diff options
Diffstat (limited to 'lib/xsd')
-rw-r--r-- | lib/xsd/charset.rb | 76 | ||||
-rw-r--r-- | lib/xsd/codegen/gensupport.rb | 17 | ||||
-rw-r--r-- | lib/xsd/datatypes.rb | 23 | ||||
-rw-r--r-- | lib/xsd/ns.rb | 2 | ||||
-rw-r--r-- | lib/xsd/qname.rb | 7 |
5 files changed, 69 insertions, 56 deletions
diff --git a/lib/xsd/charset.rb b/lib/xsd/charset.rb index ccd22a7744..15d5500fce 100644 --- a/lib/xsd/charset.rb +++ b/lib/xsd/charset.rb @@ -1,5 +1,5 @@ # XSD4R - Charset handling library. -# Copyright (C) 2001, 2003 NAKAMURA, Hiroshi <nahi@ruby-lang.org>. +# Copyright (C) 2001, 2003, 2005 NAKAMURA, Hiroshi <nahi@ruby-lang.org>. # This program is copyrighted free software by NAKAMURA, Hiroshi. You can # redistribute it and/or modify it under the same terms of Ruby's license; @@ -10,7 +10,7 @@ module XSD module Charset - @encoding = $KCODE + @internal_encoding = $KCODE class XSDError < StandardError; end class CharsetError < XSDError; end @@ -24,27 +24,40 @@ public # EncodingConvertMap = {} def Charset.init + EncodingConvertMap[['UTF8', 'X_ISO8859_1']] = + Proc.new { |str| str.unpack('U*').pack('C*') } + EncodingConvertMap[['X_ISO8859_1', 'UTF8']] = + Proc.new { |str| str.unpack('C*').pack('U*') } begin require 'xsd/iconvcharset' - @encoding = 'UTF8' - sjtag = (/(mswin|bccwin|mingw|cygwin|emx)/ =~ RUBY_PLATFORM) ? 'cp932' : 'shift_jis' - EncodingConvertMap[['UTF8', 'EUC' ]] = Proc.new { |str| IconvCharset.safe_iconv("euc-jp", "utf-8", str) } - EncodingConvertMap[['EUC' , 'UTF8']] = Proc.new { |str| IconvCharset.safe_iconv("utf-8", "euc-jp", str) } - EncodingConvertMap[['EUC' , 'SJIS']] = Proc.new { |str| IconvCharset.safe_iconv(sjtag, "euc-jp", str) } - EncodingConvertMap[['UTF8', 'SJIS']] = Proc.new { |str| IconvCharset.safe_iconv(sjtag, "utf-8", str) } - EncodingConvertMap[['SJIS', 'UTF8']] = Proc.new { |str| IconvCharset.safe_iconv("utf-8", sjtag, str) } - EncodingConvertMap[['SJIS', 'EUC' ]] = Proc.new { |str| IconvCharset.safe_iconv("euc-jp", sjtag, str) } + @internal_encoding = 'UTF8' + sjtag = (/(mswin|bccwin|mingw|cygwin|emx)/ =~ RUBY_PLATFORM) ? 'cp932' : + 'shift_jis' + EncodingConvertMap[['UTF8', 'EUC' ]] = + Proc.new { |str| IconvCharset.safe_iconv("euc-jp", "utf-8", str) } + EncodingConvertMap[['EUC' , 'UTF8']] = + Proc.new { |str| IconvCharset.safe_iconv("utf-8", "euc-jp", str) } + EncodingConvertMap[['EUC' , 'SJIS']] = + Proc.new { |str| IconvCharset.safe_iconv(sjtag, "euc-jp", str) } + EncodingConvertMap[['UTF8', 'SJIS']] = + Proc.new { |str| IconvCharset.safe_iconv(sjtag, "utf-8", str) } + EncodingConvertMap[['SJIS', 'UTF8']] = + Proc.new { |str| IconvCharset.safe_iconv("utf-8", sjtag, str) } + EncodingConvertMap[['SJIS', 'EUC' ]] = + Proc.new { |str| IconvCharset.safe_iconv("euc-jp", sjtag, str) } rescue LoadError begin require 'nkf' - EncodingConvertMap[['EUC' , 'SJIS']] = Proc.new { |str| NKF.nkf('-sXm0', str) } - EncodingConvertMap[['SJIS', 'EUC' ]] = Proc.new { |str| NKF.nkf('-eXm0', str) } + EncodingConvertMap[['EUC' , 'SJIS']] = + Proc.new { |str| NKF.nkf('-sXm0', str) } + EncodingConvertMap[['SJIS', 'EUC' ]] = + Proc.new { |str| NKF.nkf('-eXm0', str) } rescue LoadError end begin require 'uconv' - @encoding = 'UTF8' + @internal_encoding = 'UTF8' EncodingConvertMap[['UTF8', 'EUC' ]] = Uconv.method(:u8toeuc) EncodingConvertMap[['UTF8', 'SJIS']] = Uconv.method(:u8tosjis) EncodingConvertMap[['EUC' , 'UTF8']] = Uconv.method(:euctou8) @@ -60,6 +73,8 @@ public 'EUC' => 'euc-jp', 'SJIS' => 'shift_jis', 'UTF8' => 'utf-8', + 'X_ISO_8859_1' => 'iso-8859-1', + 'X_UNKNOWN' => nil, } @@ -67,24 +82,24 @@ public ## handlers # def Charset.encoding - @encoding + @internal_encoding end def Charset.encoding=(encoding) warn("xsd charset is set to #{encoding}") if $DEBUG - @encoding = encoding + @internal_encoding = encoding end - def Charset.encoding_label - charset_label(@encoding) + def Charset.xml_encoding_label + charset_label(@internal_encoding) end def Charset.encoding_to_xml(str, charset) - encoding_conv(str, @encoding, charset_str(charset)) + encoding_conv(str, @internal_encoding, charset_str(charset)) end def Charset.encoding_from_xml(str, charset) - encoding_conv(str, charset_str(charset), @encoding) + encoding_conv(str, charset_str(charset), @internal_encoding) end def Charset.encoding_conv(str, enc_from, enc_to) @@ -94,7 +109,7 @@ public converter.call(str) else raise CharsetConversionError.new( - "Converter not found: #{ enc_from } -> #{ enc_to }") + "Converter not found: #{enc_from} -> #{enc_to}") end end @@ -104,26 +119,26 @@ public def Charset.charset_str(label) if CharsetMap.respond_to?(:key) - CharsetMap.key(label.downcase) + CharsetMap.key(label.downcase) || 'X_UNKNOWN' else - CharsetMap.index(label.downcase) + CharsetMap.index(label.downcase) || 'X_UNKNOWN' end end # us_ascii = '[\x00-\x7F]' us_ascii = '[\x9\xa\xd\x20-\x7F]' # XML 1.0 restricted. - USASCIIRegexp = Regexp.new("\\A#{ us_ascii }*\\z", nil, "NONE") + USASCIIRegexp = Regexp.new("\\A#{us_ascii}*\\z", nil, "NONE") twobytes_euc = '(?:[\x8E\xA1-\xFE][\xA1-\xFE])' threebytes_euc = '(?:\x8F[\xA1-\xFE][\xA1-\xFE])' - character_euc = "(?:#{ us_ascii }|#{ twobytes_euc }|#{ threebytes_euc })" - EUCRegexp = Regexp.new("\\A#{ character_euc }*\\z", nil, "NONE") + character_euc = "(?:#{us_ascii}|#{twobytes_euc}|#{threebytes_euc})" + EUCRegexp = Regexp.new("\\A#{character_euc}*\\z", nil, "NONE") # onebyte_sjis = '[\x00-\x7F\xA1-\xDF]' onebyte_sjis = '[\x9\xa\xd\x20-\x7F\xA1-\xDF]' # XML 1.0 restricted. twobytes_sjis = '(?:[\x81-\x9F\xE0-\xFC][\x40-\x7E\x80-\xFC])' - character_sjis = "(?:#{ onebyte_sjis }|#{ twobytes_sjis })" - SJISRegexp = Regexp.new("\\A#{ character_sjis }*\\z", nil, "NONE") + character_sjis = "(?:#{onebyte_sjis}|#{twobytes_sjis})" + SJISRegexp = Regexp.new("\\A#{character_sjis}*\\z", nil, "NONE") # 0xxxxxxx # 110yyyyy 10xxxxxx @@ -132,8 +147,9 @@ public threebytes_utf8 = '(?:[\xE0-\xEF][\x80-\xBF][\x80-\xBF])' # 11110uuu 10uuuzzz 10yyyyyy 10xxxxxx fourbytes_utf8 = '(?:[\xF0-\xF7][\x80-\xBF][\x80-\xBF][\x80-\xBF])' - character_utf8 = "(?:#{ us_ascii }|#{ twobytes_utf8 }|#{ threebytes_utf8 }|#{ fourbytes_utf8 })" - UTF8Regexp = Regexp.new("\\A#{ character_utf8 }*\\z", nil, "NONE") + character_utf8 = + "(?:#{us_ascii}|#{twobytes_utf8}|#{threebytes_utf8}|#{fourbytes_utf8})" + UTF8Regexp = Regexp.new("\\A#{character_utf8}*\\z", nil, "NONE") def Charset.is_us_ascii(str) USASCIIRegexp =~ str @@ -162,7 +178,7 @@ public when 'SJIS' is_sjis(str) else - raise UnknownCharsetError.new("Unknown charset: #{ code }") + raise UnknownCharsetError.new("Unknown charset: #{code}") end end end diff --git a/lib/xsd/codegen/gensupport.rb b/lib/xsd/codegen/gensupport.rb index 854534154a..1e85d3668f 100644 --- a/lib/xsd/codegen/gensupport.rb +++ b/lib/xsd/codegen/gensupport.rb @@ -1,5 +1,5 @@ # XSD4R - Code generation support -# Copyright (C) 2004 NAKAMURA, Hiroshi <nahi@ruby-lang.org>. +# Copyright (C) 2004, 2005 NAKAMURA, Hiroshi <nahi@ruby-lang.org>. # This program is copyrighted free software by NAKAMURA, Hiroshi. You can # redistribute it and/or modify it under the same terms of Ruby's license; @@ -10,7 +10,8 @@ module XSD module CodeGen # from the file 'keywords' in 1.9. -KEYWORD = %w( +KEYWORD = {} +%w( __LINE__ __FILE__ BEGIN @@ -51,7 +52,7 @@ until when while yield -) +).each { |k| KEYWORD[k] = nil } module GenSupport def capitalize(target) @@ -96,12 +97,12 @@ module GenSupport module_function :safemethodname? def safevarname(name) - safename = name.scan(/[a-zA-Z0-9_]+/).join('_') - safename = uncapitalize(safename) + safename = uncapitalize(name.scan(/[a-zA-Z0-9_]+/).join('_')) if /^[a-z]/ !~ safename or keyword?(safename) - safename = "v_#{safename}" + "v_#{safename}" + else + safename end - safename end module_function :safevarname @@ -111,7 +112,7 @@ module GenSupport module_function :safevarname? def keyword?(word) - KEYWORD.include?(word) + KEYWORD.key?(word) end module_function :keyword? diff --git a/lib/xsd/datatypes.rb b/lib/xsd/datatypes.rb index d0fc44d69b..bbe6c8578f 100644 --- a/lib/xsd/datatypes.rb +++ b/lib/xsd/datatypes.rb @@ -560,23 +560,20 @@ module XSDDateTimeImpl def screen_data(t) # convert t to a DateTime as an internal representation. - if t.is_a?(DateTime) + if t.respond_to?(:to_datetime) # 1.9 or later + t.to_datetime + elsif t.is_a?(DateTime) t elsif t.is_a?(Date) - if t.respond_to?(:to_datetime) # from 1.9 - t.to_datetime - else - t = screen_data_str(t) - t <<= 12 if t.year < 0 - t - end + t = screen_data_str(t) + t <<= 12 if t.year < 0 + t elsif t.is_a?(Time) - sec, min, hour, mday, month, year = t.to_a[0..5] - diffday = t.usec.to_r / 1000000 / SecInDay + jd = DateTime.civil_to_jd(t.year, t.mon, t.mday, DateTime::ITALY) + fr = DateTime.time_to_day_fraction(t.hour, t.min, [t.sec, 59].min) + + t.usec.to_r / 1000000 / SecInDay of = t.utc_offset.to_r / SecInDay - data = DateTime.civil(year, month, mday, hour, min, sec, of) - data += diffday - data + DateTime.new0(DateTime.jd_to_ajd(jd, fr, of), of, DateTime::ITALY) else screen_data_str(t) end diff --git a/lib/xsd/ns.rb b/lib/xsd/ns.rb index b63a8a5cdd..53eeae7130 100644 --- a/lib/xsd/ns.rb +++ b/lib/xsd/ns.rb @@ -72,7 +72,7 @@ public if (name.namespace == @default_namespace) name.name elsif @ns2tag.key?(name.namespace) - @ns2tag[name.namespace] + ':' << name.name + "#{@ns2tag[name.namespace]}:#{name.name}" else raise FormatError.new("namespace: #{name.namespace} not defined yet") end diff --git a/lib/xsd/qname.rb b/lib/xsd/qname.rb index 18b002764b..bb9763a69a 100644 --- a/lib/xsd/qname.rb +++ b/lib/xsd/qname.rb @@ -31,9 +31,6 @@ class QName end def match(rhs) - unless self.class === rhs - return false - end if rhs.namespace and (rhs.namespace != @namespace) return false end @@ -44,7 +41,7 @@ class QName end def ==(rhs) - (self.class === rhs && @namespace == rhs.namespace && @name == rhs.name) + !rhs.nil? and @namespace == rhs.namespace and @name == rhs.name end def ===(rhs) @@ -73,6 +70,8 @@ class QName NormalizedNameRegexp =~ str self.new($1, $2) end + + EMPTY = QName.new.freeze end |