be4ea1d9 +0 -0
- data/.git/objects/b2/56d1cf3a9ba99f308daf1b4cd0dd08f11448e0 +1 -0
- data/.git/objects/bc/184b885eca1bed875171bd3990433055dda3af +0 -0
- data/.git/objects/c2/7f6559350f7adb19d43742b55b2f91d07b6550 +0 -0
- data/.git/objects/c3/e7364c6e90fb80e979779cd822bc8cdc8a3790 +0 -0
- data/.git/objects/c9/5128ff8f94e949d796afa128f2f341ba69989c +0 -0
- data/.git/objects/ce/e1f5b453e86a5aa913c626641608839bb79eef +1 -0
- data/.git/objects/cf/6add7ea568d3d90d6a1f8afb0898b0119b14ff +0 -0
- data/.git/objects/d5/e18b1fd7b189fdc75dd3bc4d208323cbc01aa8 +0 -0
- data/.git/objects/de/6e40dc4d92c1a889f8581718a48f2fcd219442 +0 -0
- data/.git/objects/df/42782d610e8c714f78cf0a018c600667da795e +0 -0
- data/.git/objects/e0/65ff01f0883142d4901a371be8dc60998fbe04 +4 -0
- data/.git/objects/e1/ad06969c3d3982b06ebd7da3777a0926501e7c +0 -0
- data/.git/objects/e4/69154c2ae4c3c596870a5e920f197b7b5be179 +2 -0
- data/.git/objects/e4/8464df56bf487e96e21ea99487330266dae3c9 +0 -0
- data/.git/objects/e6/9de29bb2d1d6434b8b29ae775ad8c2e48c5391 +0 -0
- data/.git/objects/e8/517350dbaadcb01282282218638c6022a20091 +0 -0
- data/.git/objects/eb/1bffe6adda4dc202d219cb35fe142a5e42b46b +0 -0
- data/.git/objects/eb/5d1c06d3a469df91dff68022b3b179e30cd127 +0 -0
- data/.git/objects/ec/f2f7e20cd14b0d4b5c0b7d0450aafc8c4f7066 +0 -0
- data/.git/objects/f1/ca0626d3e5dfcf88df23c9720d6f770fcbef73 +0 -0
- data/.git/objects/f2/3c73e1110a2fec82d37a1a399b9f274d30f0fe +0 -0
- data/.git/objects/f2/aec2820bf2f1c87c031d4abe262560392ac4b9 +0 -0
- data/.git/objects/f8/73dae66e0208669dbf74718181b3866647f426 +0 -0
- data/.git/refs/heads/master +1 -0
- data/.git/refs/remotes/origin/master +1 -0
- data/History.txt +4 -0
- data/License.txt +20 -0
- data/Manifest.txt +97 -0
- data/README.txt +19 -0
- data/Rakefile +4 -0
- data/config/hoe.rb +70 -0
- data/config/requirements.rb +17 -0
- data/lib/multibyte/chars.rb +135 -0
- data/lib/multibyte/generators/generate_tables.rb +145 -0
- data/lib/multibyte/handlers/passthru_handler.rb +9 -0
- data/lib/multibyte/handlers/utf8_handler.rb +564 -0
- data/lib/multibyte/handlers/utf8_handler_proc.rb +43 -0
- data/lib/multibyte/version.rb +9 -0
- data/lib/multibyte.rb +15 -0
- data/lib/unicode.rb +66 -0
- data/log/debug.log +0 -0
- data/script/destroy +14 -0
- data/script/generate +14 -0
- data/script/txt2html +74 -0
- data/setup.rb +1585 -0
- data/spec/multibyte_spec.rb +11 -0
- data/spec/spec.opts +1 -0
- data/spec/spec_helper.rb +7 -0
- data/tasks/deployment.rake +34 -0
- data/tasks/environment.rake +7 -0
- data/tasks/rspec.rake +21 -0
- data/tasks/website.rake +17 -0
- data/website/index.html +11 -0
- data/website/index.txt +39 -0
- data/website/javascripts/rounded_corners_lite.inc.js +285 -0
- data/website/stylesheets/screen.css +138 -0
- data/website/template.rhtml +48 -0
- data.tar.gz.sig +3 -0
- metadata +176 -0
- metadata.gz.sig +0 -0
@@ -0,0 +1,145 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
require 'open-uri'
|
3
|
+
require 'tmpdir'
|
4
|
+
|
5
|
+
module Multibyte::Handlers #:nodoc:
|
6
|
+
class UnicodeDatabase #:nodoc:
|
7
|
+
def self.load
|
8
|
+
[Hash.new(Codepoint.new),[],{},{}]
|
9
|
+
end
|
10
|
+
end
|
11
|
+
|
12
|
+
class UnicodeTableGenerator #:nodoc:
|
13
|
+
BASE_URI = "http://www.unicode.org/Public/#{Multibyte::UNICODE_VERSION}/ucd/"
|
14
|
+
SOURCES = {
|
15
|
+
:codepoints => BASE_URI + 'UnicodeData.txt',
|
16
|
+
:composition_exclusion => BASE_URI + 'CompositionExclusions.txt',
|
17
|
+
:grapheme_break_property => BASE_URI + 'auxiliary/GraphemeBreakProperty.txt',
|
18
|
+
:cp1252 => 'http://unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP1252.TXT'
|
19
|
+
}
|
20
|
+
|
21
|
+
def initialize
|
22
|
+
@ucd = UnicodeDatabase.new
|
23
|
+
|
24
|
+
default = Codepoint.new
|
25
|
+
default.combining_class = 0
|
26
|
+
default.uppercase_mapping = 0
|
27
|
+
default.lowercase_mapping = 0
|
28
|
+
@ucd.codepoints = Hash.new(default)
|
29
|
+
|
30
|
+
@ucd.composition_exclusion = []
|
31
|
+
@ucd.composition_map = {}
|
32
|
+
@ucd.boundary = {}
|
33
|
+
@ucd.cp1252 = {}
|
34
|
+
end
|
35
|
+
|
36
|
+
def parse_codepoints(line)
|
37
|
+
codepoint = Codepoint.new
|
38
|
+
raise "Could not parse input." unless line =~ /^
|
39
|
+
([0-9A-F]+); # code
|
40
|
+
([^;]+); # name
|
41
|
+
([A-Z]+); # general category
|
42
|
+
([0-9]+); # canonical combining class
|
43
|
+
([A-Z]+); # bidi class
|
44
|
+
(<([A-Z]*)>)? # decomposition type
|
45
|
+
((\ ?[0-9A-F]+)*); # decompomposition mapping
|
46
|
+
([0-9]*); # decimal digit
|
47
|
+
([0-9]*); # digit
|
48
|
+
([^;]*); # numeric
|
49
|
+
([YN]*); # bidi mirrored
|
50
|
+
([^;]*); # unicode 1.0 name
|
51
|
+
([^;]*); # iso comment
|
52
|
+
([0-9A-F]*); # simple uppercase mapping
|
53
|
+
([0-9A-F]*); # simple lowercase mapping
|
54
|
+
([0-9A-F]*)$/ix # simple titlecase mapping
|
55
|
+
codepoint.code = $1.hex
|
56
|
+
#codepoint.name = $2
|
57
|
+
#codepoint.category = $3
|
58
|
+
codepoint.combining_class = Integer($4)
|
59
|
+
#codepoint.bidi_class = $5
|
60
|
+
codepoint.decomp_type = $7
|
61
|
+
codepoint.decomp_mapping = ($8=='') ? nil : $8.split.collect { |element| element.hex }
|
62
|
+
#codepoint.bidi_mirrored = ($13=='Y') ? true : false
|
63
|
+
codepoint.uppercase_mapping = ($16=='') ? 0 : $16.hex
|
64
|
+
codepoint.lowercase_mapping = ($17=='') ? 0 : $17.hex
|
65
|
+
#codepoint.titlecase_mapping = ($18=='') ? nil : $18.hex
|
66
|
+
@ucd.codepoints[codepoint.code] = codepoint
|
67
|
+
end
|
68
|
+
|
69
|
+
def parse_grapheme_break_property(line)
|
70
|
+
if line =~ /^([0-9A-F\.]+)\s*;\s*([\w]+)\s*#/
|
71
|
+
type = $2.downcase.intern
|
72
|
+
@ucd.boundary[type] ||= []
|
73
|
+
if $1.include? '..'
|
74
|
+
parts = $1.split '..'
|
75
|
+
@ucd.boundary[type] << (parts[0].hex..parts[1].hex)
|
76
|
+
else
|
77
|
+
@ucd.boundary[type] << $1.hex
|
78
|
+
end
|
79
|
+
end
|
80
|
+
end
|
81
|
+
|
82
|
+
def parse_composition_exclusion(line)
|
83
|
+
if line =~ /^([0-9A-F]+)/i
|
84
|
+
@ucd.composition_exclusion << $1.hex
|
85
|
+
end
|
86
|
+
end
|
87
|
+
|
88
|
+
def parse_cp1252(line)
|
89
|
+
if line =~ /^([0-9A-Fx]+)\s([0-9A-Fx]+)/i
|
90
|
+
@ucd.cp1252[$1.hex] = $2.hex
|
91
|
+
end
|
92
|
+
end
|
93
|
+
|
94
|
+
def create_composition_map
|
95
|
+
@ucd.codepoints.each do |_, cp|
|
96
|
+
if !cp.nil? and cp.combining_class == 0 and cp.decomp_type.nil? and !cp.decomp_mapping.nil? and cp.decomp_mapping.length == 2 and @ucd[cp.decomp_mapping[0]].combining_class == 0 and !@ucd.composition_exclusion.include?(cp.code)
|
97
|
+
@ucd.composition_map[cp.decomp_mapping[0]] ||= {}
|
98
|
+
@ucd.composition_map[cp.decomp_mapping[0]][cp.decomp_mapping[1]] = cp.code
|
99
|
+
end
|
100
|
+
end
|
101
|
+
end
|
102
|
+
|
103
|
+
def normalize_boundary_map
|
104
|
+
@ucd.boundary.each do |k,v|
|
105
|
+
if [:lf, :cr].include? k
|
106
|
+
@ucd.boundary[k] = v[0]
|
107
|
+
end
|
108
|
+
end
|
109
|
+
end
|
110
|
+
|
111
|
+
def parse
|
112
|
+
SOURCES.each do |type, url|
|
113
|
+
filename = File.join(Dir.tmpdir, "#{url.split('/').last}")
|
114
|
+
unless File.exist?(filename)
|
115
|
+
$stderr.puts "Downloading #{url.split('/').last}"
|
116
|
+
File.open(filename, 'wb') do |target|
|
117
|
+
open(url) do |source|
|
118
|
+
source.each_line { |line| target.write line }
|
119
|
+
end
|
120
|
+
end
|
121
|
+
end
|
122
|
+
File.open(filename) do |file|
|
123
|
+
file.each_line { |line| send "parse_#{type}".intern, line }
|
124
|
+
end
|
125
|
+
end
|
126
|
+
create_composition_map
|
127
|
+
normalize_boundary_map
|
128
|
+
end
|
129
|
+
|
130
|
+
def dump_to(filename)
|
131
|
+
File.open(filename, 'wb') do |f|
|
132
|
+
f.write Marshal.dump([@ucd.codepoints, @ucd.composition_exclusion, @ucd.composition_map, @ucd.boundary, @ucd.cp1252])
|
133
|
+
end
|
134
|
+
end
|
135
|
+
end
|
136
|
+
end
|
137
|
+
|
138
|
+
if __FILE__ == $0
|
139
|
+
filename = Multibyte::Handlers::UnicodeDatabase.filename
|
140
|
+
generator = Multibyte::Handlers::UnicodeTableGenerator.new
|
141
|
+
generator.parse
|
142
|
+
print "Writing to: #{filename}"
|
143
|
+
generator.dump_to filename
|
144
|
+
puts " (#{File.size(filename)} bytes)"
|
145
|
+
end
|
@@ -0,0 +1,9 @@
|
|
1
|
+
# Chars uses this handler when $KCODE is not set to 'UTF8'. Because this handler doesn't define any methods all call
|
2
|
+
# will be forwarded to String.
|
3
|
+
class Multibyte::Handlers::PassthruHandler #:nodoc:
|
4
|
+
|
5
|
+
# Return the original byteoffset
|
6
|
+
def self.translate_offset(string, byte_offset) #:nodoc:
|
7
|
+
byte_offset
|
8
|
+
end
|
9
|
+
end
|
@@ -0,0 +1,564 @@
|
|
1
|
+
# Contains all the handlers and helper classes
|
2
|
+
module Multibyte::Handlers #:nodoc:
|
3
|
+
class EncodingError < ArgumentError #:nodoc:
|
4
|
+
end
|
5
|
+
|
6
|
+
class Codepoint #:nodoc:
|
7
|
+
attr_accessor :code, :combining_class, :decomp_type, :decomp_mapping, :uppercase_mapping, :lowercase_mapping
|
8
|
+
end
|
9
|
+
|
10
|
+
class UnicodeDatabase #:nodoc:
|
11
|
+
attr_writer :codepoints, :composition_exclusion, :composition_map, :boundary, :cp1252
|
12
|
+
|
13
|
+
# self-expiring methods that lazily load the Unicode database and then return the value.
|
14
|
+
[:codepoints, :composition_exclusion, :composition_map, :boundary, :cp1252].each do |attr_name|
|
15
|
+
class_eval(<<-EOS, __FILE__, __LINE__)
|
16
|
+
def #{attr_name}
|
17
|
+
load
|
18
|
+
@#{attr_name}
|
19
|
+
end
|
20
|
+
EOS
|
21
|
+
end
|
22
|
+
|
23
|
+
# Shortcut to ucd.codepoints[]
|
24
|
+
def [](index); codepoints[index]; end
|
25
|
+
|
26
|
+
# Returns the directory in which the data files are stored
|
27
|
+
def self.dirname
|
28
|
+
File.dirname(__FILE__) + '/../../values/'
|
29
|
+
end
|
30
|
+
|
31
|
+
# Returns the filename for the data file for this version
|
32
|
+
def self.filename
|
33
|
+
File.expand_path File.join(dirname, "unicode_tables.dat")
|
34
|
+
end
|
35
|
+
|
36
|
+
# Loads the unicode database and returns all the internal objects of UnicodeDatabase
|
37
|
+
# Once the values have been loaded, define attr_reader methods for the instance variables.
|
38
|
+
def load
|
39
|
+
begin
|
40
|
+
@codepoints, @composition_exclusion, @composition_map, @boundary, @cp1252 = File.open(self.class.filename, 'rb') { |f| Marshal.load f.read }
|
41
|
+
rescue Exception => e
|
42
|
+
raise IOError.new("Couldn't load the unicode tables for UTF8Handler (#{e.message}), handler is unusable")
|
43
|
+
end
|
44
|
+
@codepoints ||= Hash.new(Codepoint.new)
|
45
|
+
@composition_exclusion ||= []
|
46
|
+
@composition_map ||= {}
|
47
|
+
@boundary ||= {}
|
48
|
+
@cp1252 ||= {}
|
49
|
+
|
50
|
+
# Redefine the === method so we can write shorter rules for grapheme cluster breaks
|
51
|
+
@boundary.each do |k,_|
|
52
|
+
@boundary[k].instance_eval do
|
53
|
+
def ===(other)
|
54
|
+
detect { |i| i === other } ? true : false
|
55
|
+
end
|
56
|
+
end if @boundary[k].kind_of?(Array)
|
57
|
+
end
|
58
|
+
|
59
|
+
# define attr_reader methods for the instance variables
|
60
|
+
class << self
|
61
|
+
attr_reader :codepoints, :composition_exclusion, :composition_map, :boundary, :cp1252
|
62
|
+
end
|
63
|
+
end
|
64
|
+
end
|
65
|
+
|
66
|
+
# UTF8Handler implements Unicode aware operations for strings, these operations will be used by the Chars
|
67
|
+
# proxy when $KCODE is set to 'UTF8'.
|
68
|
+
class UTF8Handler
|
69
|
+
# Hangul character boundaries and properties
|
70
|
+
HANGUL_SBASE = 0xAC00
|
71
|
+
HANGUL_LBASE = 0x1100
|
72
|
+
HANGUL_VBASE = 0x1161
|
73
|
+
HANGUL_TBASE = 0x11A7
|
74
|
+
HANGUL_LCOUNT = 19
|
75
|
+
HANGUL_VCOUNT = 21
|
76
|
+
HANGUL_TCOUNT = 28
|
77
|
+
HANGUL_NCOUNT = HANGUL_VCOUNT * HANGUL_TCOUNT
|
78
|
+
HANGUL_SCOUNT = 11172
|
79
|
+
HANGUL_SLAST = HANGUL_SBASE + HANGUL_SCOUNT
|
80
|
+
HANGUL_JAMO_FIRST = 0x1100
|
81
|
+
HANGUL_JAMO_LAST = 0x11FF
|
82
|
+
|
83
|
+
# All the unicode whitespace
|
84
|
+
UNICODE_WHITESPACE = [
|
85
|
+
(0x0009..0x000D).to_a, # White_Space # Cc [5] <control-0009>..<control-000D>
|
86
|
+
0x0020, # White_Space # Zs SPACE
|
87
|
+
0x0085, # White_Space # Cc <control-0085>
|
88
|
+
0x00A0, # White_Space # Zs NO-BREAK SPACE
|
89
|
+
0x1680, # White_Space # Zs OGHAM SPACE MARK
|
90
|
+
0x180E, # White_Space # Zs MONGOLIAN VOWEL SEPARATOR
|
91
|
+
(0x2000..0x200A).to_a, # White_Space # Zs [11] EN QUAD..HAIR SPACE
|
92
|
+
0x2028, # White_Space # Zl LINE SEPARATOR
|
93
|
+
0x2029, # White_Space # Zp PARAGRAPH SEPARATOR
|
94
|
+
0x202F, # White_Space # Zs NARROW NO-BREAK SPACE
|
95
|
+
0x205F, # White_Space # Zs MEDIUM MATHEMATICAL SPACE
|
96
|
+
0x3000, # White_Space # Zs IDEOGRAPHIC SPACE
|
97
|
+
].flatten.freeze
|
98
|
+
|
99
|
+
# BOM (byte order mark) can also be seen as whitespace, it's a non-rendering character used to distinguish
|
100
|
+
# between little and big endian. This is not an issue in utf-8, so it must be ignored.
|
101
|
+
UNICODE_LEADERS_AND_TRAILERS = UNICODE_WHITESPACE + [65279] # ZERO-WIDTH NO-BREAK SPACE aka BOM
|
102
|
+
|
103
|
+
# Borrowed from the Kconv library by Shinji KONO - (also as seen on the W3C site)
|
104
|
+
UTF8_PAT = /\A(?:
|
105
|
+
[\x00-\x7f] |
|
106
|
+
[\xc2-\xdf] [\x80-\xbf] |
|
107
|
+
\xe0 [\xa0-\xbf] [\x80-\xbf] |
|
108
|
+
[\xe1-\xef] [\x80-\xbf] [\x80-\xbf] |
|
109
|
+
\xf0 [\x90-\xbf] [\x80-\xbf] [\x80-\xbf] |
|
110
|
+
[\xf1-\xf3] [\x80-\xbf] [\x80-\xbf] [\x80-\xbf] |
|
111
|
+
\xf4 [\x80-\x8f] [\x80-\xbf] [\x80-\xbf]
|
112
|
+
)*\z/xn
|
113
|
+
|
114
|
+
# Returns a regular expression pattern that matches the passed Unicode codepoints
|
115
|
+
def self.codepoints_to_pattern(array_of_codepoints) #:nodoc:
|
116
|
+
array_of_codepoints.collect{ |e| [e].pack 'U*' }.join('|')
|
117
|
+
end
|
118
|
+
UNICODE_TRAILERS_PAT = /(#{codepoints_to_pattern(UNICODE_LEADERS_AND_TRAILERS)})+\Z/
|
119
|
+
UNICODE_LEADERS_PAT = /\A(#{codepoints_to_pattern(UNICODE_LEADERS_AND_TRAILERS)})+/
|
120
|
+
|
121
|
+
class << self
|
122
|
+
|
123
|
+
# ///
|
124
|
+
# /// BEGIN String method overrides
|
125
|
+
# ///
|
126
|
+
|
127
|
+
# Inserts the passed string at specified codepoint offsets
|
128
|
+
def insert(str, offset, fragment)
|
129
|
+
str.replace(
|
130
|
+
u_unpack(str).insert(
|
131
|
+
offset,
|
132
|
+
u_unpack(fragment)
|
133
|
+
).flatten.pack('U*')
|
134
|
+
)
|
135
|
+
end
|
136
|
+
|
137
|
+
# Returns the position of the passed argument in the string, counting in codepoints
|
138
|
+
def index(str, *args)
|
139
|
+
bidx = str.index(*args)
|
140
|
+
bidx ? (u_unpack(str.slice(0...bidx)).size) : nil
|
141
|
+
end
|
142
|
+
|
143
|
+
# Works just like the indexed replace method on string, except instead of byte offsets you specify
|
144
|
+
# character offsets.
|
145
|
+
#
|
146
|
+
# Example:
|
147
|
+
#
|
148
|
+
# s = "Müller"
|
149
|
+
# s.chars[2] = "e" # Replace character with offset 2
|
150
|
+
# s # => "Müeler"
|
151
|
+
#
|
152
|
+
# s = "Müller"
|
153
|
+
# s.chars[1, 2] = "ö" # Replace 2 characters at character offset 1
|
154
|
+
# s # => "Möler"
|
155
|
+
def []=(str, *args)
|
156
|
+
replace_by = args.pop
|
157
|
+
# Indexed replace with regular expressions already works
|
158
|
+
return str[*args] = replace_by if args.first.is_a?(Regexp)
|
159
|
+
result = u_unpack(str)
|
160
|
+
if args[0].is_a?(Fixnum)
|
161
|
+
raise IndexError, "index #{args[0]} out of string" if args[0] >= result.length
|
162
|
+
min = args[0]
|
163
|
+
max = args[1].nil? ? min : (min + args[1] - 1)
|
164
|
+
range = Range.new(min, max)
|
165
|
+
replace_by = [replace_by].pack('U') if replace_by.is_a?(Fixnum)
|
166
|
+
elsif args.first.is_a?(Range)
|
167
|
+
raise RangeError, "#{args[0]} out of range" if args[0].min >= result.length
|
168
|
+
range = args[0]
|
169
|
+
else
|
170
|
+
needle = args[0].to_s
|
171
|
+
min = index(str, needle)
|
172
|
+
max = min + length(needle) - 1
|
173
|
+
range = Range.new(min, max)
|
174
|
+
end
|
175
|
+
result[range] = u_unpack(replace_by)
|
176
|
+
str.replace(result.pack('U*'))
|
177
|
+
end
|
178
|
+
|
179
|
+
# Works just like String#rjust, only integer specifies characters instead of bytes.
|
180
|
+
#
|
181
|
+
# Example:
|
182
|
+
#
|
183
|
+
# "¾ cup".chars.rjust(8).to_s
|
184
|
+
# # => " ¾ cup"
|
185
|
+
#
|
186
|
+
# "¾ cup".chars.rjust(8, " ").to_s # Use non-breaking whitespace
|
187
|
+
# # => " ¾ cup"
|
188
|
+
def rjust(str, integer, padstr=' ')
|
189
|
+
justify(str, integer, :right, padstr)
|
190
|
+
end
|
191
|
+
|
192
|
+
# Works just like String#ljust, only integer specifies characters instead of bytes.
|
193
|
+
#
|
194
|
+
# Example:
|
195
|
+
#
|
196
|
+
# "¾ cup".chars.rjust(8).to_s
|
197
|
+
# # => "¾ cup "
|
198
|
+
#
|
199
|
+
# "¾ cup".chars.rjust(8, " ").to_s # Use non-breaking whitespace
|
200
|
+
# # => "¾ cup "
|
201
|
+
def ljust(str, integer, padstr=' ')
|
202
|
+
justify(str, integer, :left, padstr)
|
203
|
+
end
|
204
|
+
|
205
|
+
# Works just like String#center, only integer specifies characters instead of bytes.
|
206
|
+
#
|
207
|
+
# Example:
|
208
|
+
#
|
209
|
+
# "¾ cup".chars.center(8).to_s
|
210
|
+
# # => " ¾ cup "
|
211
|
+
#
|
212
|
+
# "¾ cup".chars.center(8, " ").to_s # Use non-breaking whitespace
|
213
|
+
# # => " ¾ cup "
|
214
|
+
def center(str, integer, padstr=' ')
|
215
|
+
justify(str, integer, :center, padstr)
|
216
|
+
end
|
217
|
+
|
218
|
+
# Does Unicode-aware rstrip
|
219
|
+
def rstrip(str)
|
220
|
+
str.gsub(UNICODE_TRAILERS_PAT, '')
|
221
|
+
end
|
222
|
+
|
223
|
+
# Does Unicode-aware lstrip
|
224
|
+
def lstrip(str)
|
225
|
+
str.gsub(UNICODE_LEADERS_PAT, '')
|
226
|
+
end
|
227
|
+
|
228
|
+
# Removed leading and trailing whitespace
|
229
|
+
def strip(str)
|
230
|
+
str.gsub(UNICODE_LEADERS_PAT, '').gsub(UNICODE_TRAILERS_PAT, '')
|
231
|
+
end
|
232
|
+
|
233
|
+
# Returns the number of codepoints in the string
|
234
|
+
def size(str)
|
235
|
+
u_unpack(str).size
|
236
|
+
end
|
237
|
+
alias_method :length, :size
|
238
|
+
|
239
|
+
# Reverses codepoints in the string.
|
240
|
+
def reverse(str)
|
241
|
+
u_unpack(str).reverse.pack('U*')
|
242
|
+
end
|
243
|
+
|
244
|
+
# Implements Unicode-aware slice with codepoints. Slicing on one point returns the codepoints for that
|
245
|
+
# character.
|
246
|
+
def slice(str, *args)
|
247
|
+
if args.size > 2
|
248
|
+
raise ArgumentError, "wrong number of arguments (#{args.size} for 1)" # Do as if we were native
|
249
|
+
elsif (args.size == 2 && !(args.first.is_a?(Numeric) || args.first.is_a?(Regexp)))
|
250
|
+
raise TypeError, "cannot convert #{args.first.class} into Integer" # Do as if we were native
|
251
|
+
elsif (args.size == 2 && !args[1].is_a?(Numeric))
|
252
|
+
raise TypeError, "cannot convert #{args[1].class} into Integer" # Do as if we were native
|
253
|
+
elsif args[0].kind_of? Range
|
254
|
+
cps = u_unpack(str).slice(*args)
|
255
|
+
cps.nil? ? nil : cps.pack('U*')
|
256
|
+
elsif args[0].kind_of? Regexp
|
257
|
+
str.slice(*args)
|
258
|
+
elsif args.size == 1 && args[0].kind_of?(Numeric)
|
259
|
+
u_unpack(str)[args[0]]
|
260
|
+
else
|
261
|
+
u_unpack(str).slice(*args).pack('U*')
|
262
|
+
end
|
263
|
+
end
|
264
|
+
alias_method :[], :slice
|
265
|
+
|
266
|
+
# Convert characters in the string to uppercase
|
267
|
+
def upcase(str); to_case :uppercase_mapping, str; end
|
268
|
+
|
269
|
+
# Convert characters in the string to lowercase
|
270
|
+
def downcase(str); to_case :lowercase_mapping, str; end
|
271
|
+
|
272
|
+
# Returns a copy of +str+ with the first character converted to uppercase and the remainder to lowercase
|
273
|
+
def capitalize(str)
|
274
|
+
upcase(slice(str, 0..0)) + downcase(slice(str, 1..-1) || '')
|
275
|
+
end
|
276
|
+
|
277
|
+
# ///
|
278
|
+
# /// Extra String methods for unicode operations
|
279
|
+
# ///
|
280
|
+
|
281
|
+
# Returns the KC normalization of the string by default. NFKC is considered the best normalization form for
|
282
|
+
# passing strings to databases and validations.
|
283
|
+
#
|
284
|
+
# * <tt>str</tt> - The string to perform normalization on.
|
285
|
+
# * <tt>form</tt> - The form you want to normalize in. Should be one of the following:
|
286
|
+
# <tt>:c</tt>, <tt>:kc</tt>, <tt>:d</tt>, or <tt>:kd</tt>. Default is
|
287
|
+
# ActiveSupport::Multibyte::DEFAULT_NORMALIZATION_FORM.
|
288
|
+
def normalize(str, form=ActiveSupport::Multibyte::DEFAULT_NORMALIZATION_FORM)
|
289
|
+
# See http://www.unicode.org/reports/tr15, Table 1
|
290
|
+
codepoints = u_unpack(str)
|
291
|
+
case form
|
292
|
+
when :d
|
293
|
+
reorder_characters(decompose_codepoints(:canonical, codepoints))
|
294
|
+
when :c
|
295
|
+
compose_codepoints reorder_characters(decompose_codepoints(:canonical, codepoints))
|
296
|
+
when :kd
|
297
|
+
reorder_characters(decompose_codepoints(:compatability, codepoints))
|
298
|
+
when :kc
|
299
|
+
compose_codepoints reorder_characters(decompose_codepoints(:compatability, codepoints))
|
300
|
+
else
|
301
|
+
raise ArgumentError, "#{form} is not a valid normalization variant", caller
|
302
|
+
end.pack('U*')
|
303
|
+
end
|
304
|
+
|
305
|
+
# Perform decomposition on the characters in the string
|
306
|
+
def decompose(str)
|
307
|
+
decompose_codepoints(:canonical, u_unpack(str)).pack('U*')
|
308
|
+
end
|
309
|
+
|
310
|
+
# Perform composition on the characters in the string
|
311
|
+
def compose(str)
|
312
|
+
compose_codepoints u_unpack(str).pack('U*')
|
313
|
+
end
|
314
|
+
|
315
|
+
# ///
|
316
|
+
# /// BEGIN Helper methods for unicode operation
|
317
|
+
# ///
|
318
|
+
|
319
|
+
# Used to translate an offset from bytes to characters, for instance one received from a regular expression match
|
320
|
+
def translate_offset(str, byte_offset)
|
321
|
+
return nil if byte_offset.nil?
|
322
|
+
return 0 if str == ''
|
323
|
+
chunk = str[0..byte_offset]
|
324
|
+
begin
|
325
|
+
begin
|
326
|
+
chunk.unpack('U*').length - 1
|
327
|
+
rescue ArgumentError => e
|
328
|
+
chunk = str[0..(byte_offset+=1)]
|
329
|
+
# Stop retrying at the end of the string
|
330
|
+
raise e unless byte_offset < chunk.length
|
331
|
+
# We damaged a character, retry
|
332
|
+
retry
|
333
|
+
end
|
334
|
+
# Catch the ArgumentError so we can throw our own
|
335
|
+
rescue ArgumentError
|
336
|
+
raise EncodingError.new('malformed UTF-8 character')
|
337
|
+
end
|
338
|
+
end
|
339
|
+
|
340
|
+
# Checks if the string is valid UTF8.
|
341
|
+
def consumes?(str)
|
342
|
+
# Unpack is a little bit faster than regular expressions
|
343
|
+
begin
|
344
|
+
str.unpack('U*')
|
345
|
+
true
|
346
|
+
rescue ArgumentError
|
347
|
+
false
|
348
|
+
end
|
349
|
+
end
|
350
|
+
|
351
|
+
# Returns the number of grapheme clusters in the string. This method is very likely to be moved or renamed
|
352
|
+
# in future versions.
|
353
|
+
def g_length(str)
|
354
|
+
g_unpack(str).length
|
355
|
+
end
|
356
|
+
|
357
|
+
# Replaces all the non-utf-8 bytes by their iso-8859-1 or cp1252 equivalent resulting in a valid utf-8 string
|
358
|
+
def tidy_bytes(str)
|
359
|
+
str.split(//u).map do |c|
|
360
|
+
if !UTF8_PAT.match(c)
|
361
|
+
n = c.unpack('C')[0]
|
362
|
+
n < 128 ? n.chr :
|
363
|
+
n < 160 ? [UCD.cp1252[n] || n].pack('U') :
|
364
|
+
n < 192 ? "\xC2" + n.chr : "\xC3" + (n-64).chr
|
365
|
+
else
|
366
|
+
c
|
367
|
+
end
|
368
|
+
end.join
|
369
|
+
end
|
370
|
+
|
371
|
+
protected
|
372
|
+
|
373
|
+
# Detect whether the codepoint is in a certain character class. Primarily used by the
|
374
|
+
# grapheme cluster support.
|
375
|
+
def in_char_class?(codepoint, classes)
|
376
|
+
classes.detect { |c| UCD.boundary[c] === codepoint } ? true : false
|
377
|
+
end
|
378
|
+
|
379
|
+
# Unpack the string at codepoints boundaries
|
380
|
+
def u_unpack(str)
|
381
|
+
begin
|
382
|
+
str.unpack 'U*'
|
383
|
+
rescue ArgumentError
|
384
|
+
raise EncodingError.new('malformed UTF-8 character')
|
385
|
+
end
|
386
|
+
end
|
387
|
+
|
388
|
+
# Unpack the string at grapheme boundaries instead of codepoint boundaries
|
389
|
+
def g_unpack(str)
|
390
|
+
codepoints = u_unpack(str)
|
391
|
+
unpacked = []
|
392
|
+
pos = 0
|
393
|
+
marker = 0
|
394
|
+
eoc = codepoints.length
|
395
|
+
while(pos < eoc)
|
396
|
+
pos += 1
|
397
|
+
previous = codepoints[pos-1]
|
398
|
+
current = codepoints[pos]
|
399
|
+
if (
|
400
|
+
# CR X LF
|
401
|
+
one = ( previous == UCD.boundary[:cr] and current == UCD.boundary[:lf] ) or
|
402
|
+
# L X (L|V|LV|LVT)
|
403
|
+
two = ( UCD.boundary[:l] === previous and in_char_class?(current, [:l,:v,:lv,:lvt]) ) or
|
404
|
+
# (LV|V) X (V|T)
|
405
|
+
three = ( in_char_class?(previous, [:lv,:v]) and in_char_class?(current, [:v,:t]) ) or
|
406
|
+
# (LVT|T) X (T)
|
407
|
+
four = ( in_char_class?(previous, [:lvt,:t]) and UCD.boundary[:t] === current ) or
|
408
|
+
# X Extend
|
409
|
+
five = (UCD.boundary[:extend] === current)
|
410
|
+
)
|
411
|
+
else
|
412
|
+
unpacked << codepoints[marker..pos-1]
|
413
|
+
marker = pos
|
414
|
+
end
|
415
|
+
end
|
416
|
+
unpacked
|
417
|
+
end
|
418
|
+
|
419
|
+
# Reverse operation of g_unpack
|
420
|
+
def g_pack(unpacked)
|
421
|
+
unpacked.flatten
|
422
|
+
end
|
423
|
+
|
424
|
+
# Justifies a string in a certain way. Valid values for <tt>way</tt> are <tt>:right</tt>, <tt>:left</tt> and
|
425
|
+
# <tt>:center</tt>. Is primarily used as a helper method by <tt>rjust</tt>, <tt>ljust</tt> and <tt>center</tt>.
|
426
|
+
def justify(str, integer, way, padstr=' ')
|
427
|
+
raise ArgumentError, "zero width padding" if padstr.length == 0
|
428
|
+
padsize = integer - size(str)
|
429
|
+
padsize = padsize > 0 ? padsize : 0
|
430
|
+
case way
|
431
|
+
when :right
|
432
|
+
str.dup.insert(0, padding(padsize, padstr))
|
433
|
+
when :left
|
434
|
+
str.dup.insert(-1, padding(padsize, padstr))
|
435
|
+
when :center
|
436
|
+
lpad = padding((padsize / 2.0).floor, padstr)
|
437
|
+
rpad = padding((padsize / 2.0).ceil, padstr)
|
438
|
+
str.dup.insert(0, lpad).insert(-1, rpad)
|
439
|
+
end
|
440
|
+
end
|
441
|
+
|
442
|
+
# Generates a padding string of a certain size.
|
443
|
+
def padding(padsize, padstr=' ')
|
444
|
+
if padsize != 0
|
445
|
+
slice(padstr * ((padsize / size(padstr)) + 1), 0, padsize)
|
446
|
+
else
|
447
|
+
''
|
448
|
+
end
|
449
|
+
end
|
450
|
+
|
451
|
+
# Convert characters to a different case
|
452
|
+
def to_case(way, str)
|
453
|
+
u_unpack(str).map do |codepoint|
|
454
|
+
cp = UCD[codepoint]
|
455
|
+
unless cp.nil?
|
456
|
+
ncp = cp.send(way)
|
457
|
+
ncp > 0 ? ncp : codepoint
|
458
|
+
else
|
459
|
+
codepoint
|
460
|
+
end
|
461
|
+
end.pack('U*')
|
462
|
+
end
|
463
|
+
|
464
|
+
# Re-order codepoints so the string becomes canonical
|
465
|
+
def reorder_characters(codepoints)
|
466
|
+
length = codepoints.length- 1
|
467
|
+
pos = 0
|
468
|
+
while pos < length do
|
469
|
+
cp1, cp2 = UCD[codepoints[pos]], UCD[codepoints[pos+1]]
|
470
|
+
if (cp1.combining_class > cp2.combining_class) && (cp2.combining_class > 0)
|
471
|
+
codepoints[pos..pos+1] = cp2.code, cp1.code
|
472
|
+
pos += (pos > 0 ? -1 : 1)
|
473
|
+
else
|
474
|
+
pos += 1
|
475
|
+
end
|
476
|
+
end
|
477
|
+
codepoints
|
478
|
+
end
|
479
|
+
|
480
|
+
# Decompose composed characters to the decomposed form
|
481
|
+
def decompose_codepoints(type, codepoints)
|
482
|
+
codepoints.inject([]) do |decomposed, cp|
|
483
|
+
# if it's a hangul syllable starter character
|
484
|
+
if HANGUL_SBASE <= cp and cp < HANGUL_SLAST
|
485
|
+
sindex = cp - HANGUL_SBASE
|
486
|
+
ncp = [] # new codepoints
|
487
|
+
ncp << HANGUL_LBASE + sindex / HANGUL_NCOUNT
|
488
|
+
ncp << HANGUL_VBASE + (sindex % HANGUL_NCOUNT) / HANGUL_TCOUNT
|
489
|
+
tindex = sindex % HANGUL_TCOUNT
|
490
|
+
ncp << (HANGUL_TBASE + tindex) unless tindex == 0
|
491
|
+
decomposed.concat ncp
|
492
|
+
# if the codepoint is decomposable in with the current decomposition type
|
493
|
+
elsif (ncp = UCD[cp].decomp_mapping) and (!UCD[cp].decomp_type || type == :compatability)
|
494
|
+
decomposed.concat decompose_codepoints(type, ncp.dup)
|
495
|
+
else
|
496
|
+
decomposed << cp
|
497
|
+
end
|
498
|
+
end
|
499
|
+
end
|
500
|
+
|
501
|
+
# Compose decomposed characters to the composed form
|
502
|
+
def compose_codepoints(codepoints)
|
503
|
+
pos = 0
|
504
|
+
eoa = codepoints.length - 1
|
505
|
+
starter_pos = 0
|
506
|
+
starter_char = codepoints[0]
|
507
|
+
previous_combining_class = -1
|
508
|
+
while pos < eoa
|
509
|
+
pos += 1
|
510
|
+
lindex = starter_char - HANGUL_LBASE
|
511
|
+
# -- Hangul
|
512
|
+
if 0 <= lindex and lindex < HANGUL_LCOUNT
|
513
|
+
vindex = codepoints[starter_pos+1] - HANGUL_VBASE rescue vindex = -1
|
514
|
+
if 0 <= vindex and vindex < HANGUL_VCOUNT
|
515
|
+
tindex = codepoints[starter_pos+2] - HANGUL_TBASE rescue tindex = -1
|
516
|
+
if 0 <= tindex and tindex < HANGUL_TCOUNT
|
517
|
+
j = starter_pos + 2
|
518
|
+
eoa -= 2
|
519
|
+
else
|
520
|
+
tindex = 0
|
521
|
+
j = starter_pos + 1
|
522
|
+
eoa -= 1
|
523
|
+
end
|
524
|
+
codepoints[starter_pos..j] = (lindex * HANGUL_VCOUNT + vindex) * HANGUL_TCOUNT + tindex + HANGUL_SBASE
|
525
|
+
end
|
526
|
+
starter_pos += 1
|
527
|
+
starter_char = codepoints[starter_pos]
|
528
|
+
# -- Other characters
|
529
|
+
else
|
530
|
+
current_char = codepoints[pos]
|
531
|
+
current = UCD[current_char]
|
532
|
+
if current.combining_class > previous_combining_class
|
533
|
+
if ref = UCD.composition_map[starter_char]
|
534
|
+
composition = ref[current_char]
|
535
|
+
else
|
536
|
+
composition = nil
|
537
|
+
end
|
538
|
+
unless composition.nil?
|
539
|
+
codepoints[starter_pos] = composition
|
540
|
+
starter_char = composition
|
541
|
+
codepoints.delete_at pos
|
542
|
+
eoa -= 1
|
543
|
+
pos -= 1
|
544
|
+
previous_combining_class = -1
|
545
|
+
else
|
546
|
+
previous_combining_class = current.combining_class
|
547
|
+
end
|
548
|
+
else
|
549
|
+
previous_combining_class = current.combining_class
|
550
|
+
end
|
551
|
+
if current.combining_class == 0
|
552
|
+
starter_pos = pos
|
553
|
+
starter_char = codepoints[pos]
|
554
|
+
end
|
555
|
+
end
|
556
|
+
end
|
557
|
+
codepoints
|
558
|
+
end
|
559
|
+
|
560
|
+
# UniCode Database
|
561
|
+
UCD = UnicodeDatabase.new
|
562
|
+
end
|
563
|
+
end
|
564
|
+
end
|