bioinform 0.1.11 → 0.1.12

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (27) hide show
  1. checksums.yaml +4 -4
  2. data/Gemfile +1 -1
  3. data/TODO.txt +2 -0
  4. data/bioinform.gemspec +0 -6
  5. data/lib/bioinform/data_models/motif.rb +1 -1
  6. data/lib/bioinform/data_models/pm.rb +2 -1
  7. data/lib/bioinform/data_models/pwm.rb +4 -5
  8. data/lib/bioinform/support/multiline_squish.rb +1 -1
  9. data/lib/bioinform/support/third_part/active_support/core_ext/array/extract_options.rb +29 -0
  10. data/lib/bioinform/support/third_part/active_support/core_ext/hash/indifferent_access.rb +23 -0
  11. data/lib/bioinform/support/third_part/active_support/core_ext/hash/keys.rb +54 -0
  12. data/lib/bioinform/support/third_part/active_support/core_ext/module/attribute_accessors.rb +64 -0
  13. data/lib/bioinform/support/third_part/active_support/core_ext/object/try.rb +57 -0
  14. data/lib/bioinform/support/third_part/active_support/core_ext/string/access.rb +99 -0
  15. data/lib/bioinform/support/third_part/active_support/core_ext/string/behavior.rb +6 -0
  16. data/lib/bioinform/support/third_part/active_support/core_ext/string/filters.rb +49 -0
  17. data/lib/bioinform/support/third_part/active_support/core_ext/string/multibyte.rb +72 -0
  18. data/lib/bioinform/support/third_part/active_support/hash_with_indifferent_access.rb +178 -0
  19. data/lib/bioinform/support/third_part/active_support/multibyte/chars.rb +476 -0
  20. data/lib/bioinform/support/third_part/active_support/multibyte/exceptions.rb +8 -0
  21. data/lib/bioinform/support/third_part/active_support/multibyte/unicode.rb +393 -0
  22. data/lib/bioinform/support/third_part/active_support/multibyte/utils.rb +60 -0
  23. data/lib/bioinform/support/third_part/active_support/multibyte.rb +44 -0
  24. data/lib/bioinform/support.rb +2 -2
  25. data/lib/bioinform/version.rb +1 -1
  26. data/spec/spec_helper.rb +10 -20
  27. metadata +21 -77
@@ -0,0 +1,178 @@
1
+ require_relative 'core_ext/hash/keys'
2
+
3
+ # This class has dubious semantics and we only have it so that
4
+ # people can write <tt>params[:key]</tt> instead of <tt>params['key']</tt>
5
+ # and they get the same value for both keys.
6
+
7
+ module ActiveSupport
8
+ class HashWithIndifferentAccess < Hash
9
+
10
+ # Always returns true, so that <tt>Array#extract_options!</tt> finds members of this class.
11
+ def extractable_options?
12
+ true
13
+ end
14
+
15
+ def with_indifferent_access
16
+ dup
17
+ end
18
+
19
+ def nested_under_indifferent_access
20
+ self
21
+ end
22
+
23
+ def initialize(constructor = {})
24
+ if constructor.is_a?(Hash)
25
+ super()
26
+ update(constructor)
27
+ else
28
+ super(constructor)
29
+ end
30
+ end
31
+
32
+ def default(key = nil)
33
+ if key.is_a?(Symbol) && include?(key = key.to_s)
34
+ self[key]
35
+ else
36
+ super
37
+ end
38
+ end
39
+
40
+ def self.new_from_hash_copying_default(hash)
41
+ new(hash).tap do |new_hash|
42
+ new_hash.default = hash.default
43
+ end
44
+ end
45
+
46
+ alias_method :regular_writer, :[]= unless method_defined?(:regular_writer)
47
+ alias_method :regular_update, :update unless method_defined?(:regular_update)
48
+
49
+ # Assigns a new value to the hash:
50
+ #
51
+ # hash = HashWithIndifferentAccess.new
52
+ # hash[:key] = "value"
53
+ #
54
+ def []=(key, value)
55
+ regular_writer(convert_key(key), convert_value(value))
56
+ end
57
+
58
+ alias_method :store, :[]=
59
+
60
+ # Updates the instantized hash with values from the second:
61
+ #
62
+ # hash_1 = HashWithIndifferentAccess.new
63
+ # hash_1[:key] = "value"
64
+ #
65
+ # hash_2 = HashWithIndifferentAccess.new
66
+ # hash_2[:key] = "New Value!"
67
+ #
68
+ # hash_1.update(hash_2) # => {"key"=>"New Value!"}
69
+ #
70
+ def update(other_hash)
71
+ if other_hash.is_a? HashWithIndifferentAccess
72
+ super(other_hash)
73
+ else
74
+ other_hash.each_pair { |key, value| regular_writer(convert_key(key), convert_value(value)) }
75
+ self
76
+ end
77
+ end
78
+
79
+ alias_method :merge!, :update
80
+
81
+ # Checks the hash for a key matching the argument passed in:
82
+ #
83
+ # hash = HashWithIndifferentAccess.new
84
+ # hash["key"] = "value"
85
+ # hash.key? :key # => true
86
+ # hash.key? "key" # => true
87
+ #
88
+ def key?(key)
89
+ super(convert_key(key))
90
+ end
91
+
92
+ alias_method :include?, :key?
93
+ alias_method :has_key?, :key?
94
+ alias_method :member?, :key?
95
+
96
+ # Same as <tt>Hash#fetch</tt> where the key passed as argument can be
97
+ # either a string or a symbol:
98
+ #
99
+ # counters = HashWithIndifferentAccess.new
100
+ # counters[:foo] = 1
101
+ #
102
+ # counters.fetch("foo") # => 1
103
+ # counters.fetch(:bar, 0) # => 0
104
+ # counters.fetch(:bar) {|key| 0} # => 0
105
+ # counters.fetch(:zoo) # => KeyError: key not found: "zoo"
106
+ #
107
+ def fetch(key, *extras)
108
+ super(convert_key(key), *extras)
109
+ end
110
+
111
+ # Returns an array of the values at the specified indices:
112
+ #
113
+ # hash = HashWithIndifferentAccess.new
114
+ # hash[:a] = "x"
115
+ # hash[:b] = "y"
116
+ # hash.values_at("a", "b") # => ["x", "y"]
117
+ #
118
+ def values_at(*indices)
119
+ indices.collect {|key| self[convert_key(key)]}
120
+ end
121
+
122
+ # Returns an exact copy of the hash.
123
+ def dup
124
+ self.class.new(self).tap do |new_hash|
125
+ new_hash.default = default
126
+ end
127
+ end
128
+
129
+ # Merges the instantized and the specified hashes together, giving precedence to the values from the second hash.
130
+ # Does not overwrite the existing hash.
131
+ def merge(hash)
132
+ self.dup.update(hash)
133
+ end
134
+
135
+ # Performs the opposite of merge, with the keys and values from the first hash taking precedence over the second.
136
+ # This overloaded definition prevents returning a regular hash, if reverse_merge is called on a <tt>HashWithDifferentAccess</tt>.
137
+ def reverse_merge(other_hash)
138
+ super self.class.new_from_hash_copying_default(other_hash)
139
+ end
140
+
141
+ def reverse_merge!(other_hash)
142
+ replace(reverse_merge( other_hash ))
143
+ end
144
+
145
+ # Removes a specified key from the hash.
146
+ def delete(key)
147
+ super(convert_key(key))
148
+ end
149
+
150
+ def stringify_keys!; self end
151
+ def stringify_keys; dup end
152
+ undef :symbolize_keys!
153
+ def symbolize_keys; to_hash.symbolize_keys end
154
+ def to_options!; self end
155
+
156
+ # Convert to a Hash with String keys.
157
+ def to_hash
158
+ Hash.new(default).merge!(self)
159
+ end
160
+
161
+ protected
162
+ def convert_key(key)
163
+ key.kind_of?(Symbol) ? key.to_s : key
164
+ end
165
+
166
+ def convert_value(value)
167
+ if value.is_a? Hash
168
+ value.nested_under_indifferent_access
169
+ elsif value.is_a?(Array)
170
+ value.dup.replace(value.map { |e| convert_value(e) })
171
+ else
172
+ value
173
+ end
174
+ end
175
+ end
176
+ end
177
+
178
+ HashWithIndifferentAccess = ActiveSupport::HashWithIndifferentAccess
@@ -0,0 +1,476 @@
1
+ # encoding: utf-8
2
+ require_relative '../core_ext/string/access'
3
+ require_relative '../core_ext/string/behavior'
4
+
5
+ module ActiveSupport #:nodoc:
6
+ module Multibyte #:nodoc:
7
+ # Chars enables you to work transparently with UTF-8 encoding in the Ruby String class without having extensive
8
+ # knowledge about the encoding. A Chars object accepts a string upon initialization and proxies String methods in an
9
+ # encoding safe manner. All the normal String methods are also implemented on the proxy.
10
+ #
11
+ # String methods are proxied through the Chars object, and can be accessed through the +mb_chars+ method. Methods
12
+ # which would normally return a String object now return a Chars object so methods can be chained.
13
+ #
14
+ # "The Perfect String ".mb_chars.downcase.strip.normalize # => "the perfect string"
15
+ #
16
+ # Chars objects are perfectly interchangeable with String objects as long as no explicit class checks are made.
17
+ # If certain methods do explicitly check the class, call +to_s+ before you pass chars objects to them.
18
+ #
19
+ # bad.explicit_checking_method "T".mb_chars.downcase.to_s
20
+ #
21
+ # The default Chars implementation assumes that the encoding of the string is UTF-8, if you want to handle different
22
+ # encodings you can write your own multibyte string handler and configure it through
23
+ # ActiveSupport::Multibyte.proxy_class.
24
+ #
25
+ # class CharsForUTF32
26
+ # def size
27
+ # @wrapped_string.size / 4
28
+ # end
29
+ #
30
+ # def self.accepts?(string)
31
+ # string.length % 4 == 0
32
+ # end
33
+ # end
34
+ #
35
+ # ActiveSupport::Multibyte.proxy_class = CharsForUTF32
36
+ class Chars
37
+ attr_reader :wrapped_string
38
+ alias to_s wrapped_string
39
+ alias to_str wrapped_string
40
+
41
+ if RUBY_VERSION >= "1.9"
42
+ # Creates a new Chars instance by wrapping _string_.
43
+ def initialize(string)
44
+ @wrapped_string = string
45
+ @wrapped_string.force_encoding(Encoding::UTF_8) unless @wrapped_string.frozen?
46
+ end
47
+ else
48
+ def initialize(string) #:nodoc:
49
+ @wrapped_string = string
50
+ end
51
+ end
52
+
53
+ # Forward all undefined methods to the wrapped string.
54
+ def method_missing(method, *args, &block)
55
+ if method.to_s =~ /!$/
56
+ @wrapped_string.__send__(method, *args, &block)
57
+ self
58
+ else
59
+ result = @wrapped_string.__send__(method, *args, &block)
60
+ result.kind_of?(String) ? chars(result) : result
61
+ end
62
+ end
63
+
64
+ # Returns +true+ if _obj_ responds to the given method. Private methods are included in the search
65
+ # only if the optional second parameter evaluates to +true+.
66
+ def respond_to?(method, include_private=false)
67
+ super || @wrapped_string.respond_to?(method, include_private)
68
+ end
69
+
70
+ # Enable more predictable duck-typing on String-like classes. See Object#acts_like?.
71
+ def acts_like_string?
72
+ true
73
+ end
74
+
75
+ # Returns +true+ when the proxy class can handle the string. Returns +false+ otherwise.
76
+ def self.consumes?(string)
77
+ # Unpack is a little bit faster than regular expressions.
78
+ string.unpack('U*')
79
+ true
80
+ rescue ArgumentError
81
+ false
82
+ end
83
+
84
+ include Comparable
85
+
86
+ # Returns -1, 0, or 1, depending on whether the Chars object is to be sorted before,
87
+ # equal or after the object on the right side of the operation. It accepts any object
88
+ # that implements +to_s+:
89
+ #
90
+ # 'é'.mb_chars <=> 'ü'.mb_chars # => -1
91
+ #
92
+ # See <tt>String#<=></tt> for more details.
93
+ def <=>(other)
94
+ @wrapped_string <=> other.to_s
95
+ end
96
+
97
+ if RUBY_VERSION < "1.9"
98
+ # Returns +true+ if the Chars class can and should act as a proxy for the string _string_. Returns
99
+ # +false+ otherwise.
100
+ def self.wants?(string)
101
+ $KCODE == 'UTF8' && consumes?(string)
102
+ end
103
+
104
+ # Returns a new Chars object containing the _other_ object concatenated to the string.
105
+ #
106
+ # Example:
107
+ # ('Café'.mb_chars + ' périferôl').to_s # => "Café périferôl"
108
+ def +(other)
109
+ chars(@wrapped_string + other)
110
+ end
111
+
112
+ # Like <tt>String#=~</tt> only it returns the character offset (in codepoints) instead of the byte offset.
113
+ #
114
+ # Example:
115
+ # 'Café périferôl'.mb_chars =~ /ô/ # => 12
116
+ def =~(other)
117
+ translate_offset(@wrapped_string =~ other)
118
+ end
119
+
120
+ # Inserts the passed string at specified codepoint offsets.
121
+ #
122
+ # Example:
123
+ # 'Café'.mb_chars.insert(4, ' périferôl').to_s # => "Café périferôl"
124
+ def insert(offset, fragment)
125
+ unpacked = Unicode.u_unpack(@wrapped_string)
126
+ unless offset > unpacked.length
127
+ @wrapped_string.replace(
128
+ Unicode.u_unpack(@wrapped_string).insert(offset, *Unicode.u_unpack(fragment)).pack('U*')
129
+ )
130
+ else
131
+ raise IndexError, "index #{offset} out of string"
132
+ end
133
+ self
134
+ end
135
+
136
+ # Returns +true+ if contained string contains _other_. Returns +false+ otherwise.
137
+ #
138
+ # Example:
139
+ # 'Café'.mb_chars.include?('é') # => true
140
+ def include?(other)
141
+ # We have to redefine this method because Enumerable defines it.
142
+ @wrapped_string.include?(other)
143
+ end
144
+
145
+ # Returns the position _needle_ in the string, counting in codepoints. Returns +nil+ if _needle_ isn't found.
146
+ #
147
+ # Example:
148
+ # 'Café périferôl'.mb_chars.index('ô') # => 12
149
+ # 'Café périferôl'.mb_chars.index(/\w/u) # => 0
150
+ def index(needle, offset=0)
151
+ wrapped_offset = first(offset).wrapped_string.length
152
+ index = @wrapped_string.index(needle, wrapped_offset)
153
+ index ? (Unicode.u_unpack(@wrapped_string.slice(0...index)).size) : nil
154
+ end
155
+
156
+ # Returns the position _needle_ in the string, counting in
157
+ # codepoints, searching backward from _offset_ or the end of the
158
+ # string. Returns +nil+ if _needle_ isn't found.
159
+ #
160
+ # Example:
161
+ # 'Café périferôl'.mb_chars.rindex('é') # => 6
162
+ # 'Café périferôl'.mb_chars.rindex(/\w/u) # => 13
163
+ def rindex(needle, offset=nil)
164
+ offset ||= length
165
+ wrapped_offset = first(offset).wrapped_string.length
166
+ index = @wrapped_string.rindex(needle, wrapped_offset)
167
+ index ? (Unicode.u_unpack(@wrapped_string.slice(0...index)).size) : nil
168
+ end
169
+
170
+ # Returns the number of codepoints in the string
171
+ def size
172
+ Unicode.u_unpack(@wrapped_string).size
173
+ end
174
+ alias_method :length, :size
175
+
176
+ # Strips entire range of Unicode whitespace from the right of the string.
177
+ def rstrip
178
+ chars(@wrapped_string.gsub(Unicode::TRAILERS_PAT, ''))
179
+ end
180
+
181
+ # Strips entire range of Unicode whitespace from the left of the string.
182
+ def lstrip
183
+ chars(@wrapped_string.gsub(Unicode::LEADERS_PAT, ''))
184
+ end
185
+
186
+ # Strips entire range of Unicode whitespace from the right and left of the string.
187
+ def strip
188
+ rstrip.lstrip
189
+ end
190
+
191
+ # Returns the codepoint of the first character in the string.
192
+ #
193
+ # Example:
194
+ # 'こんにちは'.mb_chars.ord # => 12371
195
+ def ord
196
+ Unicode.u_unpack(@wrapped_string)[0]
197
+ end
198
+
199
+ # Works just like <tt>String#rjust</tt>, only integer specifies characters instead of bytes.
200
+ #
201
+ # Example:
202
+ #
203
+ # "¾ cup".mb_chars.rjust(8).to_s
204
+ # # => " ¾ cup"
205
+ #
206
+ # "¾ cup".mb_chars.rjust(8, " ").to_s # Use non-breaking whitespace
207
+ # # => "   ¾ cup"
208
+ def rjust(integer, padstr=' ')
209
+ justify(integer, :right, padstr)
210
+ end
211
+
212
+ # Works just like <tt>String#ljust</tt>, only integer specifies characters instead of bytes.
213
+ #
214
+ # Example:
215
+ #
216
+ # "¾ cup".mb_chars.rjust(8).to_s
217
+ # # => "¾ cup "
218
+ #
219
+ # "¾ cup".mb_chars.rjust(8, " ").to_s # Use non-breaking whitespace
220
+ # # => "¾ cup   "
221
+ def ljust(integer, padstr=' ')
222
+ justify(integer, :left, padstr)
223
+ end
224
+
225
+ # Works just like <tt>String#center</tt>, only integer specifies characters instead of bytes.
226
+ #
227
+ # Example:
228
+ #
229
+ # "¾ cup".mb_chars.center(8).to_s
230
+ # # => " ¾ cup "
231
+ #
232
+ # "¾ cup".mb_chars.center(8, " ").to_s # Use non-breaking whitespace
233
+ # # => " ¾ cup  "
234
+ def center(integer, padstr=' ')
235
+ justify(integer, :center, padstr)
236
+ end
237
+
238
+ else
239
+ def =~(other)
240
+ @wrapped_string =~ other
241
+ end
242
+ end
243
+
244
+ # Works just like <tt>String#split</tt>, with the exception that the items in the resulting list are Chars
245
+ # instances instead of String. This makes chaining methods easier.
246
+ #
247
+ # Example:
248
+ # 'Café périferôl'.mb_chars.split(/é/).map { |part| part.upcase.to_s } # => ["CAF", " P", "RIFERÔL"]
249
+ def split(*args)
250
+ @wrapped_string.split(*args).map { |i| i.mb_chars }
251
+ end
252
+
253
+ # Like <tt>String#[]=</tt>, except instead of byte offsets you specify character offsets.
254
+ #
255
+ # Example:
256
+ #
257
+ # s = "Müller"
258
+ # s.mb_chars[2] = "e" # Replace character with offset 2
259
+ # s
260
+ # # => "Müeler"
261
+ #
262
+ # s = "Müller"
263
+ # s.mb_chars[1, 2] = "ö" # Replace 2 characters at character offset 1
264
+ # s
265
+ # # => "Möler"
266
+ def []=(*args)
267
+ replace_by = args.pop
268
+ # Indexed replace with regular expressions already works
269
+ if args.first.is_a?(Regexp)
270
+ @wrapped_string[*args] = replace_by
271
+ else
272
+ result = Unicode.u_unpack(@wrapped_string)
273
+ case args.first
274
+ when Fixnum
275
+ raise IndexError, "index #{args[0]} out of string" if args[0] >= result.length
276
+ min = args[0]
277
+ max = args[1].nil? ? min : (min + args[1] - 1)
278
+ range = Range.new(min, max)
279
+ replace_by = [replace_by].pack('U') if replace_by.is_a?(Fixnum)
280
+ when Range
281
+ raise RangeError, "#{args[0]} out of range" if args[0].min >= result.length
282
+ range = args[0]
283
+ else
284
+ needle = args[0].to_s
285
+ min = index(needle)
286
+ max = min + Unicode.u_unpack(needle).length - 1
287
+ range = Range.new(min, max)
288
+ end
289
+ result[range] = Unicode.u_unpack(replace_by)
290
+ @wrapped_string.replace(result.pack('U*'))
291
+ end
292
+ end
293
+
294
+ # Reverses all characters in the string.
295
+ #
296
+ # Example:
297
+ # 'Café'.mb_chars.reverse.to_s # => 'éfaC'
298
+ def reverse
299
+ chars(Unicode.g_unpack(@wrapped_string).reverse.flatten.pack('U*'))
300
+ end
301
+
302
+ # Implements Unicode-aware slice with codepoints. Slicing on one point returns the codepoints for that
303
+ # character.
304
+ #
305
+ # Example:
306
+ # 'こんにちは'.mb_chars.slice(2..3).to_s # => "にち"
307
+ def slice(*args)
308
+ if args.size > 2
309
+ raise ArgumentError, "wrong number of arguments (#{args.size} for 1)" # Do as if we were native
310
+ elsif (args.size == 2 && !(args.first.is_a?(Numeric) || args.first.is_a?(Regexp)))
311
+ raise TypeError, "cannot convert #{args.first.class} into Integer" # Do as if we were native
312
+ elsif (args.size == 2 && !args[1].is_a?(Numeric))
313
+ raise TypeError, "cannot convert #{args[1].class} into Integer" # Do as if we were native
314
+ elsif args[0].kind_of? Range
315
+ cps = Unicode.u_unpack(@wrapped_string).slice(*args)
316
+ result = cps.nil? ? nil : cps.pack('U*')
317
+ elsif args[0].kind_of? Regexp
318
+ result = @wrapped_string.slice(*args)
319
+ elsif args.size == 1 && args[0].kind_of?(Numeric)
320
+ character = Unicode.u_unpack(@wrapped_string)[args[0]]
321
+ result = character && [character].pack('U')
322
+ else
323
+ cps = Unicode.u_unpack(@wrapped_string).slice(*args)
324
+ result = cps && cps.pack('U*')
325
+ end
326
+ result && chars(result)
327
+ end
328
+ alias_method :[], :slice
329
+
330
+ # Limit the byte size of the string to a number of bytes without breaking characters. Usable
331
+ # when the storage for a string is limited for some reason.
332
+ #
333
+ # Example:
334
+ # 'こんにちは'.mb_chars.limit(7).to_s # => "こん"
335
+ def limit(limit)
336
+ slice(0...translate_offset(limit))
337
+ end
338
+
339
+ # Convert characters in the string to uppercase.
340
+ #
341
+ # Example:
342
+ # 'Laurent, où sont les tests ?'.mb_chars.upcase.to_s # => "LAURENT, OÙ SONT LES TESTS ?"
343
+ def upcase
344
+ chars(Unicode.apply_mapping @wrapped_string, :uppercase_mapping)
345
+ end
346
+
347
+ # Convert characters in the string to lowercase.
348
+ #
349
+ # Example:
350
+ # 'VĚDA A VÝZKUM'.mb_chars.downcase.to_s # => "věda a výzkum"
351
+ def downcase
352
+ chars(Unicode.apply_mapping @wrapped_string, :lowercase_mapping)
353
+ end
354
+
355
+ # Converts the first character to uppercase and the remainder to lowercase.
356
+ #
357
+ # Example:
358
+ # 'über'.mb_chars.capitalize.to_s # => "Über"
359
+ def capitalize
360
+ (slice(0) || chars('')).upcase + (slice(1..-1) || chars('')).downcase
361
+ end
362
+
363
+ # Capitalizes the first letter of every word, when possible.
364
+ #
365
+ # Example:
366
+ # "ÉL QUE SE ENTERÓ".mb_chars.titleize # => "Él Que Se Enteró"
367
+ # "日本語".mb_chars.titleize # => "日本語"
368
+ def titleize
369
+ chars(downcase.to_s.gsub(/\b('?[\S])/u) { Unicode.apply_mapping $1, :uppercase_mapping })
370
+ end
371
+ alias_method :titlecase, :titleize
372
+
373
+ # Returns the KC normalization of the string by default. NFKC is considered the best normalization form for
374
+ # passing strings to databases and validations.
375
+ #
376
+ # * <tt>form</tt> - The form you want to normalize in. Should be one of the following:
377
+ # <tt>:c</tt>, <tt>:kc</tt>, <tt>:d</tt>, or <tt>:kd</tt>. Default is
378
+ # ActiveSupport::Multibyte::Unicode.default_normalization_form
379
+ def normalize(form = nil)
380
+ chars(Unicode.normalize(@wrapped_string, form))
381
+ end
382
+
383
+ # Performs canonical decomposition on all the characters.
384
+ #
385
+ # Example:
386
+ # 'é'.length # => 2
387
+ # 'é'.mb_chars.decompose.to_s.length # => 3
388
+ def decompose
389
+ chars(Unicode.decompose_codepoints(:canonical, Unicode.u_unpack(@wrapped_string)).pack('U*'))
390
+ end
391
+
392
+ # Performs composition on all the characters.
393
+ #
394
+ # Example:
395
+ # 'é'.length # => 3
396
+ # 'é'.mb_chars.compose.to_s.length # => 2
397
+ def compose
398
+ chars(Unicode.compose_codepoints(Unicode.u_unpack(@wrapped_string)).pack('U*'))
399
+ end
400
+
401
+ # Returns the number of grapheme clusters in the string.
402
+ #
403
+ # Example:
404
+ # 'क्षि'.mb_chars.length # => 4
405
+ # 'क्षि'.mb_chars.g_length # => 3
406
+ def g_length
407
+ Unicode.g_unpack(@wrapped_string).length
408
+ end
409
+
410
+ # Replaces all ISO-8859-1 or CP1252 characters by their UTF-8 equivalent resulting in a valid UTF-8 string.
411
+ #
412
+ # Passing +true+ will forcibly tidy all bytes, assuming that the string's encoding is entirely CP1252 or ISO-8859-1.
413
+ def tidy_bytes(force = false)
414
+ chars(Unicode.tidy_bytes(@wrapped_string, force))
415
+ end
416
+
417
+ %w(capitalize downcase lstrip reverse rstrip slice strip tidy_bytes upcase).each do |method|
418
+ # Only define a corresponding bang method for methods defined in the proxy; On 1.9 the proxy will
419
+ # exclude lstrip!, rstrip! and strip! because they are already work as expected on multibyte strings.
420
+ if public_method_defined?(method)
421
+ define_method("#{method}!") do |*args|
422
+ @wrapped_string = send(args.nil? ? method : method, *args).to_s
423
+ self
424
+ end
425
+ end
426
+ end
427
+
428
+ protected
429
+
430
+ def translate_offset(byte_offset) #:nodoc:
431
+ return nil if byte_offset.nil?
432
+ return 0 if @wrapped_string == ''
433
+
434
+ if @wrapped_string.respond_to?(:force_encoding)
435
+ @wrapped_string = @wrapped_string.dup.force_encoding(Encoding::ASCII_8BIT)
436
+ end
437
+
438
+ begin
439
+ @wrapped_string[0...byte_offset].unpack('U*').length
440
+ rescue ArgumentError
441
+ byte_offset -= 1
442
+ retry
443
+ end
444
+ end
445
+
446
+ def justify(integer, way, padstr=' ') #:nodoc:
447
+ raise ArgumentError, "zero width padding" if padstr.length == 0
448
+ padsize = integer - size
449
+ padsize = padsize > 0 ? padsize : 0
450
+ case way
451
+ when :right
452
+ result = @wrapped_string.dup.insert(0, padding(padsize, padstr))
453
+ when :left
454
+ result = @wrapped_string.dup.insert(-1, padding(padsize, padstr))
455
+ when :center
456
+ lpad = padding((padsize / 2.0).floor, padstr)
457
+ rpad = padding((padsize / 2.0).ceil, padstr)
458
+ result = @wrapped_string.dup.insert(0, lpad).insert(-1, rpad)
459
+ end
460
+ chars(result)
461
+ end
462
+
463
+ def padding(padsize, padstr=' ') #:nodoc:
464
+ if padsize != 0
465
+ chars(padstr * ((padsize / Unicode.u_unpack(padstr).size) + 1)).slice(0, padsize)
466
+ else
467
+ ''
468
+ end
469
+ end
470
+
471
+ def chars(string) #:nodoc:
472
+ self.class.new(string)
473
+ end
474
+ end
475
+ end
476
+ end
@@ -0,0 +1,8 @@
1
+ # encoding: utf-8
2
+
3
+ module ActiveSupport #:nodoc:
4
+ module Multibyte #:nodoc:
5
+ # Raised when a problem with the encoding was found.
6
+ class EncodingError < StandardError; end
7
+ end
8
+ end