From 22f75d539dca7b6f33cbf86e4e9d1944bb22731f Mon Sep 17 00:00:00 2001 From: Manfred Stienstra Date: Sun, 21 Sep 2008 17:21:30 +0200 Subject: Simplify ActiveSupport::Multibyte and make it run on Ruby 1.9. * Unicode methods are now defined directly on Chars instead of a handler * Updated Unicode database to Unicode 5.1.0 * Improved documentation --- activesupport/bin/generate_tables | 147 ++++ .../lib/active_support/core_ext/string.rb | 6 +- .../active_support/core_ext/string/multibyte.rb | 81 +++ .../lib/active_support/core_ext/string/unicode.rb | 66 -- activesupport/lib/active_support/multibyte.rb | 36 +- .../lib/active_support/multibyte/chars.rb | 782 +++++++++++++++++---- .../lib/active_support/multibyte/exceptions.rb | 7 + .../multibyte/generators/generate_tables.rb | 149 ---- .../multibyte/handlers/passthru_handler.rb | 9 - .../multibyte/handlers/utf8_handler.rb | 564 --------------- .../multibyte/handlers/utf8_handler_proc.rb | 43 -- .../active_support/multibyte/unicode_database.rb | 71 ++ .../lib/active_support/values/unicode_tables.dat | Bin 656156 -> 710734 bytes activesupport/test/abstract_unit.rb | 8 +- activesupport/test/multibyte_chars_test.rb | 642 +++++++++++++---- activesupport/test/multibyte_conformance.rb | 101 ++- activesupport/test/multibyte_handler_test.rb | 372 ---------- .../test/multibyte_unicode_database_test.rb | 28 + 18 files changed, 1562 insertions(+), 1550 deletions(-) create mode 100755 activesupport/bin/generate_tables create mode 100644 activesupport/lib/active_support/core_ext/string/multibyte.rb delete mode 100644 activesupport/lib/active_support/core_ext/string/unicode.rb create mode 100644 activesupport/lib/active_support/multibyte/exceptions.rb delete mode 100755 activesupport/lib/active_support/multibyte/generators/generate_tables.rb delete mode 100644 activesupport/lib/active_support/multibyte/handlers/passthru_handler.rb delete mode 100644 activesupport/lib/active_support/multibyte/handlers/utf8_handler.rb delete mode 100644 activesupport/lib/active_support/multibyte/handlers/utf8_handler_proc.rb create mode 100644 activesupport/lib/active_support/multibyte/unicode_database.rb delete mode 100644 activesupport/test/multibyte_handler_test.rb create mode 100644 activesupport/test/multibyte_unicode_database_test.rb (limited to 'activesupport') diff --git a/activesupport/bin/generate_tables b/activesupport/bin/generate_tables new file mode 100755 index 0000000000..f8e032139f --- /dev/null +++ b/activesupport/bin/generate_tables @@ -0,0 +1,147 @@ +#!/usr/bin/env ruby + +begin + $:.unshift(File.expand_path(File.dirname(__FILE__) + '/../lib')) + require 'active_support' +rescue IOError +end + +require 'open-uri' +require 'tmpdir' + +module ActiveSupport + module Multibyte + class UnicodeDatabase + def load; end + end + + class UnicodeDatabaseGenerator + BASE_URI = "http://www.unicode.org/Public/#{ActiveSupport::Multibyte::UNICODE_VERSION}/ucd/" + SOURCES = { + :codepoints => BASE_URI + 'UnicodeData.txt', + :composition_exclusion => BASE_URI + 'CompositionExclusions.txt', + :grapheme_break_property => BASE_URI + 'auxiliary/GraphemeBreakProperty.txt', + :cp1252 => 'http://unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP1252.TXT' + } + + def initialize + @ucd = UnicodeDatabase.new + + default = Codepoint.new + default.combining_class = 0 + default.uppercase_mapping = 0 + default.lowercase_mapping = 0 + @ucd.codepoints = Hash.new(default) + end + + def parse_codepoints(line) + codepoint = Codepoint.new + raise "Could not parse input." unless line =~ /^ + ([0-9A-F]+); # code + ([^;]+); # name + ([A-Z]+); # general category + ([0-9]+); # canonical combining class + ([A-Z]+); # bidi class + (<([A-Z]*)>)? # decomposition type + ((\ ?[0-9A-F]+)*); # decompomposition mapping + ([0-9]*); # decimal digit + ([0-9]*); # digit + ([^;]*); # numeric + ([YN]*); # bidi mirrored + ([^;]*); # unicode 1.0 name + ([^;]*); # iso comment + ([0-9A-F]*); # simple uppercase mapping + ([0-9A-F]*); # simple lowercase mapping + ([0-9A-F]*)$/ix # simple titlecase mapping + codepoint.code = $1.hex + #codepoint.name = $2 + #codepoint.category = $3 + codepoint.combining_class = Integer($4) + #codepoint.bidi_class = $5 + codepoint.decomp_type = $7 + codepoint.decomp_mapping = ($8=='') ? nil : $8.split.collect { |element| element.hex } + #codepoint.bidi_mirrored = ($13=='Y') ? true : false + codepoint.uppercase_mapping = ($16=='') ? 0 : $16.hex + codepoint.lowercase_mapping = ($17=='') ? 0 : $17.hex + #codepoint.titlecase_mapping = ($18=='') ? nil : $18.hex + @ucd.codepoints[codepoint.code] = codepoint + end + + def parse_grapheme_break_property(line) + if line =~ /^([0-9A-F\.]+)\s*;\s*([\w]+)\s*#/ + type = $2.downcase.intern + @ucd.boundary[type] ||= [] + if $1.include? '..' + parts = $1.split '..' + @ucd.boundary[type] << (parts[0].hex..parts[1].hex) + else + @ucd.boundary[type] << $1.hex + end + end + end + + def parse_composition_exclusion(line) + if line =~ /^([0-9A-F]+)/i + @ucd.composition_exclusion << $1.hex + end + end + + def parse_cp1252(line) + if line =~ /^([0-9A-Fx]+)\s([0-9A-Fx]+)/i + @ucd.cp1252[$1.hex] = $2.hex + end + end + + def create_composition_map + @ucd.codepoints.each do |_, cp| + 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.codepoints[cp.decomp_mapping[0]].combining_class == 0 and !@ucd.composition_exclusion.include?(cp.code) + @ucd.composition_map[cp.decomp_mapping[0]] ||= {} + @ucd.composition_map[cp.decomp_mapping[0]][cp.decomp_mapping[1]] = cp.code + end + end + end + + def normalize_boundary_map + @ucd.boundary.each do |k,v| + if [:lf, :cr].include? k + @ucd.boundary[k] = v[0] + end + end + end + + def parse + SOURCES.each do |type, url| + filename = File.join(Dir.tmpdir, "#{url.split('/').last}") + unless File.exist?(filename) + $stderr.puts "Downloading #{url.split('/').last}" + File.open(filename, 'wb') do |target| + open(url) do |source| + source.each_line { |line| target.write line } + end + end + end + File.open(filename) do |file| + file.each_line { |line| send "parse_#{type}".intern, line } + end + end + create_composition_map + normalize_boundary_map + end + + def dump_to(filename) + File.open(filename, 'wb') do |f| + f.write Marshal.dump([@ucd.codepoints, @ucd.composition_exclusion, @ucd.composition_map, @ucd.boundary, @ucd.cp1252]) + end + end + end + end +end + +if __FILE__ == $0 + filename = ActiveSupport::Multibyte::UnicodeDatabase.filename + generator = ActiveSupport::Multibyte::UnicodeDatabaseGenerator.new + generator.parse + print "Writing to: #{filename}" + generator.dump_to filename + puts " (#{File.size(filename)} bytes)" +end diff --git a/activesupport/lib/active_support/core_ext/string.rb b/activesupport/lib/active_support/core_ext/string.rb index 7ff2f11eff..16c544a577 100644 --- a/activesupport/lib/active_support/core_ext/string.rb +++ b/activesupport/lib/active_support/core_ext/string.rb @@ -1,9 +1,11 @@ +# encoding: utf-8 + require 'active_support/core_ext/string/inflections' require 'active_support/core_ext/string/conversions' require 'active_support/core_ext/string/access' require 'active_support/core_ext/string/starts_ends_with' require 'active_support/core_ext/string/iterators' -require 'active_support/core_ext/string/unicode' +require 'active_support/core_ext/string/multibyte' require 'active_support/core_ext/string/xchar' require 'active_support/core_ext/string/filters' require 'active_support/core_ext/string/behavior' @@ -15,6 +17,6 @@ class String #:nodoc: include ActiveSupport::CoreExtensions::String::Inflections include ActiveSupport::CoreExtensions::String::StartsEndsWith include ActiveSupport::CoreExtensions::String::Iterators - include ActiveSupport::CoreExtensions::String::Unicode include ActiveSupport::CoreExtensions::String::Behavior + include ActiveSupport::CoreExtensions::String::Multibyte end diff --git a/activesupport/lib/active_support/core_ext/string/multibyte.rb b/activesupport/lib/active_support/core_ext/string/multibyte.rb new file mode 100644 index 0000000000..5a2dc36f72 --- /dev/null +++ b/activesupport/lib/active_support/core_ext/string/multibyte.rb @@ -0,0 +1,81 @@ +# encoding: utf-8 + +module ActiveSupport #:nodoc: + module CoreExtensions #:nodoc: + module String #:nodoc: + # Implements multibyte methods for easier access to multibyte characters in a String instance. + module Multibyte + unless '1.9'.respond_to?(:force_encoding) + # +mb_chars+ is a multibyte safe proxy method for string methods. + # + # In Ruby 1.8 and older it creates and returns an instance of the ActiveSupport::Multibyte::Chars class which + # encapsulates the original string. A Unicode safe version of all the String methods are defined on this proxy + # class. If the proxy class doesn't respond to a certain method, it's forwarded to the encapsuled string. + # + # name = 'Claus Müller' + # name.reverse #=> "rell??M sualC" + # name.length #=> 13 + # + # name.mb_chars.reverse.to_s #=> "rellüM sualC" + # name.mb_chars.length #=> 12 + # + # In Ruby 1.9 and newer +mb_chars+ returns +self+ because String is (mostly) encoding aware so we don't need + # a proxy class any more. This means that +mb_chars+ makes it easier to write code that runs on multiple Ruby + # versions. + # + # == Method chaining + # + # All the methods on the Chars proxy which normally return a string will return a Chars object. This allows + # method chaining on the result of any of these methods. + # + # name.mb_chars.reverse.length #=> 12 + # + # == Interoperability and configuration + # + # The Char object tries to be as interchangeable with String objects as possible: sorting and comparing between + # String and Char work like expected. The bang! methods change the internal string representation in the Chars + # object. Interoperability problems can be resolved easily with a +to_s+ call. + # + # For more information about the methods defined on the Chars proxy see ActiveSupport::Multibyte::Chars. For + # information about how to change the default Multibyte behaviour, see ActiveSupport::Multibyte. + def mb_chars + if ActiveSupport::Multibyte.proxy_class.wants?(self) + ActiveSupport::Multibyte.proxy_class.new(self) + else + self + end + end + + # Returns true if the string has UTF-8 semantics (a String used for purely byte resources is unlikely to have + # them), returns false otherwise. + def is_utf8? + ActiveSupport::Multibyte::Chars.consumes?(self) + end + + unless '1.8.7 and later'.respond_to?(:chars) + alias chars mb_chars + end + else + # In Ruby 1.9 and newer +mb_chars+ returns self. In Ruby 1.8 and older +mb_chars+ creates and returns an + # Unicode safe proxy for string operations, this makes it easier to write code that runs on multiple Ruby + # versions. + def mb_chars + self + end + + # Returns true if the string has valid UTF-8 encoding. + def is_utf8? + case encoding + when Encoding::UTF_8 + valid_encoding? + when Encoding::ASCII_8BIT, Encoding::US_ASCII + dup.force_encoding(Encoding::UTF_8).valid_encoding? + else + false + end + end + end + end + end + end +end diff --git a/activesupport/lib/active_support/core_ext/string/unicode.rb b/activesupport/lib/active_support/core_ext/string/unicode.rb deleted file mode 100644 index 666f7bcb65..0000000000 --- a/activesupport/lib/active_support/core_ext/string/unicode.rb +++ /dev/null @@ -1,66 +0,0 @@ -module ActiveSupport #:nodoc: - module CoreExtensions #:nodoc: - module String #:nodoc: - # Define methods for handling unicode data. - module Unicode - def self.append_features(base) - if '1.8.7 and later'.respond_to?(:chars) - base.class_eval { remove_method :chars } - end - super - end - - unless '1.9'.respond_to?(:force_encoding) - # +chars+ is a Unicode safe proxy for string methods. It creates and returns an instance of the - # ActiveSupport::Multibyte::Chars class which encapsulates the original string. A Unicode safe version of all - # the String methods are defined on this proxy class. Undefined methods are forwarded to String, so all of the - # string overrides can also be called through the +chars+ proxy. - # - # name = 'Claus Müller' - # name.reverse # => "rell??M sualC" - # name.length # => 13 - # - # name.chars.reverse.to_s # => "rellüM sualC" - # name.chars.length # => 12 - # - # - # All the methods on the chars proxy which normally return a string will return a Chars object. This allows - # method chaining on the result of any of these methods. - # - # name.chars.reverse.length # => 12 - # - # The Char object tries to be as interchangeable with String objects as possible: sorting and comparing between - # String and Char work like expected. The bang! methods change the internal string representation in the Chars - # object. Interoperability problems can be resolved easily with a +to_s+ call. - # - # For more information about the methods defined on the Chars proxy see ActiveSupport::Multibyte::Chars and - # ActiveSupport::Multibyte::Handlers::UTF8Handler. - def chars - ActiveSupport::Multibyte::Chars.new(self) - end - - # Returns true if the string has UTF-8 semantics (a String used for purely byte resources is unlikely to have - # them), returns false otherwise. - def is_utf8? - ActiveSupport::Multibyte::Handlers::UTF8Handler.consumes?(self) - end - else - def chars #:nodoc: - self - end - - def is_utf8? #:nodoc: - case encoding - when Encoding::UTF_8 - valid_encoding? - when Encoding::ASCII_8BIT - dup.force_encoding('UTF-8').valid_encoding? - else - false - end - end - end - end - end - end -end diff --git a/activesupport/lib/active_support/multibyte.rb b/activesupport/lib/active_support/multibyte.rb index 27c0d180a5..63c0d50166 100644 --- a/activesupport/lib/active_support/multibyte.rb +++ b/activesupport/lib/active_support/multibyte.rb @@ -1,9 +1,33 @@ -module ActiveSupport +# encoding: utf-8 + +require 'active_support/multibyte/chars' +require 'active_support/multibyte/exceptions' +require 'active_support/multibyte/unicode_database' + +module ActiveSupport #:nodoc: module Multibyte #:nodoc: - DEFAULT_NORMALIZATION_FORM = :kc + # A list of all available normalization forms. See http://www.unicode.org/reports/tr15/tr15-29.html for more + # information about normalization. NORMALIZATIONS_FORMS = [:c, :kc, :d, :kd] - UNICODE_VERSION = '5.0.0' - end -end -require 'active_support/multibyte/chars' + # The Unicode version that is supported by the implementation + UNICODE_VERSION = '5.1.0' + + # The default normalization used for operations that require normalization. It can be set to any of the + # normalizations in NORMALIZATIONS_FORMS. + # + # Example: + # ActiveSupport::Multibyte.default_normalization_form = :c + mattr_accessor :default_normalization_form + self.default_normalization_form = :kc + + # The proxy class returned when calling mb_chars. You can use this accessor to configure your own proxy + # class so you can support other encodings. See the ActiveSupport::Multibyte::Chars implementation for + # an example how to do this. + # + # Example: + # ActiveSupport::Multibyte.proxy_class = CharsForUTF32 + mattr_accessor :proxy_class + self.proxy_class = ActiveSupport::Multibyte::Chars + end +end \ No newline at end of file diff --git a/activesupport/lib/active_support/multibyte/chars.rb b/activesupport/lib/active_support/multibyte/chars.rb index de2c83f8d1..c05419bfbf 100644 --- a/activesupport/lib/active_support/multibyte/chars.rb +++ b/activesupport/lib/active_support/multibyte/chars.rb @@ -1,142 +1,670 @@ -require 'active_support/multibyte/handlers/utf8_handler' -require 'active_support/multibyte/handlers/passthru_handler' - -# Encapsulates all the functionality related to the Chars proxy. -module ActiveSupport::Multibyte #:nodoc: - # Chars enables you to work transparently with multibyte encodings in the Ruby String class without having extensive - # knowledge about the encoding. A Chars object accepts a string upon initialization and proxies String methods in an - # encoding safe manner. All the normal String methods are also implemented on the proxy. - # - # String methods are proxied through the Chars object, and can be accessed through the +chars+ method. Methods - # which would normally return a String object now return a Chars object so methods can be chained. - # - # "The Perfect String ".chars.downcase.strip.normalize # => "the perfect string" - # - # Chars objects are perfectly interchangeable with String objects as long as no explicit class checks are made. - # If certain methods do explicitly check the class, call +to_s+ before you pass chars objects to them. - # - # bad.explicit_checking_method "T".chars.downcase.to_s - # - # The actual operations on the string are delegated to handlers. Theoretically handlers can be implemented for - # any encoding, but the default handler handles UTF-8. This handler is set during initialization, if you want to - # use you own handler, you can set it on the Chars class. Look at the UTF8Handler source for an example how to - # implement your own handler. If you your own handler to work on anything but UTF-8 you probably also - # want to override Chars#handler. - # - # ActiveSupport::Multibyte::Chars.handler = MyHandler - # - # Note that a few methods are defined on Chars instead of the handler because they are defined on Object or Kernel - # and method_missing can't catch them. - class Chars - - attr_reader :string # The contained string - alias_method :to_s, :string - - include Comparable - - # The magic method to make String and Chars comparable - def to_str - # Using any other ways of overriding the String itself will lead you all the way from infinite loops to - # core dumps. Don't go there. - @string - end +# encoding: utf-8 + +module ActiveSupport #:nodoc: + module Multibyte #:nodoc: + # Chars enables you to work transparently with multibyte encodings in the Ruby String class without having extensive + # knowledge about the encoding. A Chars object accepts a string upon initialization and proxies String methods in an + # encoding safe manner. All the normal String methods are also implemented on the proxy. + # + # String methods are proxied through the Chars object, and can be accessed through the +mb_chars+ method. Methods + # which would normally return a String object now return a Chars object so methods can be chained. + # + # "The Perfect String ".chars.downcase.strip.normalize #=> "the perfect string" + # + # Chars objects are perfectly interchangeable with String objects as long as no explicit class checks are made. + # If certain methods do explicitly check the class, call +to_s+ before you pass chars objects to them. + # + # bad.explicit_checking_method "T".chars.downcase.to_s + # + # The default Chars implementation assumes that the encoding of the string is UTF-8, if you want to handle different + # encodings you can write your own multibyte string handler and configure it through + # ActiveSupport::Multibyte.proxy_class. + # + # class CharsForUTF32 + # def size + # @wrapped_string.size / 4 + # end + # + # def self.accepts?(string) + # string.length % 4 == 0 + # end + # end + # + # ActiveSupport::Multibyte.proxy_class = CharsForUTF32 + class Chars + # Hangul character boundaries and properties + HANGUL_SBASE = 0xAC00 + HANGUL_LBASE = 0x1100 + HANGUL_VBASE = 0x1161 + HANGUL_TBASE = 0x11A7 + HANGUL_LCOUNT = 19 + HANGUL_VCOUNT = 21 + HANGUL_TCOUNT = 28 + HANGUL_NCOUNT = HANGUL_VCOUNT * HANGUL_TCOUNT + HANGUL_SCOUNT = 11172 + HANGUL_SLAST = HANGUL_SBASE + HANGUL_SCOUNT + HANGUL_JAMO_FIRST = 0x1100 + HANGUL_JAMO_LAST = 0x11FF + + # All the unicode whitespace + UNICODE_WHITESPACE = [ + (0x0009..0x000D).to_a, # White_Space # Cc [5] .. + 0x0020, # White_Space # Zs SPACE + 0x0085, # White_Space # Cc + 0x00A0, # White_Space # Zs NO-BREAK SPACE + 0x1680, # White_Space # Zs OGHAM SPACE MARK + 0x180E, # White_Space # Zs MONGOLIAN VOWEL SEPARATOR + (0x2000..0x200A).to_a, # White_Space # Zs [11] EN QUAD..HAIR SPACE + 0x2028, # White_Space # Zl LINE SEPARATOR + 0x2029, # White_Space # Zp PARAGRAPH SEPARATOR + 0x202F, # White_Space # Zs NARROW NO-BREAK SPACE + 0x205F, # White_Space # Zs MEDIUM MATHEMATICAL SPACE + 0x3000, # White_Space # Zs IDEOGRAPHIC SPACE + ].flatten.freeze + + # BOM (byte order mark) can also be seen as whitespace, it's a non-rendering character used to distinguish + # between little and big endian. This is not an issue in utf-8, so it must be ignored. + UNICODE_LEADERS_AND_TRAILERS = UNICODE_WHITESPACE + [65279] # ZERO-WIDTH NO-BREAK SPACE aka BOM + + # Returns a regular expression pattern that matches the passed Unicode codepoints + def self.codepoints_to_pattern(array_of_codepoints) #:nodoc: + array_of_codepoints.collect{ |e| [e].pack 'U*' }.join('|') + end + UNICODE_TRAILERS_PAT = /(#{codepoints_to_pattern(UNICODE_LEADERS_AND_TRAILERS)})+\Z/ + UNICODE_LEADERS_PAT = /\A(#{codepoints_to_pattern(UNICODE_LEADERS_AND_TRAILERS)})+/ + + # Borrowed from the Kconv library by Shinji KONO - (also as seen on the W3C site) + UTF8_PAT = /\A(?: + [\x00-\x7f] | + [\xc2-\xdf] [\x80-\xbf] | + \xe0 [\xa0-\xbf] [\x80-\xbf] | + [\xe1-\xef] [\x80-\xbf] [\x80-\xbf] | + \xf0 [\x90-\xbf] [\x80-\xbf] [\x80-\xbf] | + [\xf1-\xf3] [\x80-\xbf] [\x80-\xbf] [\x80-\xbf] | + \xf4 [\x80-\x8f] [\x80-\xbf] [\x80-\xbf] + )*\z/xn + + attr_reader :wrapped_string + alias to_s wrapped_string + alias to_str wrapped_string + + # Creates a new Chars instance. +string+ is the wrapped string. + if '1.9'.respond_to?(:force_encoding) + def initialize(string) + @wrapped_string = string + @wrapped_string.force_encoding(Encoding::UTF_8) unless @wrapped_string.frozen? + end + else + def initialize(string) + @wrapped_string = string + end + end + + # Forward all undefined methods to the wrapped string. + def method_missing(method, *args, &block) + if method.to_s =~ /!$/ + @wrapped_string.__send__(method, *args, &block) + self + else + chars(@wrapped_string.__send__(method, *args, &block)) + end + end + + # Returns +true+ if _obj_ responds to the given method. Private methods are included in the search + # only if the optional second parameter evaluates to +true+. + def respond_to?(method, include_private=false) + super || @wrapped_string.respond_to?(method, include_private) || false + end + + # Enable more predictable duck-typing on String-like classes. See Object#acts_like?. + def acts_like_string? + true + end + + # Returns +true+ if the Chars class can and should act as a proxy for the string +string+. Returns + # +false+ otherwise. + def self.wants?(string) + RUBY_VERSION < '1.9' && $KCODE == 'UTF8' && consumes?(string) + end - # Make duck-typing with String possible - def respond_to?(method, include_priv = false) - super || @string.respond_to?(method, include_priv) || - handler.respond_to?(method, include_priv) || - (method.to_s =~ /(.*)!/ && handler.respond_to?($1, include_priv)) || + # Returns +true+ when the proxy class can handle the string. Returns +false+ otherwise. + def self.consumes?(string) + # Unpack is a little bit faster than regular expressions. + string.unpack('U*') + true + rescue ArgumentError false - end + end - # Enable more predictable duck-typing on String-like classes. See Object#acts_like?. - def acts_like_string? - true - end + include Comparable - # Create a new Chars instance. - def initialize(str) - @string = str.respond_to?(:string) ? str.string : str - end - - # Returns -1, 0 or +1 depending on whether the Chars object is to be sorted before, equal or after the - # object on the right side of the operation. It accepts any object that implements +to_s+. See String.<=> - # for more details. - def <=>(other); @string <=> other.to_s; end - - # Works just like String#split, with the exception that the items in the resulting list are Chars - # instances instead of String. This makes chaining methods easier. - def split(*args) - @string.split(*args).map { |i| i.chars } - end - - # Gsub works exactly the same as gsub on a normal string. - def gsub(*a, &b); @string.gsub(*a, &b).chars; end - - # Like String.=~ only it returns the character offset (in codepoints) instead of the byte offset. - def =~(other) - handler.translate_offset(@string, @string =~ other) - end - - # Try to forward all undefined methods to the handler, when a method is not defined on the handler, send it to - # the contained string. Method_missing is also responsible for making the bang! methods destructive. - def method_missing(m, *a, &b) - begin - # Simulate methods with a ! at the end because we can't touch the enclosed string from the handlers. - if m.to_s =~ /^(.*)\!$/ && handler.respond_to?($1) - result = handler.send($1, @string, *a, &b) - if result == @string - result = nil + # Returns -1, 0 or +1 depending on whether the Chars object is to be sorted before, equal or after the + # object on the right side of the operation. It accepts any object that implements +to_s+. See String.<=> + # for more details. + # + # Example: + # 'é'.mb_chars <=> 'ü'.mb_chars #=> -1 + def <=>(other) + @wrapped_string <=> other.to_s + end + + # Returns a new Chars object containing the other object concatenated to the string. + # + # Example: + # ('Café'.mb_chars + ' périferôl').to_s #=> "Café périferôl" + def +(other) + self << other + end + + # Like String.=~ only it returns the character offset (in codepoints) instead of the byte offset. + # + # Example: + # 'Café périferôl'.mb_chars =~ /ô/ #=> 12 + def =~(other) + translate_offset(@wrapped_string =~ other) + end + + # Works just like String#split, with the exception that the items in the resulting list are Chars + # instances instead of String. This makes chaining methods easier. + # + # Example: + # 'Café périferôl'.mb_chars.split(/é/).map { |part| part.upcase.to_s } #=> ["CAF", " P", "RIFERÔL"] + def split(*args) + @wrapped_string.split(*args).map { |i| i.mb_chars } + end + + # Inserts the passed string at specified codepoint offsets + # + # Example: + # 'Café'.mb_chars.insert(4, ' périferôl').to_s #=> "Café périferôl" + def insert(offset, fragment) + unpacked = self.class.u_unpack(@wrapped_string) + unless offset > unpacked.length + @wrapped_string.replace( + self.class.u_unpack(@wrapped_string).insert(offset, *self.class.u_unpack(fragment)).pack('U*') + ) + else + raise IndexError, "index #{offset} out of string" + end + self + end + + # Returns true if contained string contains +other+. Returns false otherwise. + # + # Example: + # 'Café'.mb_chars.include?('é') #=> true + def include?(other) + # We have to redefine this method because Enumerable defines it. + @wrapped_string.include?(other) + end + + # Returns the position of the passed argument in the string, counting in codepoints + # + # Example: + # 'Café périferôl'.mb_chars.index('ô') #=> 12 + def index(*args) + index = @wrapped_string.index(*args) + index ? (self.class.u_unpack(@wrapped_string.slice(0...index)).size) : nil + end + + # Works just like the indexed replace method on string, except instead of byte offsets you specify + # character offsets. + # + # Example: + # + # s = "Müller" + # s.chars[2] = "e" # Replace character with offset 2 + # s + # #=> "Müeler" + # + # s = "Müller" + # s.chars[1, 2] = "ö" # Replace 2 characters at character offset 1 + # s + # #=> "Möler" + def []=(*args) + replace_by = args.pop + # Indexed replace with regular expressions already works + if args.first.is_a?(Regexp) + @wrapped_string[*args] = replace_by + else + result = self.class.u_unpack(@wrapped_string) + if args[0].is_a?(Fixnum) + raise IndexError, "index #{args[0]} out of string" if args[0] >= result.length + min = args[0] + max = args[1].nil? ? min : (min + args[1] - 1) + range = Range.new(min, max) + replace_by = [replace_by].pack('U') if replace_by.is_a?(Fixnum) + elsif args.first.is_a?(Range) + raise RangeError, "#{args[0]} out of range" if args[0].min >= result.length + range = args[0] else - @string.replace result + needle = args[0].to_s + min = index(needle) + max = min + self.class.u_unpack(needle).length - 1 + range = Range.new(min, max) end - elsif handler.respond_to?(m) - result = handler.send(m, @string, *a, &b) - else - result = @string.send(m, *a, &b) + result[range] = self.class.u_unpack(replace_by) + @wrapped_string.replace(result.pack('U*')) end - rescue Handlers::EncodingError - @string.replace handler.tidy_bytes(@string) - retry + self end - - if result.kind_of?(String) - result.chars - else - result + + # Works just like String#rjust, only integer specifies characters instead of bytes. + # + # Example: + # + # "¾ cup".chars.rjust(8).to_s + # #=> " ¾ cup" + # + # "¾ cup".chars.rjust(8, " ").to_s # Use non-breaking whitespace + # #=> "   ¾ cup" + def rjust(integer, padstr=' ') + justify(integer, :right, padstr) end - end - - # Set the handler class for the Char objects. - def self.handler=(klass) - @@handler = klass - end - # Returns the proper handler for the contained string depending on $KCODE and the encoding of the string. This - # method is used internally to always redirect messages to the proper classes depending on the context. - def handler - if utf8_pragma? - @@handler - else - ActiveSupport::Multibyte::Handlers::PassthruHandler + # Works just like String#ljust, only integer specifies characters instead of bytes. + # + # Example: + # + # "¾ cup".chars.rjust(8).to_s + # #=> "¾ cup " + # + # "¾ cup".chars.rjust(8, " ").to_s # Use non-breaking whitespace + # #=> "¾ cup   " + def ljust(integer, padstr=' ') + justify(integer, :left, padstr) + end + + # Works just like String#center, only integer specifies characters instead of bytes. + # + # Example: + # + # "¾ cup".chars.center(8).to_s + # #=> " ¾ cup " + # + # "¾ cup".chars.center(8, " ").to_s # Use non-breaking whitespace + # #=> " ¾ cup  " + def center(integer, padstr=' ') + justify(integer, :center, padstr) end - end - private + # Strips entire range of Unicode whitespace from the right of the string. + def rstrip + chars(@wrapped_string.gsub(UNICODE_TRAILERS_PAT, '')) + end + + # Strips entire range of Unicode whitespace from the left of the string. + def lstrip + chars(@wrapped_string.gsub(UNICODE_LEADERS_PAT, '')) + end + + # Strips entire range of Unicode whitespace from the right and left of the string. + def strip + rstrip.lstrip + end + + # Returns the number of codepoints in the string + def size + self.class.u_unpack(@wrapped_string).size + end + alias_method :length, :size - # +utf8_pragma+ checks if it can send this string to the handlers. It makes sure @string isn't nil and $KCODE is - # set to 'UTF8'. - def utf8_pragma? - !@string.nil? && ($KCODE == 'UTF8') + # Reverses all characters in the string + # + # Example: + # 'Café'.mb_chars.reverse.to_s #=> 'éfaC' + def reverse + chars(self.class.u_unpack(@wrapped_string).reverse.pack('U*')) end + + # Implements Unicode-aware slice with codepoints. Slicing on one point returns the codepoints for that + # character. + # + # Example: + # 'こにちわ'.mb_chars.slice(2..3).to_s #=> "ちわ" + def slice(*args) + if args.size > 2 + raise ArgumentError, "wrong number of arguments (#{args.size} for 1)" # Do as if we were native + elsif (args.size == 2 && !(args.first.is_a?(Numeric) || args.first.is_a?(Regexp))) + raise TypeError, "cannot convert #{args.first.class} into Integer" # Do as if we were native + elsif (args.size == 2 && !args[1].is_a?(Numeric)) + raise TypeError, "cannot convert #{args[1].class} into Integer" # Do as if we were native + elsif args[0].kind_of? Range + cps = self.class.u_unpack(@wrapped_string).slice(*args) + result = cps.nil? ? nil : cps.pack('U*') + elsif args[0].kind_of? Regexp + result = @wrapped_string.slice(*args) + elsif args.size == 1 && args[0].kind_of?(Numeric) + character = self.class.u_unpack(@wrapped_string)[args[0]] + result = character.nil? ? nil : [character].pack('U') + else + result = self.class.u_unpack(@wrapped_string).slice(*args).pack('U*') + end + result.nil? ? nil : chars(result) + end + alias_method :[], :slice + + # Convert characters in the string to uppercase + # + # Example: + # 'Laurent, òu sont les tests?'.mb_chars.upcase.to_s #=> "LAURENT, ÒU SONT LES TESTS?" + def upcase + apply_mapping :uppercase_mapping + end + + # Convert characters in the string to lowercase + # + # Example: + # 'VĚDA A VÝZKUM'.mb_chars.downcase.to_s #=> "věda a výzkum" + def downcase + apply_mapping :lowercase_mapping + end + + # Converts the first character to uppercase and the remainder to lowercase + # + # Example: + # 'über'.mb_chars.capitalize.to_s #=> "Über" + def capitalize + (slice(0) || '').upcase + (slice(1..-1) || '').downcase + end + + # Returns the KC normalization of the string by default. NFKC is considered the best normalization form for + # passing strings to databases and validations. + # + # * str - The string to perform normalization on. + # * form - The form you want to normalize in. Should be one of the following: + # :c, :kc, :d, or :kd. Default is + # ActiveSupport::Multibyte.default_normalization_form + def normalize(form=ActiveSupport::Multibyte.default_normalization_form) + # See http://www.unicode.org/reports/tr15, Table 1 + codepoints = self.class.u_unpack(@wrapped_string) + chars(case form + when :d + self.class.reorder_characters(self.class.decompose_codepoints(:canonical, codepoints)) + when :c + self.class.compose_codepoints(self.class.reorder_characters(self.class.decompose_codepoints(:canonical, codepoints))) + when :kd + self.class.reorder_characters(self.class.decompose_codepoints(:compatability, codepoints)) + when :kc + self.class.compose_codepoints(self.class.reorder_characters(self.class.decompose_codepoints(:compatability, codepoints))) + else + raise ArgumentError, "#{form} is not a valid normalization variant", caller + end.pack('U*')) + end + + # Performs canonical decomposition on all the characters. + # + # Example: + # 'é'.length #=> 2 + # 'é'.mb_chars.decompose.to_s.length #=> 3 + def decompose + chars(self.class.decompose_codepoints(:canonical, self.class.u_unpack(@wrapped_string)).pack('U*')) + end + + # Performs composition on all the characters. + # + # Example: + # 'é'.length #=> 3 + # 'é'.mb_chars.compose.to_s.length #=> 2 + def compose + chars(self.class.compose_codepoints(self.class.u_unpack(@wrapped_string)).pack('U*')) + end + + # Returns the number of grapheme clusters in the string. + # + # Example: + # 'क्षि'.mb_chars.length #=> 4 + # 'क्षि'.mb_chars.g_length #=> 3 + def g_length + self.class.g_unpack(@wrapped_string).length + end + + def tidy_bytes + chars(self.class.tidy_bytes(@wrapped_string)) + end + + %w(lstrip rstrip strip reverse upcase downcase slice tidy_bytes capitalize).each do |method| + define_method("#{method}!") do |*args| + unless args.nil? + @wrapped_string = send(method, *args).to_s + else + @wrapped_string = send(method).to_s + end + self + end + end + + class << self + + # Unpack the string at codepoints boundaries + def u_unpack(str) + begin + str.unpack 'U*' + rescue ArgumentError + raise EncodingError.new('malformed UTF-8 character') + end + end + + # Detect whether the codepoint is in a certain character class. Primarily used by the + # grapheme cluster support. + def in_char_class?(codepoint, classes) + classes.detect { |c| UCD.boundary[c] === codepoint } ? true : false + end + + # Unpack the string at grapheme boundaries + def g_unpack(str) + codepoints = u_unpack(str) + unpacked = [] + pos = 0 + marker = 0 + eoc = codepoints.length + while(pos < eoc) + pos += 1 + previous = codepoints[pos-1] + current = codepoints[pos] + if ( + # CR X LF + one = ( previous == UCD.boundary[:cr] and current == UCD.boundary[:lf] ) or + # L X (L|V|LV|LVT) + two = ( UCD.boundary[:l] === previous and in_char_class?(current, [:l,:v,:lv,:lvt]) ) or + # (LV|V) X (V|T) + three = ( in_char_class?(previous, [:lv,:v]) and in_char_class?(current, [:v,:t]) ) or + # (LVT|T) X (T) + four = ( in_char_class?(previous, [:lvt,:t]) and UCD.boundary[:t] === current ) or + # X Extend + five = (UCD.boundary[:extend] === current) + ) + else + unpacked << codepoints[marker..pos-1] + marker = pos + end + end + unpacked + end + + # Reverse operation of g_unpack + def g_pack(unpacked) + (unpacked.flatten).pack('U*') + end + + # Generates a padding string of a certain size. + def padding(padsize, padstr=' ') + if padsize != 0 + new(padstr * ((padsize / u_unpack(padstr).size) + 1)).slice(0, padsize) + else + '' + end + end + + # Re-order codepoints so the string becomes canonical + def reorder_characters(codepoints) + length = codepoints.length- 1 + pos = 0 + while pos < length do + cp1, cp2 = UCD.codepoints[codepoints[pos]], UCD.codepoints[codepoints[pos+1]] + if (cp1.combining_class > cp2.combining_class) && (cp2.combining_class > 0) + codepoints[pos..pos+1] = cp2.code, cp1.code + pos += (pos > 0 ? -1 : 1) + else + pos += 1 + end + end + codepoints + end + + # Decompose composed characters to the decomposed form + def decompose_codepoints(type, codepoints) + codepoints.inject([]) do |decomposed, cp| + # if it's a hangul syllable starter character + if HANGUL_SBASE <= cp and cp < HANGUL_SLAST + sindex = cp - HANGUL_SBASE + ncp = [] # new codepoints + ncp << HANGUL_LBASE + sindex / HANGUL_NCOUNT + ncp << HANGUL_VBASE + (sindex % HANGUL_NCOUNT) / HANGUL_TCOUNT + tindex = sindex % HANGUL_TCOUNT + ncp << (HANGUL_TBASE + tindex) unless tindex == 0 + decomposed.concat ncp + # if the codepoint is decomposable in with the current decomposition type + elsif (ncp = UCD.codepoints[cp].decomp_mapping) and (!UCD.codepoints[cp].decomp_type || type == :compatability) + decomposed.concat decompose_codepoints(type, ncp.dup) + else + decomposed << cp + end + end + end + + # Compose decomposed characters to the composed form + def compose_codepoints(codepoints) + pos = 0 + eoa = codepoints.length - 1 + starter_pos = 0 + starter_char = codepoints[0] + previous_combining_class = -1 + while pos < eoa + pos += 1 + lindex = starter_char - HANGUL_LBASE + # -- Hangul + if 0 <= lindex and lindex < HANGUL_LCOUNT + vindex = codepoints[starter_pos+1] - HANGUL_VBASE rescue vindex = -1 + if 0 <= vindex and vindex < HANGUL_VCOUNT + tindex = codepoints[starter_pos+2] - HANGUL_TBASE rescue tindex = -1 + if 0 <= tindex and tindex < HANGUL_TCOUNT + j = starter_pos + 2 + eoa -= 2 + else + tindex = 0 + j = starter_pos + 1 + eoa -= 1 + end + codepoints[starter_pos..j] = (lindex * HANGUL_VCOUNT + vindex) * HANGUL_TCOUNT + tindex + HANGUL_SBASE + end + starter_pos += 1 + starter_char = codepoints[starter_pos] + # -- Other characters + else + current_char = codepoints[pos] + current = UCD.codepoints[current_char] + if current.combining_class > previous_combining_class + if ref = UCD.composition_map[starter_char] + composition = ref[current_char] + else + composition = nil + end + unless composition.nil? + codepoints[starter_pos] = composition + starter_char = composition + codepoints.delete_at pos + eoa -= 1 + pos -= 1 + previous_combining_class = -1 + else + previous_combining_class = current.combining_class + end + else + previous_combining_class = current.combining_class + end + if current.combining_class == 0 + starter_pos = pos + starter_char = codepoints[pos] + end + end + end + codepoints + end + + # Replaces all the non-UTF-8 bytes by their iso-8859-1 or cp1252 equivalent resulting in a valid UTF-8 string + def tidy_bytes(str) + str.split(//u).map do |c| + if !UTF8_PAT.match(c) + n = c.unpack('C')[0] + n < 128 ? n.chr : + n < 160 ? [UCD.cp1252[n] || n].pack('U') : + n < 192 ? "\xC2" + n.chr : "\xC3" + (n-64).chr + else + c + end + end.join + end + end + + protected + + # Translate a byte offset in the wrapped string to a character offset by looking for the character boundary + def translate_offset(byte_offset) + return nil if byte_offset.nil? + return 0 if @wrapped_string == '' + chunk = @wrapped_string[0..byte_offset] + begin + begin + chunk.unpack('U*').length - 1 + rescue ArgumentError => e + chunk = @wrapped_string[0..(byte_offset+=1)] + # Stop retrying at the end of the string + raise e unless byte_offset < chunk.length + # We damaged a character, retry + retry + end + # Catch the ArgumentError so we can throw our own + rescue ArgumentError + raise EncodingError, 'malformed UTF-8 character' + end + end + + # Justifies a string in a certain way. Valid values for way are :right, :left and + # :center. + def justify(integer, way, padstr=' ') + raise ArgumentError, "zero width padding" if padstr.length == 0 + padsize = integer - size + padsize = padsize > 0 ? padsize : 0 + case way + when :right + result = @wrapped_string.dup.insert(0, self.class.padding(padsize, padstr)) + when :left + result = @wrapped_string.dup.insert(-1, self.class.padding(padsize, padstr)) + when :center + lpad = self.class.padding((padsize / 2.0).floor, padstr) + rpad = self.class.padding((padsize / 2.0).ceil, padstr) + result = @wrapped_string.dup.insert(0, lpad).insert(-1, rpad) + end + chars(result) + end + + # Map codepoints to one of it's attributes. + def apply_mapping(mapping) + chars(self.class.u_unpack(@wrapped_string).map do |codepoint| + cp = UCD.codepoints[codepoint] + if cp and (ncp = cp.send(mapping)) and ncp > 0 + ncp + else + codepoint + end + end.pack('U*')) + end + + # Creates a new instance + def chars(str) + self.class.new(str) + end + end end -end - -# When we can load the utf8proc library, override normalization with the faster methods -begin - require 'utf8proc_native' - require 'active_support/multibyte/handlers/utf8_handler_proc' - ActiveSupport::Multibyte::Chars.handler = ActiveSupport::Multibyte::Handlers::UTF8HandlerProc -rescue LoadError - ActiveSupport::Multibyte::Chars.handler = ActiveSupport::Multibyte::Handlers::UTF8Handler -end +end \ No newline at end of file diff --git a/activesupport/lib/active_support/multibyte/exceptions.rb b/activesupport/lib/active_support/multibyte/exceptions.rb new file mode 100644 index 0000000000..af760cc561 --- /dev/null +++ b/activesupport/lib/active_support/multibyte/exceptions.rb @@ -0,0 +1,7 @@ +# encoding: utf-8 + +module ActiveSupport #:nodoc: + module Multibyte #:nodoc: + class EncodingError < StandardError; end + end +end \ No newline at end of file diff --git a/activesupport/lib/active_support/multibyte/generators/generate_tables.rb b/activesupport/lib/active_support/multibyte/generators/generate_tables.rb deleted file mode 100755 index 7f807585c5..0000000000 --- a/activesupport/lib/active_support/multibyte/generators/generate_tables.rb +++ /dev/null @@ -1,149 +0,0 @@ -#!/usr/bin/env ruby -begin - require File.dirname(__FILE__) + '/../../../active_support' -rescue IOError -end -require 'open-uri' -require 'tmpdir' - -module ActiveSupport::Multibyte::Handlers #:nodoc: - class UnicodeDatabase #:nodoc: - def self.load - [Hash.new(Codepoint.new),[],{},{}] - end - end - - class UnicodeTableGenerator #:nodoc: - BASE_URI = "http://www.unicode.org/Public/#{ActiveSupport::Multibyte::UNICODE_VERSION}/ucd/" - SOURCES = { - :codepoints => BASE_URI + 'UnicodeData.txt', - :composition_exclusion => BASE_URI + 'CompositionExclusions.txt', - :grapheme_break_property => BASE_URI + 'auxiliary/GraphemeBreakProperty.txt', - :cp1252 => 'http://unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP1252.TXT' - } - - def initialize - @ucd = UnicodeDatabase.new - - default = Codepoint.new - default.combining_class = 0 - default.uppercase_mapping = 0 - default.lowercase_mapping = 0 - @ucd.codepoints = Hash.new(default) - - @ucd.composition_exclusion = [] - @ucd.composition_map = {} - @ucd.boundary = {} - @ucd.cp1252 = {} - end - - def parse_codepoints(line) - codepoint = Codepoint.new - raise "Could not parse input." unless line =~ /^ - ([0-9A-F]+); # code - ([^;]+); # name - ([A-Z]+); # general category - ([0-9]+); # canonical combining class - ([A-Z]+); # bidi class - (<([A-Z]*)>)? # decomposition type - ((\ ?[0-9A-F]+)*); # decompomposition mapping - ([0-9]*); # decimal digit - ([0-9]*); # digit - ([^;]*); # numeric - ([YN]*); # bidi mirrored - ([^;]*); # unicode 1.0 name - ([^;]*); # iso comment - ([0-9A-F]*); # simple uppercase mapping - ([0-9A-F]*); # simple lowercase mapping - ([0-9A-F]*)$/ix # simple titlecase mapping - codepoint.code = $1.hex - #codepoint.name = $2 - #codepoint.category = $3 - codepoint.combining_class = Integer($4) - #codepoint.bidi_class = $5 - codepoint.decomp_type = $7 - codepoint.decomp_mapping = ($8=='') ? nil : $8.split.collect { |element| element.hex } - #codepoint.bidi_mirrored = ($13=='Y') ? true : false - codepoint.uppercase_mapping = ($16=='') ? 0 : $16.hex - codepoint.lowercase_mapping = ($17=='') ? 0 : $17.hex - #codepoint.titlecase_mapping = ($18=='') ? nil : $18.hex - @ucd.codepoints[codepoint.code] = codepoint - end - - def parse_grapheme_break_property(line) - if line =~ /^([0-9A-F\.]+)\s*;\s*([\w]+)\s*#/ - type = $2.downcase.intern - @ucd.boundary[type] ||= [] - if $1.include? '..' - parts = $1.split '..' - @ucd.boundary[type] << (parts[0].hex..parts[1].hex) - else - @ucd.boundary[type] << $1.hex - end - end - end - - def parse_composition_exclusion(line) - if line =~ /^([0-9A-F]+)/i - @ucd.composition_exclusion << $1.hex - end - end - - def parse_cp1252(line) - if line =~ /^([0-9A-Fx]+)\s([0-9A-Fx]+)/i - @ucd.cp1252[$1.hex] = $2.hex - end - end - - def create_composition_map - @ucd.codepoints.each do |_, cp| - 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) - @ucd.composition_map[cp.decomp_mapping[0]] ||= {} - @ucd.composition_map[cp.decomp_mapping[0]][cp.decomp_mapping[1]] = cp.code - end - end - end - - def normalize_boundary_map - @ucd.boundary.each do |k,v| - if [:lf, :cr].include? k - @ucd.boundary[k] = v[0] - end - end - end - - def parse - SOURCES.each do |type, url| - filename = File.join(Dir.tmpdir, "#{url.split('/').last}") - unless File.exist?(filename) - $stderr.puts "Downloading #{url.split('/').last}" - File.open(filename, 'wb') do |target| - open(url) do |source| - source.each_line { |line| target.write line } - end - end - end - File.open(filename) do |file| - file.each_line { |line| send "parse_#{type}".intern, line } - end - end - create_composition_map - normalize_boundary_map - end - - def dump_to(filename) - File.open(filename, 'wb') do |f| - f.write Marshal.dump([@ucd.codepoints, @ucd.composition_exclusion, @ucd.composition_map, @ucd.boundary, @ucd.cp1252]) - end - end - end -end - -if __FILE__ == $0 - filename = ActiveSupport::Multibyte::Handlers::UnicodeDatabase.filename - generator = ActiveSupport::Multibyte::Handlers::UnicodeTableGenerator.new - generator.parse - print "Writing to: #{filename}" - generator.dump_to filename - puts " (#{File.size(filename)} bytes)" -end diff --git a/activesupport/lib/active_support/multibyte/handlers/passthru_handler.rb b/activesupport/lib/active_support/multibyte/handlers/passthru_handler.rb deleted file mode 100644 index 916215c2ce..0000000000 --- a/activesupport/lib/active_support/multibyte/handlers/passthru_handler.rb +++ /dev/null @@ -1,9 +0,0 @@ -# Chars uses this handler when $KCODE is not set to 'UTF8'. Because this handler doesn't define any methods all call -# will be forwarded to String. -class ActiveSupport::Multibyte::Handlers::PassthruHandler #:nodoc: - - # Return the original byteoffset - def self.translate_offset(string, byte_offset) #:nodoc: - byte_offset - end -end \ No newline at end of file diff --git a/activesupport/lib/active_support/multibyte/handlers/utf8_handler.rb b/activesupport/lib/active_support/multibyte/handlers/utf8_handler.rb deleted file mode 100644 index aa9c16f575..0000000000 --- a/activesupport/lib/active_support/multibyte/handlers/utf8_handler.rb +++ /dev/null @@ -1,564 +0,0 @@ -# Contains all the handlers and helper classes -module ActiveSupport::Multibyte::Handlers #:nodoc: - class EncodingError < ArgumentError #:nodoc: - end - - class Codepoint #:nodoc: - attr_accessor :code, :combining_class, :decomp_type, :decomp_mapping, :uppercase_mapping, :lowercase_mapping - end - - class UnicodeDatabase #:nodoc: - attr_writer :codepoints, :composition_exclusion, :composition_map, :boundary, :cp1252 - - # self-expiring methods that lazily load the Unicode database and then return the value. - [:codepoints, :composition_exclusion, :composition_map, :boundary, :cp1252].each do |attr_name| - class_eval(<<-EOS, __FILE__, __LINE__) - def #{attr_name} - load - @#{attr_name} - end - EOS - end - - # Shortcut to ucd.codepoints[] - def [](index); codepoints[index]; end - - # Returns the directory in which the data files are stored - def self.dirname - File.dirname(__FILE__) + '/../../values/' - end - - # Returns the filename for the data file for this version - def self.filename - File.expand_path File.join(dirname, "unicode_tables.dat") - end - - # Loads the unicode database and returns all the internal objects of UnicodeDatabase - # Once the values have been loaded, define attr_reader methods for the instance variables. - def load - begin - @codepoints, @composition_exclusion, @composition_map, @boundary, @cp1252 = File.open(self.class.filename, 'rb') { |f| Marshal.load f.read } - rescue Exception => e - raise IOError.new("Couldn't load the unicode tables for UTF8Handler (#{e.message}), handler is unusable") - end - @codepoints ||= Hash.new(Codepoint.new) - @composition_exclusion ||= [] - @composition_map ||= {} - @boundary ||= {} - @cp1252 ||= {} - - # Redefine the === method so we can write shorter rules for grapheme cluster breaks - @boundary.each do |k,_| - @boundary[k].instance_eval do - def ===(other) - detect { |i| i === other } ? true : false - end - end if @boundary[k].kind_of?(Array) - end - - # define attr_reader methods for the instance variables - class << self - attr_reader :codepoints, :composition_exclusion, :composition_map, :boundary, :cp1252 - end - end - end - - # UTF8Handler implements Unicode aware operations for strings, these operations will be used by the Chars - # proxy when $KCODE is set to 'UTF8'. - class UTF8Handler - # Hangul character boundaries and properties - HANGUL_SBASE = 0xAC00 - HANGUL_LBASE = 0x1100 - HANGUL_VBASE = 0x1161 - HANGUL_TBASE = 0x11A7 - HANGUL_LCOUNT = 19 - HANGUL_VCOUNT = 21 - HANGUL_TCOUNT = 28 - HANGUL_NCOUNT = HANGUL_VCOUNT * HANGUL_TCOUNT - HANGUL_SCOUNT = 11172 - HANGUL_SLAST = HANGUL_SBASE + HANGUL_SCOUNT - HANGUL_JAMO_FIRST = 0x1100 - HANGUL_JAMO_LAST = 0x11FF - - # All the unicode whitespace - UNICODE_WHITESPACE = [ - (0x0009..0x000D).to_a, # White_Space # Cc [5] .. - 0x0020, # White_Space # Zs SPACE - 0x0085, # White_Space # Cc - 0x00A0, # White_Space # Zs NO-BREAK SPACE - 0x1680, # White_Space # Zs OGHAM SPACE MARK - 0x180E, # White_Space # Zs MONGOLIAN VOWEL SEPARATOR - (0x2000..0x200A).to_a, # White_Space # Zs [11] EN QUAD..HAIR SPACE - 0x2028, # White_Space # Zl LINE SEPARATOR - 0x2029, # White_Space # Zp PARAGRAPH SEPARATOR - 0x202F, # White_Space # Zs NARROW NO-BREAK SPACE - 0x205F, # White_Space # Zs MEDIUM MATHEMATICAL SPACE - 0x3000, # White_Space # Zs IDEOGRAPHIC SPACE - ].flatten.freeze - - # BOM (byte order mark) can also be seen as whitespace, it's a non-rendering character used to distinguish - # between little and big endian. This is not an issue in utf-8, so it must be ignored. - UNICODE_LEADERS_AND_TRAILERS = UNICODE_WHITESPACE + [65279] # ZERO-WIDTH NO-BREAK SPACE aka BOM - - # Borrowed from the Kconv library by Shinji KONO - (also as seen on the W3C site) - UTF8_PAT = /\A(?: - [\x00-\x7f] | - [\xc2-\xdf] [\x80-\xbf] | - \xe0 [\xa0-\xbf] [\x80-\xbf] | - [\xe1-\xef] [\x80-\xbf] [\x80-\xbf] | - \xf0 [\x90-\xbf] [\x80-\xbf] [\x80-\xbf] | - [\xf1-\xf3] [\x80-\xbf] [\x80-\xbf] [\x80-\xbf] | - \xf4 [\x80-\x8f] [\x80-\xbf] [\x80-\xbf] - )*\z/xn - - # Returns a regular expression pattern that matches the passed Unicode codepoints - def self.codepoints_to_pattern(array_of_codepoints) #:nodoc: - array_of_codepoints.collect{ |e| [e].pack 'U*' }.join('|') - end - UNICODE_TRAILERS_PAT = /(#{codepoints_to_pattern(UNICODE_LEADERS_AND_TRAILERS)})+\Z/ - UNICODE_LEADERS_PAT = /\A(#{codepoints_to_pattern(UNICODE_LEADERS_AND_TRAILERS)})+/ - - class << self - - # /// - # /// BEGIN String method overrides - # /// - - # Inserts the passed string at specified codepoint offsets - def insert(str, offset, fragment) - str.replace( - u_unpack(str).insert( - offset, - u_unpack(fragment) - ).flatten.pack('U*') - ) - end - - # Returns the position of the passed argument in the string, counting in codepoints - def index(str, *args) - bidx = str.index(*args) - bidx ? (u_unpack(str.slice(0...bidx)).size) : nil - end - - # Works just like the indexed replace method on string, except instead of byte offsets you specify - # character offsets. - # - # Example: - # - # s = "Müller" - # s.chars[2] = "e" # Replace character with offset 2 - # s # => "Müeler" - # - # s = "Müller" - # s.chars[1, 2] = "ö" # Replace 2 characters at character offset 1 - # s # => "Möler" - def []=(str, *args) - replace_by = args.pop - # Indexed replace with regular expressions already works - return str[*args] = replace_by if args.first.is_a?(Regexp) - result = u_unpack(str) - if args[0].is_a?(Fixnum) - raise IndexError, "index #{args[0]} out of string" if args[0] >= result.length - min = args[0] - max = args[1].nil? ? min : (min + args[1] - 1) - range = Range.new(min, max) - replace_by = [replace_by].pack('U') if replace_by.is_a?(Fixnum) - elsif args.first.is_a?(Range) - raise RangeError, "#{args[0]} out of range" if args[0].min >= result.length - range = args[0] - else - needle = args[0].to_s - min = index(str, needle) - max = min + length(needle) - 1 - range = Range.new(min, max) - end - result[range] = u_unpack(replace_by) - str.replace(result.pack('U*')) - end - - # Works just like String#rjust, only integer specifies characters instead of bytes. - # - # Example: - # - # "¾ cup".chars.rjust(8).to_s - # # => " ¾ cup" - # - # "¾ cup".chars.rjust(8, " ").to_s # Use non-breaking whitespace - # # => "   ¾ cup" - def rjust(str, integer, padstr=' ') - justify(str, integer, :right, padstr) - end - - # Works just like String#ljust, only integer specifies characters instead of bytes. - # - # Example: - # - # "¾ cup".chars.rjust(8).to_s - # # => "¾ cup " - # - # "¾ cup".chars.rjust(8, " ").to_s # Use non-breaking whitespace - # # => "¾ cup   " - def ljust(str, integer, padstr=' ') - justify(str, integer, :left, padstr) - end - - # Works just like String#center, only integer specifies characters instead of bytes. - # - # Example: - # - # "¾ cup".chars.center(8).to_s - # # => " ¾ cup " - # - # "¾ cup".chars.center(8, " ").to_s # Use non-breaking whitespace - # # => " ¾ cup  " - def center(str, integer, padstr=' ') - justify(str, integer, :center, padstr) - end - - # Does Unicode-aware rstrip - def rstrip(str) - str.gsub(UNICODE_TRAILERS_PAT, '') - end - - # Does Unicode-aware lstrip - def lstrip(str) - str.gsub(UNICODE_LEADERS_PAT, '') - end - - # Removed leading and trailing whitespace - def strip(str) - str.gsub(UNICODE_LEADERS_PAT, '').gsub(UNICODE_TRAILERS_PAT, '') - end - - # Returns the number of codepoints in the string - def size(str) - u_unpack(str).size - end - alias_method :length, :size - - # Reverses codepoints in the string. - def reverse(str) - u_unpack(str).reverse.pack('U*') - end - - # Implements Unicode-aware slice with codepoints. Slicing on one point returns the codepoints for that - # character. - def slice(str, *args) - if args.size > 2 - raise ArgumentError, "wrong number of arguments (#{args.size} for 1)" # Do as if we were native - elsif (args.size == 2 && !(args.first.is_a?(Numeric) || args.first.is_a?(Regexp))) - raise TypeError, "cannot convert #{args.first.class} into Integer" # Do as if we were native - elsif (args.size == 2 && !args[1].is_a?(Numeric)) - raise TypeError, "cannot convert #{args[1].class} into Integer" # Do as if we were native - elsif args[0].kind_of? Range - cps = u_unpack(str).slice(*args) - cps.nil? ? nil : cps.pack('U*') - elsif args[0].kind_of? Regexp - str.slice(*args) - elsif args.size == 1 && args[0].kind_of?(Numeric) - u_unpack(str)[args[0]] - else - u_unpack(str).slice(*args).pack('U*') - end - end - alias_method :[], :slice - - # Convert characters in the string to uppercase - def upcase(str); to_case :uppercase_mapping, str; end - - # Convert characters in the string to lowercase - def downcase(str); to_case :lowercase_mapping, str; end - - # Returns a copy of +str+ with the first character converted to uppercase and the remainder to lowercase - def capitalize(str) - upcase(slice(str, 0..0)) + downcase(slice(str, 1..-1) || '') - end - - # /// - # /// Extra String methods for unicode operations - # /// - - # Returns the KC normalization of the string by default. NFKC is considered the best normalization form for - # passing strings to databases and validations. - # - # * str - The string to perform normalization on. - # * form - The form you want to normalize in. Should be one of the following: - # :c, :kc, :d, or :kd. Default is - # ActiveSupport::Multibyte::DEFAULT_NORMALIZATION_FORM. - def normalize(str, form=ActiveSupport::Multibyte::DEFAULT_NORMALIZATION_FORM) - # See http://www.unicode.org/reports/tr15, Table 1 - codepoints = u_unpack(str) - case form - when :d - reorder_characters(decompose_codepoints(:canonical, codepoints)) - when :c - compose_codepoints reorder_characters(decompose_codepoints(:canonical, codepoints)) - when :kd - reorder_characters(decompose_codepoints(:compatability, codepoints)) - when :kc - compose_codepoints reorder_characters(decompose_codepoints(:compatability, codepoints)) - else - raise ArgumentError, "#{form} is not a valid normalization variant", caller - end.pack('U*') - end - - # Perform decomposition on the characters in the string - def decompose(str) - decompose_codepoints(:canonical, u_unpack(str)).pack('U*') - end - - # Perform composition on the characters in the string - def compose(str) - compose_codepoints u_unpack(str).pack('U*') - end - - # /// - # /// BEGIN Helper methods for unicode operation - # /// - - # Used to translate an offset from bytes to characters, for instance one received from a regular expression match - def translate_offset(str, byte_offset) - return nil if byte_offset.nil? - return 0 if str == '' - chunk = str[0..byte_offset] - begin - begin - chunk.unpack('U*').length - 1 - rescue ArgumentError => e - chunk = str[0..(byte_offset+=1)] - # Stop retrying at the end of the string - raise e unless byte_offset < chunk.length - # We damaged a character, retry - retry - end - # Catch the ArgumentError so we can throw our own - rescue ArgumentError - raise EncodingError.new('malformed UTF-8 character') - end - end - - # Checks if the string is valid UTF8. - def consumes?(str) - # Unpack is a little bit faster than regular expressions - begin - str.unpack('U*') - true - rescue ArgumentError - false - end - end - - # Returns the number of grapheme clusters in the string. This method is very likely to be moved or renamed - # in future versions. - def g_length(str) - g_unpack(str).length - end - - # Replaces all the non-utf-8 bytes by their iso-8859-1 or cp1252 equivalent resulting in a valid utf-8 string - def tidy_bytes(str) - str.split(//u).map do |c| - if !UTF8_PAT.match(c) - n = c.unpack('C')[0] - n < 128 ? n.chr : - n < 160 ? [UCD.cp1252[n] || n].pack('U') : - n < 192 ? "\xC2" + n.chr : "\xC3" + (n-64).chr - else - c - end - end.join - end - - protected - - # Detect whether the codepoint is in a certain character class. Primarily used by the - # grapheme cluster support. - def in_char_class?(codepoint, classes) - classes.detect { |c| UCD.boundary[c] === codepoint } ? true : false - end - - # Unpack the string at codepoints boundaries - def u_unpack(str) - begin - str.unpack 'U*' - rescue ArgumentError - raise EncodingError.new('malformed UTF-8 character') - end - end - - # Unpack the string at grapheme boundaries instead of codepoint boundaries - def g_unpack(str) - codepoints = u_unpack(str) - unpacked = [] - pos = 0 - marker = 0 - eoc = codepoints.length - while(pos < eoc) - pos += 1 - previous = codepoints[pos-1] - current = codepoints[pos] - if ( - # CR X LF - one = ( previous == UCD.boundary[:cr] and current == UCD.boundary[:lf] ) or - # L X (L|V|LV|LVT) - two = ( UCD.boundary[:l] === previous and in_char_class?(current, [:l,:v,:lv,:lvt]) ) or - # (LV|V) X (V|T) - three = ( in_char_class?(previous, [:lv,:v]) and in_char_class?(current, [:v,:t]) ) or - # (LVT|T) X (T) - four = ( in_char_class?(previous, [:lvt,:t]) and UCD.boundary[:t] === current ) or - # X Extend - five = (UCD.boundary[:extend] === current) - ) - else - unpacked << codepoints[marker..pos-1] - marker = pos - end - end - unpacked - end - - # Reverse operation of g_unpack - def g_pack(unpacked) - unpacked.flatten - end - - # Justifies a string in a certain way. Valid values for way are :right, :left and - # :center. Is primarily used as a helper method by rjust, ljust and center. - def justify(str, integer, way, padstr=' ') - raise ArgumentError, "zero width padding" if padstr.length == 0 - padsize = integer - size(str) - padsize = padsize > 0 ? padsize : 0 - case way - when :right - str.dup.insert(0, padding(padsize, padstr)) - when :left - str.dup.insert(-1, padding(padsize, padstr)) - when :center - lpad = padding((padsize / 2.0).floor, padstr) - rpad = padding((padsize / 2.0).ceil, padstr) - str.dup.insert(0, lpad).insert(-1, rpad) - end - end - - # Generates a padding string of a certain size. - def padding(padsize, padstr=' ') - if padsize != 0 - slice(padstr * ((padsize / size(padstr)) + 1), 0, padsize) - else - '' - end - end - - # Convert characters to a different case - def to_case(way, str) - u_unpack(str).map do |codepoint| - cp = UCD[codepoint] - unless cp.nil? - ncp = cp.send(way) - ncp > 0 ? ncp : codepoint - else - codepoint - end - end.pack('U*') - end - - # Re-order codepoints so the string becomes canonical - def reorder_characters(codepoints) - length = codepoints.length- 1 - pos = 0 - while pos < length do - cp1, cp2 = UCD[codepoints[pos]], UCD[codepoints[pos+1]] - if (cp1.combining_class > cp2.combining_class) && (cp2.combining_class > 0) - codepoints[pos..pos+1] = cp2.code, cp1.code - pos += (pos > 0 ? -1 : 1) - else - pos += 1 - end - end - codepoints - end - - # Decompose composed characters to the decomposed form - def decompose_codepoints(type, codepoints) - codepoints.inject([]) do |decomposed, cp| - # if it's a hangul syllable starter character - if HANGUL_SBASE <= cp and cp < HANGUL_SLAST - sindex = cp - HANGUL_SBASE - ncp = [] # new codepoints - ncp << HANGUL_LBASE + sindex / HANGUL_NCOUNT - ncp << HANGUL_VBASE + (sindex % HANGUL_NCOUNT) / HANGUL_TCOUNT - tindex = sindex % HANGUL_TCOUNT - ncp << (HANGUL_TBASE + tindex) unless tindex == 0 - decomposed.concat ncp - # if the codepoint is decomposable in with the current decomposition type - elsif (ncp = UCD[cp].decomp_mapping) and (!UCD[cp].decomp_type || type == :compatability) - decomposed.concat decompose_codepoints(type, ncp.dup) - else - decomposed << cp - end - end - end - - # Compose decomposed characters to the composed form - def compose_codepoints(codepoints) - pos = 0 - eoa = codepoints.length - 1 - starter_pos = 0 - starter_char = codepoints[0] - previous_combining_class = -1 - while pos < eoa - pos += 1 - lindex = starter_char - HANGUL_LBASE - # -- Hangul - if 0 <= lindex and lindex < HANGUL_LCOUNT - vindex = codepoints[starter_pos+1] - HANGUL_VBASE rescue vindex = -1 - if 0 <= vindex and vindex < HANGUL_VCOUNT - tindex = codepoints[starter_pos+2] - HANGUL_TBASE rescue tindex = -1 - if 0 <= tindex and tindex < HANGUL_TCOUNT - j = starter_pos + 2 - eoa -= 2 - else - tindex = 0 - j = starter_pos + 1 - eoa -= 1 - end - codepoints[starter_pos..j] = (lindex * HANGUL_VCOUNT + vindex) * HANGUL_TCOUNT + tindex + HANGUL_SBASE - end - starter_pos += 1 - starter_char = codepoints[starter_pos] - # -- Other characters - else - current_char = codepoints[pos] - current = UCD[current_char] - if current.combining_class > previous_combining_class - if ref = UCD.composition_map[starter_char] - composition = ref[current_char] - else - composition = nil - end - unless composition.nil? - codepoints[starter_pos] = composition - starter_char = composition - codepoints.delete_at pos - eoa -= 1 - pos -= 1 - previous_combining_class = -1 - else - previous_combining_class = current.combining_class - end - else - previous_combining_class = current.combining_class - end - if current.combining_class == 0 - starter_pos = pos - starter_char = codepoints[pos] - end - end - end - codepoints - end - - # UniCode Database - UCD = UnicodeDatabase.new - end - end -end diff --git a/activesupport/lib/active_support/multibyte/handlers/utf8_handler_proc.rb b/activesupport/lib/active_support/multibyte/handlers/utf8_handler_proc.rb deleted file mode 100644 index f10eecc622..0000000000 --- a/activesupport/lib/active_support/multibyte/handlers/utf8_handler_proc.rb +++ /dev/null @@ -1,43 +0,0 @@ -# Methods in this handler call functions in the utf8proc ruby extension. These are significantly faster than the -# pure ruby versions. Chars automatically uses this handler when it can load the utf8proc extension. For -# documentation on handler methods see UTF8Handler. -class ActiveSupport::Multibyte::Handlers::UTF8HandlerProc < ActiveSupport::Multibyte::Handlers::UTF8Handler #:nodoc: - class << self - def normalize(str, form=ActiveSupport::Multibyte::DEFAULT_NORMALIZATION_FORM) #:nodoc: - codepoints = str.unpack('U*') - case form - when :d - utf8map(str, :stable) - when :c - utf8map(str, :stable, :compose) - when :kd - utf8map(str, :stable, :compat) - when :kc - utf8map(str, :stable, :compose, :compat) - else - raise ArgumentError, "#{form} is not a valid normalization variant", caller - end - end - - def decompose(str) #:nodoc: - utf8map(str, :stable) - end - - def downcase(str) #:nodoc:c - utf8map(str, :casefold) - end - - protected - - def utf8map(str, *option_array) #:nodoc: - options = 0 - option_array.each do |option| - flag = Utf8Proc::Options[option] - raise ArgumentError, "Unknown argument given to utf8map." unless - flag - options |= flag - end - return Utf8Proc::utf8map(str, options) - end - end -end diff --git a/activesupport/lib/active_support/multibyte/unicode_database.rb b/activesupport/lib/active_support/multibyte/unicode_database.rb new file mode 100644 index 0000000000..3b8cf8f9eb --- /dev/null +++ b/activesupport/lib/active_support/multibyte/unicode_database.rb @@ -0,0 +1,71 @@ +# encoding: utf-8 + +module ActiveSupport #:nodoc: + module Multibyte #:nodoc: + # Holds data about a codepoint in the Unicode database + class Codepoint + attr_accessor :code, :combining_class, :decomp_type, :decomp_mapping, :uppercase_mapping, :lowercase_mapping + end + + # Holds static data from the Unicode database + class UnicodeDatabase + ATTRIBUTES = :codepoints, :composition_exclusion, :composition_map, :boundary, :cp1252 + + attr_writer(*ATTRIBUTES) + + def initialize + @codepoints = Hash.new(Codepoint.new) + @composition_exclusion = [] + @composition_map = {} + @boundary = {} + @cp1252 = {} + end + + # Lazy load the Unicode database so it's only loaded when it's actually used + ATTRIBUTES.each do |attr_name| + class_eval(<<-EOS, __FILE__, __LINE__) + def #{attr_name} + load + @#{attr_name} + end + EOS + end + + # Loads the Unicode database and returns all the internal objects of UnicodeDatabase. + def load + begin + @codepoints, @composition_exclusion, @composition_map, @boundary, @cp1252 = File.open(self.class.filename, 'rb') { |f| Marshal.load f.read } + rescue Exception => e + raise IOError.new("Couldn't load the Unicode tables for UTF8Handler (#{e.message}), ActiveSupport::Multibyte is unusable") + end + + # Redefine the === method so we can write shorter rules for grapheme cluster breaks + @boundary.each do |k,_| + @boundary[k].instance_eval do + def ===(other) + detect { |i| i === other } ? true : false + end + end if @boundary[k].kind_of?(Array) + end + + # define attr_reader methods for the instance variables + class << self + attr_reader(*ATTRIBUTES) + end + end + + # Returns the directory in which the data files are stored + def self.dirname + File.dirname(__FILE__) + '/../values/' + end + + # Returns the filename for the data file for this version + def self.filename + File.expand_path File.join(dirname, "unicode_tables.dat") + end + end + + # UniCode Database + UCD = UnicodeDatabase.new + end +end \ No newline at end of file diff --git a/activesupport/lib/active_support/values/unicode_tables.dat b/activesupport/lib/active_support/values/unicode_tables.dat index 35edb148c3..74b333d416 100644 Binary files a/activesupport/lib/active_support/values/unicode_tables.dat and b/activesupport/lib/active_support/values/unicode_tables.dat differ diff --git a/activesupport/test/abstract_unit.rb b/activesupport/test/abstract_unit.rb index cce8d5d220..f39f264ae1 100644 --- a/activesupport/test/abstract_unit.rb +++ b/activesupport/test/abstract_unit.rb @@ -1,9 +1,15 @@ +# encoding: utf-8 + require 'test/unit' $:.unshift "#{File.dirname(__FILE__)}/../lib" $:.unshift File.dirname(__FILE__) require 'active_support' +if RUBY_VERSION < '1.9' + $KCODE = 'UTF8' +end + def uses_gem(gem_name, test_name, version = '> 0') require 'rubygems' gem gem_name.to_s, version @@ -21,4 +27,4 @@ unless defined? uses_mocha end # Show backtraces for deprecated behavior for quicker cleanup. -ActiveSupport::Deprecation.debug = true +ActiveSupport::Deprecation.debug = true \ No newline at end of file diff --git a/activesupport/test/multibyte_chars_test.rb b/activesupport/test/multibyte_chars_test.rb index a87309b038..31b8f1b760 100644 --- a/activesupport/test/multibyte_chars_test.rb +++ b/activesupport/test/multibyte_chars_test.rb @@ -1,186 +1,528 @@ # encoding: utf-8 + require 'abstract_unit' -if RUBY_VERSION < '1.9' +module MultibyteTest + UNICODE_STRING = 'こにちわ' + ASCII_STRING = 'ohayo' + BYTE_STRING = "\270\236\010\210\245" + + def chars(str) + ActiveSupport::Multibyte::Chars.new(str) + end + + def inspect_codepoints(str) + str.to_s.unpack("U*").map{|cp| cp.to_s(16) }.join(' ') + end + + def assert_equal_codepoints(expected, actual, message=nil) + assert_equal(inspect_codepoints(expected), inspect_codepoints(actual), message) + end +end + +class String + def __string_for_multibyte_testing; self; end + def __string_for_multibyte_testing!; self; end +end -$KCODE = 'UTF8' +class MultibyteCharsTest < Test::Unit::TestCase + include MultibyteTest -class CharsTest < Test::Unit::TestCase - def setup - @s = { - :utf8 => "Abcd Блå ffi блa 埋", - :ascii => "asci ias c iia s", - :bytes => "\270\236\010\210\245" - } + @chars = ActiveSupport::Multibyte::Chars.new UNICODE_STRING end - - def test_sanity - @s.each do |t, s| - assert s.respond_to?(:chars), "All string should have the chars method (#{t})" - assert s.respond_to?(:to_s), "All string should have the to_s method (#{t})" - assert_kind_of ActiveSupport::Multibyte::Chars, s.chars, "#chars should return an instance of Chars (#{t})" - end + + def test_wraps_the_original_string + assert_equal UNICODE_STRING, @chars.to_s + assert_equal UNICODE_STRING, @chars.wrapped_string end - - def test_comparability - @s.each do |t, s| - assert_equal s, s.chars.to_s, "Chars#to_s should return enclosed string unchanged" - end + + def test_should_allow_method_calls_to_string assert_nothing_raised do - assert_equal "a", "a", "Normal string comparisons should be unaffected" - assert_not_equal "a", "b", "Normal string comparisons should be unaffected" - assert_not_equal "a".chars, "b".chars, "Chars objects should be comparable" - assert_equal "a".chars, "A".downcase.chars, "Chars objects should be comparable to each other" - assert_equal "a".chars, "A".downcase, "Chars objects should be comparable to strings coming from elsewhere" + @chars.__string_for_multibyte_testing end - - assert !@s[:utf8].eql?(@s[:utf8].chars), "Strict comparison is not supported" - assert_equal @s[:utf8], @s[:utf8].chars, "Chars should be compared by their enclosed string" - - other_string = @s[:utf8].dup - assert_equal other_string, @s[:utf8].chars, "Chars should be compared by their enclosed string" - assert_equal other_string.chars, @s[:utf8].chars, "Chars should be compared by their enclosed string" - - strings = ['builder'.chars, 'armor'.chars, 'zebra'.chars] - strings.sort! - assert_equal ['armor', 'builder', 'zebra'], strings, "Chars should be sortable based on their enclosed string" - - # This leads to a StackLevelTooDeep exception if the comparison is not wired properly - assert_raise(NameError) do - Chars + assert_raises NoMethodError do + @chars.__unknown_method end end - - def test_utf8? - assert @s[:utf8].is_utf8?, "UTF-8 strings are UTF-8" - assert @s[:ascii].is_utf8?, "All ASCII strings are also valid UTF-8" - assert !@s[:bytes].is_utf8?, "This bytestring isn't UTF-8" - end - - # The test for the following methods are defined here because they can only be defined on the Chars class for - # various reasons - - def test_gsub - assert_equal 'éxa', 'éda'.chars.gsub(/d/, 'x') - with_kcode('none') do - assert_equal 'éxa', 'éda'.chars.gsub(/d/, 'x') - end + + def test_forwarded_method_calls_should_return_new_chars_instance + assert @chars.__string_for_multibyte_testing.kind_of?(ActiveSupport::Multibyte::Chars) + assert_not_equal @chars.object_id, @chars.__string_for_multibyte_testing.object_id end - - def test_split - word = "efficient" - chars = ["e", "ffi", "c", "i", "e", "n", "t"] - assert_equal chars, word.split(//) - assert_equal chars, word.chars.split(//) - assert_kind_of ActiveSupport::Multibyte::Chars, word.chars.split(//).first, "Split should return Chars instances" - end - - def test_regexp - with_kcode('none') do - assert_equal 12, (@s[:utf8].chars =~ /ffi/), - "Regex matching should be bypassed to String" + + def test_forwarded_bang_method_calls_should_return_the_original_chars_instance + assert @chars.__string_for_multibyte_testing!.kind_of?(ActiveSupport::Multibyte::Chars) + assert_equal @chars.object_id, @chars.__string_for_multibyte_testing!.object_id + end + + def test_methods_are_forwarded_to_wrapped_string_for_byte_strings + assert_equal BYTE_STRING.class, BYTE_STRING.mb_chars.class + end + + def test_should_concatenate + assert_equal 'ab', 'a'.mb_chars + 'b' + assert_equal 'ab', 'a' + 'b'.mb_chars + assert_equal 'ab', 'a'.mb_chars + 'b'.mb_chars + + assert_equal 'ab', 'a'.mb_chars << 'b' + assert_equal 'ab', 'a' << 'b'.mb_chars + assert_equal 'ab', 'a'.mb_chars << 'b'.mb_chars + end + + if RUBY_VERSION < '1.9' + def test_concatenation_should_return_a_proxy_class_instance + assert_equal ActiveSupport::Multibyte.proxy_class, ('a'.mb_chars + 'b').class + assert_equal ActiveSupport::Multibyte.proxy_class, ('a'.mb_chars << 'b').class end - with_kcode('UTF8') do - assert_equal 9, (@s[:utf8].chars =~ /ffi/), - "Regex matching should be unicode aware" - assert_nil((''.chars =~ /\d+/), - "Non-matching regular expressions should return nil") + + def test_ascii_strings_are_treated_at_utf8_strings + assert_equal ActiveSupport::Multibyte.proxy_class, ASCII_STRING.mb_chars.class + end + + def test_concatenate_should_return_proxy_instance + assert(('a'.mb_chars + 'b').kind_of?(ActiveSupport::Multibyte::Chars)) + assert(('a'.mb_chars + 'b'.mb_chars).kind_of?(ActiveSupport::Multibyte::Chars)) + assert(('a'.mb_chars << 'b').kind_of?(ActiveSupport::Multibyte::Chars)) + assert(('a'.mb_chars << 'b'.mb_chars).kind_of?(ActiveSupport::Multibyte::Chars)) end end +end - def test_pragma - if RUBY_VERSION < '1.9' - with_kcode('UTF8') do - assert " ".chars.send(:utf8_pragma?), "UTF8 pragma should be on because KCODE is UTF8" - end - with_kcode('none') do - assert !" ".chars.send(:utf8_pragma?), "UTF8 pragma should be off because KCODE is not UTF8" +class MultibyteCharsUTF8BehaviourTest < Test::Unit::TestCase + include MultibyteTest + + def setup + @chars = UNICODE_STRING.dup.mb_chars + + # NEWLINE, SPACE, EM SPACE + @whitespace = "\n#{[32, 8195].pack('U*')}" + @whitespace.force_encoding(Encoding::UTF_8) if @whitespace.respond_to?(:force_encoding) + @byte_order_mark = [65279].pack('U') + end + + if RUBY_VERSION < '1.9' + def test_split_should_return_an_array_of_chars_instances + @chars.split(//).each do |character| + assert character.kind_of?(ActiveSupport::Multibyte::Chars) end - else - assert !" ".chars.send(:utf8_pragma?), "UTF8 pragma should be off in Ruby 1.9" end - end - def test_handler_setting - handler = ''.chars.handler - - ActiveSupport::Multibyte::Chars.handler = :first - assert_equal :first, ''.chars.handler - ActiveSupport::Multibyte::Chars.handler = :second - assert_equal :second, ''.chars.handler - assert_raise(NoMethodError) do - ''.chars.handler.split + def test_indexed_insert_accepts_fixnums + @chars[2] = 32 + assert_equal 'こに わ', @chars end - - ActiveSupport::Multibyte::Chars.handler = handler - end - - def test_method_chaining - assert_kind_of ActiveSupport::Multibyte::Chars, ''.chars.downcase - assert_kind_of ActiveSupport::Multibyte::Chars, ''.chars.strip, "Strip should return a Chars object" - assert_kind_of ActiveSupport::Multibyte::Chars, ''.chars.downcase.strip, "The Chars object should be " + - "forwarded down the call path for chaining" - assert_equal 'foo', " FOO ".chars.normalize.downcase.strip, "The Chars that results from the " + - " operations should be comparable to the string value of the result" - end - - def test_passthrough_on_kcode - # The easiest way to check if the passthrough is in place is through #size - with_kcode('none') do - assert_equal 26, @s[:utf8].chars.size + + def test_overridden_bang_methods_return_self + [:rstrip!, :lstrip!, :strip!, :reverse!, :upcase!, :downcase!, :capitalize!].each do |method| + assert_equal @chars.object_id, @chars.send(method).object_id + end + assert_equal @chars.object_id, @chars.slice!(1).object_id end - with_kcode('UTF8') do - assert_equal 17, @s[:utf8].chars.size + + def test_overridden_bang_methods_change_wrapped_string + [:rstrip!, :lstrip!, :strip!, :reverse!, :upcase!, :downcase!].each do |method| + original = ' Café ' + proxy = chars(original.dup) + proxy.send(method) + assert_not_equal original, proxy.to_s + end + proxy = chars('Café') + proxy.slice!(3) + assert_equal 'é', proxy.to_s + + proxy = chars('òu') + proxy.capitalize! + assert_equal 'Òu', proxy.to_s end end - - def test_destructiveness - # Note that we're testing the destructiveness here and not the correct behaviour of the methods - str = 'ac' - str.chars.insert(1, 'b') - assert_equal 'abc', str, 'Insert should be destructive for a string' - - str = 'ac' - str.chars.reverse! - assert_equal 'ca', str, 'reverse! should be destructive for a string' - end - - def test_resilience - assert_nothing_raised do - assert_equal 5, @s[:bytes].chars.size, "The sequence contains five interpretable bytes" - end - reversed = [0xb8, 0x17e, 0x8, 0x2c6, 0xa5].reverse.pack('U*') - assert_nothing_raised do - assert_equal reversed, @s[:bytes].chars.reverse.to_s, "Reversing the string should only yield interpretable bytes" + + if RUBY_VERSION >= '1.9' + def test_unicode_string_should_have_utf8_encoding + assert_equal Encoding::UTF_8, UNICODE_STRING.encoding end - assert_nothing_raised do - @s[:bytes].chars.reverse! - assert_equal reversed, @s[:bytes].to_s, "Reversing the string should only yield interpretable bytes" + end + + def test_should_be_equal_to_the_wrapped_string + assert_equal UNICODE_STRING, @chars + assert_equal @chars, UNICODE_STRING + end + + def test_should_not_be_equal_to_an_other_string + assert_not_equal @chars, 'other' + assert_not_equal 'other', @chars + end + + def test_should_return_character_offset_for_regexp_matches + assert_nil(@chars =~ /wrong/u) + assert_equal 0, (@chars =~ /こ/u) + assert_equal 1, (@chars =~ /に/u) + assert_equal 3, (@chars =~ /わ/u) + end + + def test_should_use_character_offsets_for_insert_offsets + assert_equal '', ''.mb_chars.insert(0, '') + assert_equal 'こわにちわ', @chars.insert(1, 'わ') + assert_equal 'こわわわにちわ', @chars.insert(2, 'わわ') + assert_equal 'わこわわわにちわ', @chars.insert(0, 'わ') + assert_equal 'わこわわわにちわ', @chars.wrapped_string if RUBY_VERSION < '1.9' + end + + def test_insert_should_be_destructive + @chars.insert(1, 'わ') + assert_equal 'こわにちわ', @chars + end + + def test_insert_throws_index_error + assert_raises(IndexError) { @chars.insert(12, 'わ') } + end + + def test_should_know_if_one_includes_the_other + assert @chars.include?('') + assert @chars.include?('ち') + assert @chars.include?('わ') + assert !@chars.include?('こちわ') + assert !@chars.include?('a') + end + + def test_include_raises_type_error_when_nil_is_passed + assert_raises(TypeError) do + @chars.include?(nil) end end - - def test_duck_typing - assert_equal true, 'test'.chars.respond_to?(:strip) - assert_equal true, 'test'.chars.respond_to?(:normalize) - assert_equal true, 'test'.chars.respond_to?(:normalize!) - assert_equal false, 'test'.chars.respond_to?(:a_method_that_doesnt_exist) + + def test_index_should_return_character_offset + assert_nil @chars.index('u') + assert_equal 0, @chars.index('こに') + assert_equal 2, @chars.index('ち') + assert_equal 3, @chars.index('わ') + end + + def test_indexed_insert_should_take_character_offsets + @chars[2] = 'a' + assert_equal 'こにaわ', @chars + @chars[2] = 'ηη' + assert_equal 'こにηηわ', @chars + @chars[3, 2] = 'λλλ' + assert_equal 'こにηλλλ', @chars + @chars[1, 0] = "λ" + assert_equal 'こλにηλλλ', @chars + @chars[4..6] = "ηη" + assert_equal 'こλにηηη', @chars + @chars[/ηη/] = "λλλ" + assert_equal 'こλにλλλη', @chars + @chars[/(λλ)(.)/, 2] = "α" + assert_equal 'こλにλλαη', @chars + @chars["α"] = "¢" + assert_equal 'こλにλλ¢η', @chars + @chars["λλ"] = "ααα" + assert_equal 'こλにααα¢η', @chars + end + + def test_indexed_insert_should_raise_on_index_overflow + before = @chars.to_s + assert_raises(IndexError) { @chars[10] = 'a' } + assert_raises(IndexError) { @chars[10, 4] = 'a' } + assert_raises(IndexError) { @chars[/ii/] = 'a' } + assert_raises(IndexError) { @chars[/()/, 10] = 'a' } + assert_equal before, @chars + end + + def test_indexed_insert_should_raise_on_range_overflow + before = @chars.to_s + assert_raises(RangeError) { @chars[10..12] = 'a' } + assert_equal before, @chars end - + + def test_rjust_should_raise_argument_errors_on_bad_arguments + assert_raises(ArgumentError) { @chars.rjust(10, '') } + assert_raises(ArgumentError) { @chars.rjust } + end + + def test_rjust_should_count_characters_instead_of_bytes + assert_equal UNICODE_STRING, @chars.rjust(-3) + assert_equal UNICODE_STRING, @chars.rjust(0) + assert_equal UNICODE_STRING, @chars.rjust(4) + assert_equal " #{UNICODE_STRING}", @chars.rjust(5) + assert_equal " #{UNICODE_STRING}", @chars.rjust(7) + assert_equal "---#{UNICODE_STRING}", @chars.rjust(7, '-') + assert_equal "ααα#{UNICODE_STRING}", @chars.rjust(7, 'α') + assert_equal "aba#{UNICODE_STRING}", @chars.rjust(7, 'ab') + assert_equal "αηα#{UNICODE_STRING}", @chars.rjust(7, 'αη') + assert_equal "αηαη#{UNICODE_STRING}", @chars.rjust(8, 'αη') + end + + def test_ljust_should_raise_argument_errors_on_bad_arguments + assert_raises(ArgumentError) { @chars.ljust(10, '') } + assert_raises(ArgumentError) { @chars.ljust } + end + + def test_ljust_should_count_characters_instead_of_bytes + assert_equal UNICODE_STRING, @chars.ljust(-3) + assert_equal UNICODE_STRING, @chars.ljust(0) + assert_equal UNICODE_STRING, @chars.ljust(4) + assert_equal "#{UNICODE_STRING} ", @chars.ljust(5) + assert_equal "#{UNICODE_STRING} ", @chars.ljust(7) + assert_equal "#{UNICODE_STRING}---", @chars.ljust(7, '-') + assert_equal "#{UNICODE_STRING}ααα", @chars.ljust(7, 'α') + assert_equal "#{UNICODE_STRING}aba", @chars.ljust(7, 'ab') + assert_equal "#{UNICODE_STRING}αηα", @chars.ljust(7, 'αη') + assert_equal "#{UNICODE_STRING}αηαη", @chars.ljust(8, 'αη') + end + + def test_center_should_raise_argument_errors_on_bad_arguments + assert_raises(ArgumentError) { @chars.center(10, '') } + assert_raises(ArgumentError) { @chars.center } + end + + def test_center_should_count_charactes_instead_of_bytes + assert_equal UNICODE_STRING, @chars.center(-3) + assert_equal UNICODE_STRING, @chars.center(0) + assert_equal UNICODE_STRING, @chars.center(4) + assert_equal "#{UNICODE_STRING} ", @chars.center(5) + assert_equal " #{UNICODE_STRING} ", @chars.center(6) + assert_equal " #{UNICODE_STRING} ", @chars.center(7) + assert_equal "--#{UNICODE_STRING}--", @chars.center(8, '-') + assert_equal "--#{UNICODE_STRING}---", @chars.center(9, '-') + assert_equal "αα#{UNICODE_STRING}αα", @chars.center(8, 'α') + assert_equal "αα#{UNICODE_STRING}ααα", @chars.center(9, 'α') + assert_equal "a#{UNICODE_STRING}ab", @chars.center(7, 'ab') + assert_equal "ab#{UNICODE_STRING}ab", @chars.center(8, 'ab') + assert_equal "abab#{UNICODE_STRING}abab", @chars.center(12, 'ab') + assert_equal "α#{UNICODE_STRING}αη", @chars.center(7, 'αη') + assert_equal "αη#{UNICODE_STRING}αη", @chars.center(8, 'αη') + end + + def test_lstrip_strips_whitespace_from_the_left_of_the_string + assert_equal UNICODE_STRING, UNICODE_STRING.mb_chars.lstrip + assert_equal UNICODE_STRING, (@whitespace + UNICODE_STRING).mb_chars.lstrip + assert_equal UNICODE_STRING + @whitespace, (@whitespace + UNICODE_STRING + @whitespace).mb_chars.lstrip + end + + def test_rstrip_strips_whitespace_from_the_right_of_the_string + assert_equal UNICODE_STRING, UNICODE_STRING.mb_chars.rstrip + assert_equal UNICODE_STRING, (UNICODE_STRING + @whitespace).mb_chars.rstrip + assert_equal @whitespace + UNICODE_STRING, (@whitespace + UNICODE_STRING + @whitespace).mb_chars.rstrip + end + + def test_strip_strips_whitespace + assert_equal UNICODE_STRING, UNICODE_STRING.mb_chars.strip + assert_equal UNICODE_STRING, (@whitespace + UNICODE_STRING).mb_chars.strip + assert_equal UNICODE_STRING, (UNICODE_STRING + @whitespace).mb_chars.strip + assert_equal UNICODE_STRING, (@whitespace + UNICODE_STRING + @whitespace).mb_chars.strip + end + + def test_stripping_whitespace_leaves_whitespace_within_the_string_intact + string_with_whitespace = UNICODE_STRING + @whitespace + UNICODE_STRING + assert_equal string_with_whitespace, string_with_whitespace.mb_chars.strip + assert_equal string_with_whitespace, string_with_whitespace.mb_chars.lstrip + assert_equal string_with_whitespace, string_with_whitespace.mb_chars.rstrip + end + + def test_size_returns_characters_instead_of_bytes + assert_equal 0, ''.mb_chars.size + assert_equal 4, @chars.size + assert_equal 4, @chars.length + assert_equal 5, ASCII_STRING.mb_chars.size + end + + def test_reverse_reverses_characters + assert_equal '', ''.mb_chars.reverse + assert_equal 'わちにこ', @chars.reverse + end + + def test_slice_should_take_character_offsets + assert_equal nil, ''.mb_chars.slice(0) + assert_equal 'こ', @chars.slice(0) + assert_equal 'わ', @chars.slice(3) + assert_equal nil, ''.mb_chars.slice(-1..1) + assert_equal '', ''.mb_chars.slice(0..10) + assert_equal 'にちわ', @chars.slice(1..3) + assert_equal 'にちわ', @chars.slice(1, 3) + assert_equal 'こ', @chars.slice(0, 1) + assert_equal 'ちわ', @chars.slice(2..10) + assert_equal '', @chars.slice(4..10) + assert_equal 'に', @chars.slice(/に/u) + assert_equal 'にち', @chars.slice(/に\w/u) + assert_equal nil, @chars.slice(/unknown/u) + assert_equal 'にち', @chars.slice(/(にち)/u, 1) + assert_equal nil, @chars.slice(/(にち)/u, 2) + assert_equal nil, @chars.slice(7..6) + end + + def test_slice_should_throw_exceptions_on_invalid_arguments + assert_raise(TypeError) { @chars.slice(2..3, 1) } + assert_raise(TypeError) { @chars.slice(1, 2..3) } + assert_raise(ArgumentError) { @chars.slice(1, 1, 1) } + end + + def test_upcase_should_upcase_ascii_characters + assert_equal '', ''.mb_chars.upcase + assert_equal 'ABC', 'aBc'.mb_chars.upcase + end + + def test_downcase_should_downcase_ascii_characters + assert_equal '', ''.mb_chars.downcase + assert_equal 'abc', 'aBc'.mb_chars.downcase + end + + def test_capitalize_should_work_on_ascii_characters + assert_equal '', ''.mb_chars.capitalize + assert_equal 'Abc', 'abc'.mb_chars.capitalize + end + + def test_respond_to_knows_which_methods_the_proxy_responds_to + assert ''.mb_chars.respond_to?(:slice) # Defined on Chars + assert ''.mb_chars.respond_to?(:capitalize!) # Defined on Chars + assert ''.mb_chars.respond_to?(:gsub) # Defined on String + assert !''.mb_chars.respond_to?(:undefined_method) # Not defined + end + def test_acts_like_string - assert 'Bambi'.chars.acts_like_string? + assert 'Bambi'.mb_chars.acts_like_string? end +end - protected +# The default Multibyte Chars proxy has more features than the normal string implementation. Tests +# for the implementation of these features should run on all Ruby versions and shouldn't be tested +# through the proxy methods. +class MultibyteCharsExtrasTest < Test::Unit::TestCase + include MultibyteTest - def with_kcode(kcode) - old_kcode, $KCODE = $KCODE, kcode - begin - yield - ensure - $KCODE = old_kcode + if RUBY_VERSION >= '1.9' + def test_tidy_bytes_is_broken_on_1_9_0 + assert_raises(ArgumentError) do + assert_equal_codepoints [0xfffd].pack('U'), chars("\xef\xbf\xbd").tidy_bytes + end end end -end + def test_upcase_should_be_unicode_aware + assert_equal "АБВГД\0F", chars("аБвгд\0f").upcase + assert_equal 'こにちわ', chars('こにちわ').upcase + end + + def test_downcase_should_be_unicode_aware + assert_equal "абвгд\0f", chars("аБвгд\0f").downcase + assert_equal 'こにちわ', chars('こにちわ').downcase + end + + def test_capitalize_should_be_unicode_aware + { 'аБвг аБвг' => 'Абвг абвг', + 'аБвг АБВГ' => 'Абвг абвг', + 'АБВГ АБВГ' => 'Абвг абвг', + '' => '' }.each do |f,t| + assert_equal t, chars(f).capitalize + end + end + + def test_composition_exclusion_is_set_up_properly + # Normalization of DEVANAGARI LETTER QA breaks when composition exclusion isn't used correctly + qa = [0x915, 0x93c].pack('U*') + assert_equal qa, chars(qa).normalize(:c) + end + + # Test for the Public Review Issue #29, bad explanation of composition might lead to a + # bad implementation: http://www.unicode.org/review/pr-29.html + def test_normalization_C_pri_29 + [ + [0x0B47, 0x0300, 0x0B3E], + [0x1100, 0x0300, 0x1161] + ].map { |c| c.pack('U*') }.each do |c| + assert_equal_codepoints c, chars(c).normalize(:c) + end + end + + def test_normalization_shouldnt_strip_null_bytes + null_byte_str = "Test\0test" + + assert_equal null_byte_str, chars(null_byte_str).normalize(:kc) + assert_equal null_byte_str, chars(null_byte_str).normalize(:c) + assert_equal null_byte_str, chars(null_byte_str).normalize(:d) + assert_equal null_byte_str, chars(null_byte_str).normalize(:kd) + assert_equal null_byte_str, chars(null_byte_str).decompose + assert_equal null_byte_str, chars(null_byte_str).compose + end + + def test_simple_normalization + comp_str = [ + 44, # LATIN CAPITAL LETTER D + 307, # COMBINING DOT ABOVE + 328, # COMBINING OGONEK + 323 # COMBINING DOT BELOW + ].pack("U*") + + assert_equal_codepoints '', chars('').normalize + assert_equal_codepoints [44,105,106,328,323].pack("U*"), chars(comp_str).normalize(:kc).to_s + assert_equal_codepoints [44,307,328,323].pack("U*"), chars(comp_str).normalize(:c).to_s + assert_equal_codepoints [44,307,110,780,78,769].pack("U*"), chars(comp_str).normalize(:d).to_s + assert_equal_codepoints [44,105,106,110,780,78,769].pack("U*"), chars(comp_str).normalize(:kd).to_s + end + + def test_should_compute_grapheme_length + [ + ['', 0], + ['abc', 3], + ['こにちわ', 4], + [[0x0924, 0x094D, 0x0930].pack('U*'), 2], + [%w(cr lf), 1], + [%w(l l), 1], + [%w(l v), 1], + [%w(l lv), 1], + [%w(l lvt), 1], + [%w(lv v), 1], + [%w(lv t), 1], + [%w(v v), 1], + [%w(v t), 1], + [%w(lvt t), 1], + [%w(t t), 1], + [%w(n extend), 1], + [%w(n n), 2], + [%w(n cr lf n), 3], + [%w(n l v t), 2] + ].each do |input, expected_length| + if input.kind_of?(Array) + str = string_from_classes(input) + else + str = input + end + assert_equal expected_length, chars(str).g_length + end + end + + def test_tidy_bytes_should_tidy_bytes + byte_string = "\270\236\010\210\245" + tidy_string = [0xb8, 0x17e, 0x8, 0x2c6, 0xa5].pack('U*') + ascii_padding = 'aa' + utf8_padding = 'éé' + + assert_equal_codepoints tidy_string, chars(byte_string).tidy_bytes + + assert_equal_codepoints ascii_padding.dup.insert(1, tidy_string), + chars(ascii_padding.dup.insert(1, byte_string)).tidy_bytes + assert_equal_codepoints utf8_padding.dup.insert(2, tidy_string), + chars(utf8_padding.dup.insert(2, byte_string)).tidy_bytes + assert_nothing_raised { chars(byte_string).tidy_bytes.to_s.unpack('U*') } + + assert_equal_codepoints "\xC3\xA7", chars("\xE7").tidy_bytes # iso_8859_1: small c cedilla + assert_equal_codepoints "\xE2\x80\x9C", chars("\x93").tidy_bytes # win_1252: left smart quote + assert_equal_codepoints "\xE2\x82\xAC", chars("\x80").tidy_bytes # win_1252: euro + assert_equal_codepoints "\x00", chars("\x00").tidy_bytes # null char + assert_equal_codepoints [0xfffd].pack('U'), chars("\xef\xbf\xbd").tidy_bytes # invalid char + rescue ArgumentError => e + raise e if RUBY_VERSION < '1.9' + end + + private + + def string_from_classes(classes) + # Characters from the character classes as described in UAX #29 + character_from_class = { + :l => 0x1100, :v => 0x1160, :t => 0x11A8, :lv => 0xAC00, :lvt => 0xAC01, :cr => 0x000D, :lf => 0x000A, + :extend => 0x094D, :n => 0x64 + } + classes.collect do |k| + character_from_class[k.intern] + end.pack('U*') + end end diff --git a/activesupport/test/multibyte_conformance.rb b/activesupport/test/multibyte_conformance.rb index 05fb9ef7a7..fe6cd70c70 100644 --- a/activesupport/test/multibyte_conformance.rb +++ b/activesupport/test/multibyte_conformance.rb @@ -1,13 +1,7 @@ require 'abstract_unit' +require 'fileutils' require 'open-uri' - -if RUBY_VERSION < '1.9' - -$KCODE = 'UTF8' - -UNIDATA_URL = "http://www.unicode.org/Public/#{ActiveSupport::Multibyte::UNICODE_VERSION}/ucd" -UNIDATA_FILE = '/NormalizationTest.txt' -CACHE_DIR = File.dirname(__FILE__) + '/cache' +require 'tmpdir' class Downloader def self.download(from, to) @@ -27,17 +21,19 @@ class Downloader end end -class String - # Unicode Inspect returns the codepoints of the string in hex - def ui - "#{self} " + ("[%s]" % unpack("U*").map{|cp| cp.to_s(16) }.join(' ')) - end unless ''.respond_to?(:ui) -end - -Dir.mkdir(CACHE_DIR) unless File.exist?(CACHE_DIR) -Downloader.download(UNIDATA_URL + UNIDATA_FILE, CACHE_DIR + UNIDATA_FILE) - -module ConformanceTest +class MultibyteConformanceTest < Test::Unit::TestCase + include MultibyteTest + + UNIDATA_URL = "http://www.unicode.org/Public/#{ActiveSupport::Multibyte::UNICODE_VERSION}/ucd" + UNIDATA_FILE = '/NormalizationTest.txt' + CACHE_DIR = File.join(Dir.tmpdir, 'cache') + + def setup + FileUtils.mkdir_p(CACHE_DIR) + Downloader.download(UNIDATA_URL + UNIDATA_FILE, CACHE_DIR + UNIDATA_FILE) + @proxy = ActiveSupport::Multibyte::Chars + end + def test_normalizations_C each_line_of_norm_tests do |*cols| col1, col2, col3, col4, col5, comment = *cols @@ -47,13 +43,13 @@ module ConformanceTest # # NFC # c2 == NFC(c1) == NFC(c2) == NFC(c3) - assert_equal col2.ui, @handler.normalize(col1, :c).ui, "Form C - Col 2 has to be NFC(1) - #{comment}" - assert_equal col2.ui, @handler.normalize(col2, :c).ui, "Form C - Col 2 has to be NFC(2) - #{comment}" - assert_equal col2.ui, @handler.normalize(col3, :c).ui, "Form C - Col 2 has to be NFC(3) - #{comment}" + assert_equal_codepoints col2, @proxy.new(col1).normalize(:c), "Form C - Col 2 has to be NFC(1) - #{comment}" + assert_equal_codepoints col2, @proxy.new(col2).normalize(:c), "Form C - Col 2 has to be NFC(2) - #{comment}" + assert_equal_codepoints col2, @proxy.new(col3).normalize(:c), "Form C - Col 2 has to be NFC(3) - #{comment}" # # c4 == NFC(c4) == NFC(c5) - assert_equal col4.ui, @handler.normalize(col4, :c).ui, "Form C - Col 4 has to be C(4) - #{comment}" - assert_equal col4.ui, @handler.normalize(col5, :c).ui, "Form C - Col 4 has to be C(5) - #{comment}" + assert_equal_codepoints col4, @proxy.new(col4).normalize(:c), "Form C - Col 4 has to be C(4) - #{comment}" + assert_equal_codepoints col4, @proxy.new(col5).normalize(:c), "Form C - Col 4 has to be C(5) - #{comment}" end end @@ -63,12 +59,12 @@ module ConformanceTest # # NFD # c3 == NFD(c1) == NFD(c2) == NFD(c3) - assert_equal col3.ui, @handler.normalize(col1, :d).ui, "Form D - Col 3 has to be NFD(1) - #{comment}" - assert_equal col3.ui, @handler.normalize(col2, :d).ui, "Form D - Col 3 has to be NFD(2) - #{comment}" - assert_equal col3.ui, @handler.normalize(col3, :d).ui, "Form D - Col 3 has to be NFD(3) - #{comment}" + assert_equal_codepoints col3, @proxy.new(col1).normalize(:d), "Form D - Col 3 has to be NFD(1) - #{comment}" + assert_equal_codepoints col3, @proxy.new(col2).normalize(:d), "Form D - Col 3 has to be NFD(2) - #{comment}" + assert_equal_codepoints col3, @proxy.new(col3).normalize(:d), "Form D - Col 3 has to be NFD(3) - #{comment}" # c5 == NFD(c4) == NFD(c5) - assert_equal col5.ui, @handler.normalize(col4, :d).ui, "Form D - Col 5 has to be NFD(4) - #{comment}" - assert_equal col5.ui, @handler.normalize(col5, :d).ui, "Form D - Col 5 has to be NFD(5) - #{comment}" + assert_equal_codepoints col5, @proxy.new(col4).normalize(:d), "Form D - Col 5 has to be NFD(4) - #{comment}" + assert_equal_codepoints col5, @proxy.new(col5).normalize(:d), "Form D - Col 5 has to be NFD(5) - #{comment}" end end @@ -78,11 +74,11 @@ module ConformanceTest # # NFKC # c4 == NFKC(c1) == NFKC(c2) == NFKC(c3) == NFKC(c4) == NFKC(c5) - assert_equal col4.ui, @handler.normalize(col1, :kc).ui, "Form D - Col 4 has to be NFKC(1) - #{comment}" - assert_equal col4.ui, @handler.normalize(col2, :kc).ui, "Form D - Col 4 has to be NFKC(2) - #{comment}" - assert_equal col4.ui, @handler.normalize(col3, :kc).ui, "Form D - Col 4 has to be NFKC(3) - #{comment}" - assert_equal col4.ui, @handler.normalize(col4, :kc).ui, "Form D - Col 4 has to be NFKC(4) - #{comment}" - assert_equal col4.ui, @handler.normalize(col5, :kc).ui, "Form D - Col 4 has to be NFKC(5) - #{comment}" + assert_equal_codepoints col4, @proxy.new(col1).normalize(:kc), "Form D - Col 4 has to be NFKC(1) - #{comment}" + assert_equal_codepoints col4, @proxy.new(col2).normalize(:kc), "Form D - Col 4 has to be NFKC(2) - #{comment}" + assert_equal_codepoints col4, @proxy.new(col3).normalize(:kc), "Form D - Col 4 has to be NFKC(3) - #{comment}" + assert_equal_codepoints col4, @proxy.new(col4).normalize(:kc), "Form D - Col 4 has to be NFKC(4) - #{comment}" + assert_equal_codepoints col4, @proxy.new(col5).normalize(:kc), "Form D - Col 4 has to be NFKC(5) - #{comment}" end end @@ -92,11 +88,11 @@ module ConformanceTest # # NFKD # c5 == NFKD(c1) == NFKD(c2) == NFKD(c3) == NFKD(c4) == NFKD(c5) - assert_equal col5.ui, @handler.normalize(col1, :kd).ui, "Form KD - Col 5 has to be NFKD(1) - #{comment}" - assert_equal col5.ui, @handler.normalize(col2, :kd).ui, "Form KD - Col 5 has to be NFKD(2) - #{comment}" - assert_equal col5.ui, @handler.normalize(col3, :kd).ui, "Form KD - Col 5 has to be NFKD(3) - #{comment}" - assert_equal col5.ui, @handler.normalize(col4, :kd).ui, "Form KD - Col 5 has to be NFKD(4) - #{comment}" - assert_equal col5.ui, @handler.normalize(col5, :kd).ui, "Form KD - Col 5 has to be NFKD(5) - #{comment}" + assert_equal_codepoints col5, @proxy.new(col1).normalize(:kd), "Form KD - Col 5 has to be NFKD(1) - #{comment}" + assert_equal_codepoints col5, @proxy.new(col2).normalize(:kd), "Form KD - Col 5 has to be NFKD(2) - #{comment}" + assert_equal_codepoints col5, @proxy.new(col3).normalize(:kd), "Form KD - Col 5 has to be NFKD(3) - #{comment}" + assert_equal_codepoints col5, @proxy.new(col4).normalize(:kd), "Form KD - Col 5 has to be NFKD(4) - #{comment}" + assert_equal_codepoints col5, @proxy.new(col5).normalize(:kd), "Form KD - Col 5 has to be NFKD(5) - #{comment}" end end @@ -104,7 +100,7 @@ module ConformanceTest def each_line_of_norm_tests(&block) lines = 0 max_test_lines = 0 # Don't limit below 38, because that's the header of the testfile - File.open(File.dirname(__FILE__) + '/cache' + UNIDATA_FILE, 'r') do | f | + File.open(File.join(CACHE_DIR, UNIDATA_FILE), 'r') do | f | until f.eof? || (max_test_lines > 38 and lines > max_test_lines) lines += 1 line = f.gets.chomp! @@ -122,25 +118,8 @@ module ConformanceTest end end end -end - -begin - require_library_or_gem('utf8proc_native') - require 'active_record/multibyte/handlers/utf8_handler_proc' - class ConformanceTestProc < Test::Unit::TestCase - include ConformanceTest - def setup - @handler = ::ActiveSupport::Multibyte::Handlers::UTF8HandlerProc + + def inspect_codepoints(str) + str.to_s.unpack("U*").map{|cp| cp.to_s(16) }.join(' ') end - end -rescue LoadError -end - -class ConformanceTestPure < Test::Unit::TestCase - include ConformanceTest - def setup - @handler = ::ActiveSupport::Multibyte::Handlers::UTF8Handler - end -end - -end +end \ No newline at end of file diff --git a/activesupport/test/multibyte_handler_test.rb b/activesupport/test/multibyte_handler_test.rb deleted file mode 100644 index 5575ecc32d..0000000000 --- a/activesupport/test/multibyte_handler_test.rb +++ /dev/null @@ -1,372 +0,0 @@ -# encoding: utf-8 -require 'abstract_unit' - -if RUBY_VERSION < '1.9' - -$KCODE = 'UTF8' - -class String - # Unicode Inspect returns the codepoints of the string in hex - def ui - "#{self} " + ("[%s]" % unpack("U*").map{|cp| cp.to_s(16) }.join(' ')) - end unless ''.respond_to?(:ui) -end - -module UTF8HandlingTest - - def common_setup - # This is an ASCII string with some russian strings and a ligature. It's nicely calibrated, because - # slicing it at some specific bytes will kill your characters if you use standard Ruby routines. - # It has both capital and standard letters, so that we can test case conversions easily. - # It has 26 characters and 28 when the ligature gets split during normalization. - @string = "Abcd Блå ffi бла бла бла бла" - @string_kd = "Abcd Блå ffi бла бла бла бла" - @string_kc = "Abcd Блå ffi бла бла бла бла" - @string_c = "Abcd Блå ffi бла бла бла бла" - @string_d = "Abcd Блå ffi бла бла бла бла" - @bytestring = "\270\236\010\210\245" # Not UTF-8 - - # Characters from the character classes as described in UAX #29 - @character_from_class = { - :l => 0x1100, :v => 0x1160, :t => 0x11A8, :lv => 0xAC00, :lvt => 0xAC01, :cr => 0x000D, :lf => 0x000A, - :extend => 0x094D, :n => 0x64 - } - end - - def test_utf8_recognition - assert ActiveSupport::Multibyte::Handlers::UTF8Handler.consumes?(@string), - "Should recognize as a valid UTF-8 string" - assert !ActiveSupport::Multibyte::Handlers::UTF8Handler.consumes?(@bytestring), "This is bytestring, not UTF-8" - end - - def test_simple_normalization - # Normalization of DEVANAGARI LETTER QA breaks when composition exclusion isn't used correctly - assert_equal [0x915, 0x93c].pack('U*').ui, [0x915, 0x93c].pack('U*').chars.normalize(:c).to_s.ui - - null_byte_str = "Test\0test" - - assert_equal '', @handler.normalize(''), "Empty string should not break things" - assert_equal null_byte_str.ui, @handler.normalize(null_byte_str, :kc).ui, "Null byte should remain" - assert_equal null_byte_str.ui, @handler.normalize(null_byte_str, :c).ui, "Null byte should remain" - assert_equal null_byte_str.ui, @handler.normalize(null_byte_str, :d).ui, "Null byte should remain" - assert_equal null_byte_str.ui, @handler.normalize(null_byte_str, :kd).ui, "Null byte should remain" - assert_equal null_byte_str.ui, @handler.decompose(null_byte_str).ui, "Null byte should remain" - assert_equal null_byte_str.ui, @handler.compose(null_byte_str).ui, "Null byte should remain" - - comp_str = [ - 44, # LATIN CAPITAL LETTER D - 307, # COMBINING DOT ABOVE - 328, # COMBINING OGONEK - 323 # COMBINING DOT BELOW - ].pack("U*") - norm_str_KC = [44,105,106,328,323].pack("U*") - norm_str_C = [44,307,328,323].pack("U*") - norm_str_D = [44,307,110,780,78,769].pack("U*") - norm_str_KD = [44,105,106,110,780,78,769].pack("U*") - - assert_equal norm_str_KC.ui, @handler.normalize(comp_str, :kc).ui, "Should normalize KC" - assert_equal norm_str_C.ui, @handler.normalize(comp_str, :c).ui, "Should normalize C" - assert_equal norm_str_D.ui, @handler.normalize(comp_str, :d).ui, "Should normalize D" - assert_equal norm_str_KD.ui, @handler.normalize(comp_str, :kd).ui, "Should normalize KD" - - assert_raise(ActiveSupport::Multibyte::Handlers::EncodingError) { @handler.normalize(@bytestring) } - end - - # Test for the Public Review Issue #29, bad explanation of composition might lead to a - # bad implementation: http://www.unicode.org/review/pr-29.html - def test_normalization_C_pri_29 - [ - [0x0B47, 0x0300, 0x0B3E], - [0x1100, 0x0300, 0x1161] - ].map { |c| c.pack('U*') }.each do |c| - assert_equal c.ui, @handler.normalize(c, :c).ui, "Composition is implemented incorrectly" - end - end - - def test_casefolding - simple_str = "abCdef" - simple_str_upcase = "ABCDEF" - simple_str_downcase = "abcdef" - - assert_equal '', @handler.downcase(@handler.upcase('')), "Empty string should not break things" - assert_equal simple_str_upcase, @handler.upcase(simple_str), "should upcase properly" - assert_equal simple_str_downcase, @handler.downcase(simple_str), "should downcase properly" - assert_equal simple_str_downcase, @handler.downcase(@handler.upcase(simple_str_downcase)), "upcase and downcase should be mirrors" - - rus_str = "аБвгд\0f" - rus_str_upcase = "АБВГД\0F" - rus_str_downcase = "абвгд\0f" - assert_equal rus_str_upcase, @handler.upcase(rus_str), "should upcase properly honoring null-byte" - assert_equal rus_str_downcase, @handler.downcase(rus_str), "should downcase properly honoring null-byte" - - jap_str = "の埋め込み化対応はほぼ完成" - assert_equal jap_str, @handler.upcase(jap_str), "Japanse has no upcase, should remain unchanged" - assert_equal jap_str, @handler.downcase(jap_str), "Japanse has no downcase, should remain unchanged" - - assert_raise(ActiveSupport::Multibyte::Handlers::EncodingError) { @handler.upcase(@bytestring) } - end - - def test_capitalize - { 'аБвг аБвг' => 'Абвг абвг', - 'аБвг АБВГ' => 'Абвг абвг', - 'АБВГ АБВГ' => 'Абвг абвг', - '' => '' }.each do |f,t| - assert_equal t, @handler.capitalize(f), "Capitalize should work as expected" - end - assert_raise(ActiveSupport::Multibyte::Handlers::EncodingError) { @handler.capitalize(@bytestring) } - end - - def test_translate_offset - str = "Блaå" # [2, 2, 1, 2] bytes - assert_equal 0, @handler.translate_offset('', 0), "Offset for an empty string makes no sense, return 0" - assert_equal 0, @handler.translate_offset(str, 0), "First character, first byte" - assert_equal 0, @handler.translate_offset(str, 1), "First character, second byte" - assert_equal 1, @handler.translate_offset(str, 2), "Second character, third byte" - assert_equal 1, @handler.translate_offset(str, 3), "Second character, fourth byte" - assert_equal 2, @handler.translate_offset(str, 4), "Third character, fifth byte" - assert_equal 3, @handler.translate_offset(str, 5), "Fourth character, sixth byte" - assert_equal 3, @handler.translate_offset(str, 6), "Fourth character, seventh byte" - assert_raise(ActiveSupport::Multibyte::Handlers::EncodingError) { @handler.translate_offset(@bytestring, 3) } - end - - def test_insert - assert_equal '', @handler.insert('', 0, ''), "Empty string should not break things" - assert_equal "Abcd Блå ffiБУМ бла бла бла бла", @handler.insert(@string, 10, "БУМ"), - "Text should be inserted at right codepoints" - assert_equal "Abcd Блå ffiБУМ бла бла бла бла", @string, "Insert should be destructive" - assert_raise(ActiveSupport::Multibyte::Handlers::EncodingError) do - @handler.insert(@bytestring, 2, "\210") - end - end - - def test_reverse - str = "wБлåa \n" - rev = "\n aåлБw" - assert_equal '', @handler.reverse(''), "Empty string shouldn't change" - assert_equal rev.ui, @handler.reverse(str).ui, "Should reverse properly" - assert_raise(ActiveSupport::Multibyte::Handlers::EncodingError) { @handler.reverse(@bytestring) } - end - - def test_size - assert_equal 0, @handler.size(''), "Empty string has size 0" - assert_equal 26, @handler.size(@string), "String length should be 26" - assert_equal 26, @handler.length(@string), "String length method should be properly aliased" - assert_raise(ActiveSupport::Multibyte::Handlers::EncodingError) { @handler.size(@bytestring) } - end - - def test_slice - assert_equal 0x41, @handler.slice(@string, 0), "Singular characters should return codepoints" - assert_equal 0xE5, @handler.slice(@string, 7), "Singular characters should return codepoints" - assert_equal nil, @handler.slice('', -1..1), "Broken range should return nil" - assert_equal '', @handler.slice('', 0..10), "Empty string should not break things" - assert_equal "d Блå ffi", @handler.slice(@string, 3..9), "Unicode characters have to be returned" - assert_equal "d Блå ffi", @handler.slice(@string, 3, 7), "Unicode characters have to be returned" - assert_equal "A", @handler.slice(@string, 0, 1), "Slicing from an offset should return characters" - assert_equal " Блå ffi ", @handler.slice(@string, 4..10), "Unicode characters have to be returned" - assert_equal "ffi бла", @handler.slice(@string, /ffi бла/u), "Slicing on Regexps should be supported" - assert_equal "ffi бла", @handler.slice(@string, /ffi \w\wа/u), "Slicing on Regexps should be supported" - assert_equal nil, @handler.slice(@string, /unknown/u), "Slicing on Regexps with no match should return nil" - assert_equal "ffi бла", @handler.slice(@string, /(ffi бла)/u,1), "Slicing on Regexps with a match group should be supported" - assert_equal nil, @handler.slice(@string, /(ffi)/u,2), "Slicing with a Regexp and asking for an invalid match group should return nil" - assert_equal "", @handler.slice(@string, 7..6), "Range is empty, should return an empty string" - assert_raise(ActiveSupport::Multibyte::Handlers::EncodingError) { @handler.slice(@bytestring, 2..3) } - assert_raise(TypeError, "With 2 args, should raise TypeError for non-Numeric or Regexp first argument") { @handler.slice(@string, 2..3, 1) } - assert_raise(TypeError, "With 2 args, should raise TypeError for non-Numeric or Regexp second argument") { @handler.slice(@string, 1, 2..3) } - assert_raise(ArgumentError, "Should raise ArgumentError when there are more than 2 args") { @handler.slice(@string, 1, 1, 1) } - end - - def test_grapheme_cluster_length - assert_equal 0, @handler.g_length(''), "String should count 0 grapheme clusters" - assert_equal 2, @handler.g_length([0x0924, 0x094D, 0x0930].pack('U*')), "String should count 2 grapheme clusters" - assert_equal 1, @handler.g_length(string_from_classes(%w(cr lf))), "Don't cut between CR and LF" - assert_equal 1, @handler.g_length(string_from_classes(%w(l l))), "Don't cut between L" - assert_equal 1, @handler.g_length(string_from_classes(%w(l v))), "Don't cut between L and V" - assert_equal 1, @handler.g_length(string_from_classes(%w(l lv))), "Don't cut between L and LV" - assert_equal 1, @handler.g_length(string_from_classes(%w(l lvt))), "Don't cut between L and LVT" - assert_equal 1, @handler.g_length(string_from_classes(%w(lv v))), "Don't cut between LV and V" - assert_equal 1, @handler.g_length(string_from_classes(%w(lv t))), "Don't cut between LV and T" - assert_equal 1, @handler.g_length(string_from_classes(%w(v v))), "Don't cut between V and V" - assert_equal 1, @handler.g_length(string_from_classes(%w(v t))), "Don't cut between V and T" - assert_equal 1, @handler.g_length(string_from_classes(%w(lvt t))), "Don't cut between LVT and T" - assert_equal 1, @handler.g_length(string_from_classes(%w(t t))), "Don't cut between T and T" - assert_equal 1, @handler.g_length(string_from_classes(%w(n extend))), "Don't cut before Extend" - assert_equal 2, @handler.g_length(string_from_classes(%w(n n))), "Cut between normal characters" - assert_equal 3, @handler.g_length(string_from_classes(%w(n cr lf n))), "Don't cut between CR and LF" - assert_equal 2, @handler.g_length(string_from_classes(%w(n l v t))), "Don't cut between L, V and T" - assert_raise(ActiveSupport::Multibyte::Handlers::EncodingError) { @handler.g_length(@bytestring) } - end - - def test_index - s = "Καλημέρα κόσμε!" - assert_equal 0, @handler.index('', ''), "The empty string is always found at the beginning of the string" - assert_equal 0, @handler.index('haystack', ''), "The empty string is always found at the beginning of the string" - assert_equal 0, @handler.index(s, 'Κ'), "Greek K is at 0" - assert_equal 1, @handler.index(s, 'α'), "Greek Alpha is at 1" - - assert_equal nil, @handler.index(@bytestring, 'a') - assert_raise(ActiveSupport::Multibyte::Handlers::EncodingError) { @handler.index(@bytestring, "\010") } - end - - def test_indexed_insert - s = "Καλη!" - @handler[s, 2] = "a" - assert_equal "Καaη!", s - @handler[s, 2] = "ηη" - assert_equal "Καηηη!", s - assert_raises(IndexError) { @handler[s, 10] = 'a' } - assert_equal "Καηηη!", s - @handler[s, 2] = 32 - assert_equal "Κα ηη!", s - @handler[s, 3, 2] = "λλλ" - assert_equal "Κα λλλ!", s - @handler[s, 1, 0] = "λ" - assert_equal "Κλα λλλ!", s - assert_raises(IndexError) { @handler[s, 10, 4] = 'a' } - assert_equal "Κλα λλλ!", s - @handler[s, 4..6] = "ηη" - assert_equal "Κλα ηη!", s - assert_raises(RangeError) { @handler[s, 10..12] = 'a' } - assert_equal "Κλα ηη!", s - @handler[s, /ηη/] = "λλλ" - assert_equal "Κλα λλλ!", s - assert_raises(IndexError) { @handler[s, /ii/] = 'a' } - assert_equal "Κλα λλλ!", s - @handler[s, /(λλ)(.)/, 2] = "α" - assert_equal "Κλα λλα!", s - assert_raises(IndexError) { @handler[s, /()/, 10] = 'a' } - assert_equal "Κλα λλα!", s - @handler[s, "α"] = "η" - assert_equal "Κλη λλα!", s - @handler[s, "λλ"] = "ααα" - assert_equal "Κλη αααα!", s - end - - def test_rjust - s = "Καη" - assert_raises(ArgumentError) { @handler.rjust(s, 10, '') } - assert_raises(ArgumentError) { @handler.rjust(s) } - assert_equal "Καη", @handler.rjust(s, -3) - assert_equal "Καη", @handler.rjust(s, 0) - assert_equal "Καη", @handler.rjust(s, 3) - assert_equal " Καη", @handler.rjust(s, 5) - assert_equal " Καη", @handler.rjust(s, 7) - assert_equal "----Καη", @handler.rjust(s, 7, '-') - assert_equal "ααααΚαη", @handler.rjust(s, 7, 'α') - assert_equal "abaΚαη", @handler.rjust(s, 6, 'ab') - assert_equal "αηαΚαη", @handler.rjust(s, 6, 'αη') - end - - def test_ljust - s = "Καη" - assert_raises(ArgumentError) { @handler.ljust(s, 10, '') } - assert_raises(ArgumentError) { @handler.ljust(s) } - assert_equal "Καη", @handler.ljust(s, -3) - assert_equal "Καη", @handler.ljust(s, 0) - assert_equal "Καη", @handler.ljust(s, 3) - assert_equal "Καη ", @handler.ljust(s, 5) - assert_equal "Καη ", @handler.ljust(s, 7) - assert_equal "Καη----", @handler.ljust(s, 7, '-') - assert_equal "Καηαααα", @handler.ljust(s, 7, 'α') - assert_equal "Καηaba", @handler.ljust(s, 6, 'ab') - assert_equal "Καηαηα", @handler.ljust(s, 6, 'αη') - end - - def test_center - s = "Καη" - assert_raises(ArgumentError) { @handler.center(s, 10, '') } - assert_raises(ArgumentError) { @handler.center(s) } - assert_equal "Καη", @handler.center(s, -3) - assert_equal "Καη", @handler.center(s, 0) - assert_equal "Καη", @handler.center(s, 3) - assert_equal "Καη ", @handler.center(s, 4) - assert_equal " Καη ", @handler.center(s, 5) - assert_equal " Καη ", @handler.center(s, 6) - assert_equal "--Καη--", @handler.center(s, 7, '-') - assert_equal "--Καη---", @handler.center(s, 8, '-') - assert_equal "ααΚαηαα", @handler.center(s, 7, 'α') - assert_equal "ααΚαηααα", @handler.center(s, 8, 'α') - assert_equal "aΚαηab", @handler.center(s, 6, 'ab') - assert_equal "abΚαηab", @handler.center(s, 7, 'ab') - assert_equal "ababΚαηabab", @handler.center(s, 11, 'ab') - assert_equal "αΚαηαη", @handler.center(s, 6, 'αη') - assert_equal "αηΚαηαη", @handler.center(s, 7, 'αη') - end - - def test_strip - # A unicode aware version of strip should strip all 26 types of whitespace. This includes the NO BREAK SPACE - # aka BOM (byte order mark). The byte order mark has no place in UTF-8 because it's used to detect LE and BE. - b = "\n" + [ - 32, # SPACE - 8195, # EM SPACE - 8199, # FIGURE SPACE, - 8201, # THIN SPACE - 8202, # HAIR SPACE - 65279, # NO BREAK SPACE (ZW) - ].pack('U*') - m = "word блин\n\n\n word" - e = [ - 65279, # NO BREAK SPACE (ZW) - 8201, # THIN SPACE - 8199, # FIGURE SPACE, - 32, # SPACE - ].pack('U*') - string = b+m+e - - assert_equal '', @handler.strip(''), "Empty string should stay empty" - assert_equal m+e, @handler.lstrip(string), "Whitespace should be gone on the left" - assert_equal b+m, @handler.rstrip(string), "Whitespace should be gone on the right" - assert_equal m, @handler.strip(string), "Whitespace should be stripped on both sides" - - bs = "\n #{@bytestring} \n\n" - assert_equal @bytestring, @handler.strip(bs), "Invalid unicode strings should still strip" - end - - def test_tidy_bytes - result = [0xb8, 0x17e, 0x8, 0x2c6, 0xa5].pack('U*') - assert_equal result, @handler.tidy_bytes(@bytestring) - assert_equal "a#{result}a", @handler.tidy_bytes('a' + @bytestring + 'a'), - 'tidy_bytes should leave surrounding characters intact' - assert_equal "é#{result}é", @handler.tidy_bytes('é' + @bytestring + 'é'), - 'tidy_bytes should leave surrounding characters intact' - assert_nothing_raised { @handler.tidy_bytes(@bytestring).unpack('U*') } - - assert_equal "\xC3\xA7", @handler.tidy_bytes("\xE7") # iso_8859_1: small c cedilla - assert_equal "\xC2\xA9", @handler.tidy_bytes("\xA9") # iso_8859_1: copyright symbol - assert_equal "\xE2\x80\x9C", @handler.tidy_bytes("\x93") # win_1252: left smart quote - assert_equal "\xE2\x82\xAC", @handler.tidy_bytes("\x80") # win_1252: euro - assert_equal "\x00", @handler.tidy_bytes("\x00") # null char - assert_equal [0xfffd].pack('U'), @handler.tidy_bytes("\xef\xbf\xbd") # invalid char - end - - protected - - def string_from_classes(classes) - classes.collect do |k| - @character_from_class[k.intern] - end.pack('U*') - end -end - - -begin - require_library_or_gem('utf8proc_native') - require 'active_record/multibyte/handlers/utf8_handler_proc' - class UTF8HandlingTestProc < Test::Unit::TestCase - include UTF8HandlingTest - def setup - common_setup - @handler = ::ActiveSupport::Multibyte::Handlers::UTF8HandlerProc - end - end -rescue LoadError -end - -class UTF8HandlingTestPure < Test::Unit::TestCase - include UTF8HandlingTest - def setup - common_setup - @handler = ::ActiveSupport::Multibyte::Handlers::UTF8Handler - end -end - -end diff --git a/activesupport/test/multibyte_unicode_database_test.rb b/activesupport/test/multibyte_unicode_database_test.rb new file mode 100644 index 0000000000..fb415e08d3 --- /dev/null +++ b/activesupport/test/multibyte_unicode_database_test.rb @@ -0,0 +1,28 @@ +# encoding: utf-8 + +require 'abstract_unit' + +uses_mocha "MultibyteUnicodeDatabaseTest" do + +class MultibyteUnicodeDatabaseTest < Test::Unit::TestCase + + def setup + @ucd = ActiveSupport::Multibyte::UnicodeDatabase.new + end + + ActiveSupport::Multibyte::UnicodeDatabase::ATTRIBUTES.each do |attribute| + define_method "test_lazy_loading_on_attribute_access_of_#{attribute}" do + @ucd.expects(:load) + @ucd.send(attribute) + end + end + + def test_load + @ucd.load + ActiveSupport::Multibyte::UnicodeDatabase::ATTRIBUTES.each do |attribute| + assert @ucd.send(attribute).length > 1 + end + end +end + +end \ No newline at end of file -- cgit v1.2.3