summaryrefslogtreecommitdiff
path: root/ruby_1_8_5/lib/rexml/encoding.rb
diff options
context:
space:
mode:
Diffstat (limited to 'ruby_1_8_5/lib/rexml/encoding.rb')
-rw-r--r--ruby_1_8_5/lib/rexml/encoding.rb64
1 files changed, 64 insertions, 0 deletions
diff --git a/ruby_1_8_5/lib/rexml/encoding.rb b/ruby_1_8_5/lib/rexml/encoding.rb
new file mode 100644
index 0000000000..644957439e
--- /dev/null
+++ b/ruby_1_8_5/lib/rexml/encoding.rb
@@ -0,0 +1,64 @@
+# -*- mode: ruby; ruby-indent-level: 2; indent-tabs-mode: t; tab-width: 2 -*- vim: sw=2 ts=2
+module REXML
+ module Encoding
+ @encoding_methods = {}
+ def self.register(enc, &block)
+ @encoding_methods[enc] = block
+ end
+ def self.apply(obj, enc)
+ @encoding_methods[enc][obj]
+ end
+ def self.encoding_method(enc)
+ @encoding_methods[enc]
+ end
+
+ # Native, default format is UTF-8, so it is declared here rather than in
+ # an encodings/ definition.
+ UTF_8 = 'UTF-8'
+ UTF_16 = 'UTF-16'
+ UNILE = 'UNILE'
+
+ # ID ---> Encoding name
+ attr_reader :encoding
+ def encoding=( enc )
+ old_verbosity = $VERBOSE
+ begin
+ $VERBOSE = false
+ return if defined? @encoding and enc == @encoding
+ if enc and enc != UTF_8
+ @encoding = enc.upcase
+ begin
+ require 'rexml/encodings/ICONV.rb'
+ Encoding.apply(self, "ICONV")
+ rescue LoadError, Exception => err
+ raise ArgumentError, "Bad encoding name #@encoding" unless @encoding =~ /^[\w-]+$/
+ @encoding.untaint
+ enc_file = File.join( "rexml", "encodings", "#@encoding.rb" )
+ begin
+ require enc_file
+ Encoding.apply(self, @encoding)
+ rescue LoadError
+ puts $!.message
+ raise ArgumentError, "No decoder found for encoding #@encoding. Please install iconv."
+ end
+ end
+ else
+ @encoding = UTF_8
+ require 'rexml/encodings/UTF-8.rb'
+ Encoding.apply(self, @encoding)
+ end
+ ensure
+ $VERBOSE = old_verbosity
+ end
+ end
+
+ def check_encoding str
+ # We have to recognize UTF-16, LSB UTF-16, and UTF-8
+ return UTF_16 if str[0] == 254 && str[1] == 255
+ return UNILE if str[0] == 255 && str[1] == 254
+ str =~ /^\s*<?xml\s*version=(['"]).*?\2\s*encoding=(["'])(.*?)\2/um
+ return $1.upcase if $1
+ return UTF_8
+ end
+ end
+end