Chris@1295: if RUBY_VERSION < '1.9' Chris@1295: require 'iconv' Chris@1295: end Chris@441: Chris@441: module Redmine Chris@441: module CodesetUtil Chris@441: Chris@441: def self.replace_invalid_utf8(str) Chris@441: return str if str.nil? Chris@441: if str.respond_to?(:force_encoding) Chris@441: str.force_encoding('UTF-8') Chris@441: if ! str.valid_encoding? Chris@441: str = str.encode("US-ASCII", :invalid => :replace, Chris@441: :undef => :replace, :replace => '?').encode("UTF-8") Chris@441: end Chris@909: elsif RUBY_PLATFORM == 'java' Chris@909: begin Chris@909: ic = Iconv.new('UTF-8', 'UTF-8') Chris@909: str = ic.iconv(str) Chris@909: rescue Chris@909: str = str.gsub(%r{[^\r\n\t\x20-\x7e]}, '?') Chris@909: end Chris@441: else Chris@441: ic = Iconv.new('UTF-8', 'UTF-8') Chris@441: txtar = "" Chris@441: begin Chris@441: txtar += ic.iconv(str) Chris@441: rescue Iconv::IllegalSequence Chris@441: txtar += $!.success Chris@441: str = '?' + $!.failed[1,$!.failed.length] Chris@441: retry Chris@441: rescue Chris@441: txtar += $!.success Chris@441: end Chris@441: str = txtar Chris@441: end Chris@441: str Chris@441: end Chris@909: Chris@909: def self.to_utf8(str, encoding) Chris@909: return str if str.nil? Chris@909: str.force_encoding("ASCII-8BIT") if str.respond_to?(:force_encoding) Chris@909: if str.empty? Chris@909: str.force_encoding("UTF-8") if str.respond_to?(:force_encoding) Chris@909: return str Chris@909: end Chris@909: enc = encoding.blank? ? "UTF-8" : encoding Chris@909: if str.respond_to?(:force_encoding) Chris@909: if enc.upcase != "UTF-8" Chris@909: str.force_encoding(enc) Chris@909: str = str.encode("UTF-8", :invalid => :replace, Chris@909: :undef => :replace, :replace => '?') Chris@909: else Chris@909: str.force_encoding("UTF-8") Chris@909: if ! str.valid_encoding? Chris@909: str = str.encode("US-ASCII", :invalid => :replace, Chris@909: :undef => :replace, :replace => '?').encode("UTF-8") Chris@909: end Chris@909: end Chris@909: elsif RUBY_PLATFORM == 'java' Chris@909: begin Chris@909: ic = Iconv.new('UTF-8', enc) Chris@909: str = ic.iconv(str) Chris@909: rescue Chris@909: str = str.gsub(%r{[^\r\n\t\x20-\x7e]}, '?') Chris@909: end Chris@909: else Chris@909: ic = Iconv.new('UTF-8', enc) Chris@909: txtar = "" Chris@909: begin Chris@909: txtar += ic.iconv(str) Chris@909: rescue Iconv::IllegalSequence Chris@909: txtar += $!.success Chris@909: str = '?' + $!.failed[1,$!.failed.length] Chris@909: retry Chris@909: rescue Chris@909: txtar += $!.success Chris@909: end Chris@909: str = txtar Chris@909: end Chris@909: str Chris@909: end Chris@909: Chris@909: def self.to_utf8_by_setting(str) Chris@909: return str if str.nil? Chris@909: str = self.to_utf8_by_setting_internal(str) Chris@909: if str.respond_to?(:force_encoding) Chris@909: str.force_encoding('UTF-8') Chris@909: end Chris@909: str Chris@909: end Chris@909: Chris@909: def self.to_utf8_by_setting_internal(str) Chris@909: return str if str.nil? Chris@909: if str.respond_to?(:force_encoding) Chris@909: str.force_encoding('ASCII-8BIT') Chris@909: end Chris@909: return str if str.empty? Chris@909: return str if /\A[\r\n\t\x20-\x7e]*\Z/n.match(str) # for us-ascii Chris@909: if str.respond_to?(:force_encoding) Chris@909: str.force_encoding('UTF-8') Chris@909: end Chris@909: encodings = Setting.repositories_encodings.split(',').collect(&:strip) Chris@909: encodings.each do |encoding| Chris@1295: if str.respond_to?(:force_encoding) Chris@1295: begin Chris@1295: str.force_encoding(encoding) Chris@1295: utf8 = str.encode('UTF-8') Chris@1295: return utf8 if utf8.valid_encoding? Chris@1295: rescue Chris@1295: # do nothing here and try the next encoding Chris@1295: end Chris@1295: else Chris@1295: begin Chris@1295: return Iconv.conv('UTF-8', encoding, str) Chris@1295: rescue Iconv::Failure Chris@1295: # do nothing here and try the next encoding Chris@1295: end Chris@909: end Chris@909: end Chris@909: str = self.replace_invalid_utf8(str) Chris@909: if str.respond_to?(:force_encoding) Chris@909: str.force_encoding('UTF-8') Chris@909: end Chris@909: str Chris@909: end Chris@909: Chris@909: def self.from_utf8(str, encoding) Chris@909: str ||= '' Chris@909: if str.respond_to?(:force_encoding) Chris@909: str.force_encoding('UTF-8') Chris@909: if encoding.upcase != 'UTF-8' Chris@909: str = str.encode(encoding, :invalid => :replace, Chris@909: :undef => :replace, :replace => '?') Chris@909: else Chris@909: str = self.replace_invalid_utf8(str) Chris@909: end Chris@909: elsif RUBY_PLATFORM == 'java' Chris@909: begin Chris@909: ic = Iconv.new(encoding, 'UTF-8') Chris@909: str = ic.iconv(str) Chris@909: rescue Chris@909: str = str.gsub(%r{[^\r\n\t\x20-\x7e]}, '?') Chris@909: end Chris@909: else Chris@909: ic = Iconv.new(encoding, 'UTF-8') Chris@909: txtar = "" Chris@909: begin Chris@909: txtar += ic.iconv(str) Chris@909: rescue Iconv::IllegalSequence Chris@909: txtar += $!.success Chris@909: str = '?' + $!.failed[1, $!.failed.length] Chris@909: retry Chris@909: rescue Chris@909: txtar += $!.success Chris@909: end Chris@909: str = txtar Chris@909: end Chris@909: end Chris@441: end Chris@441: end