summaryrefslogtreecommitdiff
path: root/lib/xsd
diff options
context:
space:
mode:
authornahi <nahi@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>2003-09-24 15:18:44 +0000
committernahi <nahi@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>2003-09-24 15:18:44 +0000
commitdb9445103c082a306ba085f7677da02ea94b8841 (patch)
treea311d59f031ae5def87f68be71ed1f58abadc031 /lib/xsd
parent8c2fb77787d1f20b4c19c9c52552856c339b86e9 (diff)
* lib/soap/* (29 files): SOAP4R added.
* lib/wsdl/* (42 files): WSDL4R added. * lib/xsd/* (12 files): XSD4R added. * test/soap/* (16 files): added. * test/wsdl/* (2 files): added. * test/xsd/* (3 files): added. * sample/soap/* (27 files): added. * sample/wsdl/* (13 files): added. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@4591 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Diffstat (limited to 'lib/xsd')
-rw-r--r--lib/xsd/charset.rb175
-rw-r--r--lib/xsd/datatypes.rb1094
-rw-r--r--lib/xsd/datatypes1999.rb31
-rw-r--r--lib/xsd/iconvcharset.rb44
-rw-r--r--lib/xsd/namedelements.rb86
-rw-r--r--lib/xsd/ns.rb141
-rw-r--r--lib/xsd/qname.rb77
-rw-r--r--lib/xsd/xmlparser.rb72
-rw-r--r--lib/xsd/xmlparser/parser.rb107
-rw-r--r--lib/xsd/xmlparser/rexmlparser.rb65
-rw-r--r--lib/xsd/xmlparser/xmlparser.rb61
-rw-r--r--lib/xsd/xmlparser/xmlscanner.rb158
12 files changed, 2111 insertions, 0 deletions
diff --git a/lib/xsd/charset.rb b/lib/xsd/charset.rb
new file mode 100644
index 0000000000..6dda959155
--- /dev/null
+++ b/lib/xsd/charset.rb
@@ -0,0 +1,175 @@
+=begin
+XSD4R - Charset handling library.
+Copyright (C) 2001, 2003 NAKAMURA, Hiroshi.
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 2 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PRATICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+this program; if not, write to the Free Software Foundation, Inc., 675 Mass
+Ave, Cambridge, MA 02139, USA.
+=end
+
+
+module XSD
+
+
+module Charset
+ @encoding = $KCODE
+
+ class XSDError < StandardError; end
+ class CharsetError < XSDError; end
+ class UnknownCharsetError < CharsetError; end
+ class CharsetConversionError < CharsetError; end
+
+public
+
+ ###
+ ## Maps
+ #
+ EncodingConvertMap = {}
+ def Charset.init
+ begin
+ require 'xsd/iconvcharset'
+ @encoding = 'UTF8'
+ 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("shift-jis", "euc-jp", str) }
+ if /(mswin|bccwin|mingw|cygwin)/ =~ RUBY_PLATFORM
+ EncodingConvertMap[['UTF8', 'SJIS']] = Proc.new { |str| IconvCharset.safe_iconv("cp932", "utf-8", str) }
+ EncodingConvertMap[['SJIS', 'UTF8']] = Proc.new { |str| IconvCharset.safe_iconv("utf-8", "cp932", str) }
+ EncodingConvertMap[['SJIS', 'EUC' ]] = Proc.new { |str| IconvCharset.safe_iconv("euc-jp", "cp932", str) }
+ else
+ EncodingConvertMap[['UTF8', 'SJIS']] = Proc.new { |str| IconvCharset.safe_iconv("shift-jis", "utf-8", str) }
+ EncodingConvertMap[['SJIS', 'UTF8']] = Proc.new { |str| IconvCharset.safe_iconv("utf-8", "shift-jis", str) }
+ EncodingConvertMap[['SJIS', 'EUC' ]] = Proc.new { |str| IconvCharset.safe_iconv("euc-jp", "shift-jis", str) }
+ end
+ 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) }
+ rescue LoadError
+ end
+
+ begin
+ require 'uconv'
+ @encoding = 'UTF8'
+ EncodingConvertMap[['UTF8', 'EUC' ]] = Uconv.method(:u8toeuc)
+ EncodingConvertMap[['UTF8', 'SJIS']] = Uconv.method(:u8tosjis)
+ EncodingConvertMap[['EUC' , 'UTF8']] = Uconv.method(:euctou8)
+ EncodingConvertMap[['SJIS', 'UTF8']] = Uconv.method(:sjistou8)
+ rescue LoadError
+ end
+ end
+ end
+ self.init
+
+ CharsetMap = {
+ 'NONE' => 'us-ascii',
+ 'EUC' => 'euc-jp',
+ 'SJIS' => 'shift_jis',
+ 'UTF8' => 'utf-8',
+ }
+
+
+ ###
+ ## handlers
+ #
+ def Charset.encoding
+ @encoding
+ end
+
+ def Charset.encoding_label
+ charset_label(@encoding)
+ end
+
+ def Charset.encoding_to_xml(str, charset)
+ encoding_conv(str, @encoding, charset_str(charset))
+ end
+
+ def Charset.encoding_from_xml(str, charset)
+ encoding_conv(str, charset_str(charset), @encoding)
+ end
+
+ def Charset.encoding_conv(str, enc_from, enc_to)
+ if enc_from == enc_to or enc_from == 'NONE' or enc_to == 'NONE'
+ str
+ elsif converter = EncodingConvertMap[[enc_from, enc_to]]
+ converter.call(str)
+ else
+ raise CharsetConversionError.new(
+ "Converter not found: #{ enc_from } -> #{ enc_to }")
+ end
+ end
+
+ def Charset.charset_label(encoding)
+ CharsetMap[encoding.upcase]
+ end
+
+ def Charset.charset_str(label)
+ CharsetMap.index(label.downcase)
+ end
+
+ # Original regexps: http://www.din.or.jp/~ohzaki/perl.htm
+ # ascii_euc = '[\x00-\x7F]'
+ ascii_euc = '[\x9\xa\xd\x20-\x7F]' # XML 1.0 restricted.
+ twobytes_euc = '(?:[\x8E\xA1-\xFE][\xA1-\xFE])'
+ threebytes_euc = '(?:\x8F[\xA1-\xFE][\xA1-\xFE])'
+ character_euc = "(?:#{ ascii_euc }|#{ 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")
+
+ # 0xxxxxxx
+ #ascii_utf8 = '[\0-\x7F]'
+ ascii_utf8 = '[\x9\xA\xD\x20-\x7F]' # XML 1.0 restricted.
+ # 110yyyyy 10xxxxxx
+ twobytes_utf8 = '(?:[\xC0-\xDF][\x80-\xBF])'
+ # 1110zzzz 10yyyyyy 10xxxxxx
+ threebytes_utf8 = '(?:[\xE0-\xEF][\x80-\xBF][\x80-\xBF])'
+ # 11110uuu 10uuuzzz 10yyyyyy 10xxxxxx
+ fourbytes_utf8 = '(?:[\xF0-\xF7][\x80-\xBF][\x80-\xBF][\x80-\xBF])'
+ character_utf8 = "(?:#{ ascii_utf8 }|#{ twobytes_utf8 }|#{ threebytes_utf8 }|#{ fourbytes_utf8 })"
+ UTF8Regexp = Regexp.new("\\A#{ character_utf8 }*\\z", nil, "NONE")
+
+ def Charset.is_utf8(str)
+ UTF8Regexp =~ str
+ end
+
+ def Charset.is_euc(str)
+ EUCRegexp =~ str
+ end
+
+ def Charset.is_sjis(str)
+ SJISRegexp =~ str
+ end
+
+ def Charset.is_ces(str, code = $KCODE)
+ case code
+ when 'NONE'
+ true
+ when 'UTF8'
+ is_utf8(str)
+ when 'EUC'
+ is_euc(str)
+ when 'SJIS'
+ is_sjis(str)
+ else
+ raise UnknownCharsetError.new("Unknown charset: #{ code }")
+ end
+ end
+end
+
+
+end
diff --git a/lib/xsd/datatypes.rb b/lib/xsd/datatypes.rb
new file mode 100644
index 0000000000..03e42d21d2
--- /dev/null
+++ b/lib/xsd/datatypes.rb
@@ -0,0 +1,1094 @@
+=begin
+XSD4R - XML Schema Datatype implementation.
+Copyright (C) 2000, 2001, 2002, 2003 NAKAMURA, Hiroshi.
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 2 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PRATICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+this program; if not, write to the Free Software Foundation, Inc., 675 Mass
+Ave, Cambridge, MA 02139, USA.
+=end
+
+
+require 'xsd/qname'
+require 'xsd/charset'
+require 'uri'
+
+
+###
+## XMLSchamaDatatypes general definitions.
+#
+module XSD
+
+
+Namespace = 'http://www.w3.org/2001/XMLSchema'
+InstanceNamespace = 'http://www.w3.org/2001/XMLSchema-instance'
+
+AttrType = 'type'
+NilValue = 'true'
+
+AnyTypeLiteral = 'anyType'
+AnySimpleTypeLiteral = 'anySimpleType'
+NilLiteral = 'nil'
+StringLiteral = 'string'
+BooleanLiteral = 'boolean'
+DecimalLiteral = 'decimal'
+FloatLiteral = 'float'
+DoubleLiteral = 'double'
+DurationLiteral = 'duration'
+DateTimeLiteral = 'dateTime'
+TimeLiteral = 'time'
+DateLiteral = 'date'
+GYearMonthLiteral = 'gYearMonth'
+GYearLiteral = 'gYear'
+GMonthDayLiteral = 'gMonthDay'
+GDayLiteral = 'gDay'
+GMonthLiteral = 'gMonth'
+HexBinaryLiteral = 'hexBinary'
+Base64BinaryLiteral = 'base64Binary'
+AnyURILiteral = 'anyURI'
+QNameLiteral = 'QName'
+
+NormalizedStringLiteral = 'normalizedString'
+IntegerLiteral = 'integer'
+LongLiteral = 'long'
+IntLiteral = 'int'
+ShortLiteral = 'short'
+
+AttrTypeName = QName.new(InstanceNamespace, AttrType)
+AttrNilName = QName.new(InstanceNamespace, NilLiteral)
+
+AnyTypeName = QName.new(Namespace, AnyTypeLiteral)
+AnySimpleTypeName = QName.new(Namespace, AnySimpleTypeLiteral)
+
+class Error < StandardError; end
+class ValueSpaceError < Error; end
+
+
+###
+## The base class of all datatypes with Namespace.
+#
+class NSDBase
+ @@types = []
+
+ attr_accessor :type
+
+ def self.inherited(klass)
+ @@types << klass
+ end
+
+ def self.types
+ @@types
+ end
+
+ def initialize
+ @type = nil
+ end
+end
+
+
+###
+## The base class of XSD datatypes.
+#
+class XSDAnySimpleType < NSDBase
+ include XSD
+ Type = QName.new(Namespace, AnySimpleTypeLiteral)
+
+ # @data represents canonical space (ex. Integer: 123).
+ attr_reader :data
+ # @is_nil represents this data is nil or not.
+ attr_accessor :is_nil
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ @data = nil
+ @is_nil = true
+ set(value) if value
+ end
+
+ # set accepts a string which follows lexical space (ex. String: "+123"), or
+ # an object which follows canonical space (ex. Integer: 123).
+ def set(value)
+ if value.nil?
+ @is_nil = true
+ @data = nil
+ else
+ @is_nil = false
+ _set(value)
+ end
+ end
+
+ # to_s creates a string which follows lexical space (ex. String: "123").
+ def to_s()
+ if @is_nil
+ ""
+ else
+ _to_s
+ end
+ end
+
+ def trim(data)
+ data.sub(/\A\s*(\S*)\s*\z/, '\1')
+ end
+
+private
+
+ def _set(value)
+ @data = value
+ end
+
+ def _to_s
+ @data.to_s
+ end
+end
+
+class XSDNil < XSDAnySimpleType
+ Type = QName.new(Namespace, NilLiteral)
+ Value = 'true'
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value)
+ end
+
+private
+
+ def _set(value)
+ @data = value
+ end
+end
+
+
+###
+## Primitive datatypes.
+#
+class XSDString < XSDAnySimpleType
+ Type = QName.new(Namespace, StringLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ @encoding = nil
+ set(value) if value
+ end
+
+private
+
+ def _set(value)
+ unless XSD::Charset.is_ces(value, XSD::Charset.encoding)
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ value }'.")
+ end
+ @data = value
+ end
+end
+
+class XSDBoolean < XSDAnySimpleType
+ Type = QName.new(Namespace, BooleanLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value)
+ end
+
+private
+
+ def _set(value)
+ if value.is_a?(String)
+ str = trim(value)
+ if str == 'true' || str == '1'
+ @data = true
+ elsif str == 'false' || str == '0'
+ @data = false
+ else
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+ else
+ @data = value ? true : false
+ end
+ end
+end
+
+class XSDDecimal < XSDAnySimpleType
+ Type = QName.new(Namespace, DecimalLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ @sign = ''
+ @number = ''
+ @point = 0
+ set(value) if value
+ end
+
+ def nonzero?
+ (@number != '0')
+ end
+
+private
+
+ def _set(d)
+ if d.is_a?(String)
+ # Integer("00012") => 10 in Ruby.
+ d.sub!(/^([+\-]?)0*(?=\d)/, "\\1")
+ end
+ set_str(d)
+ end
+
+ def set_str(str)
+ /^([+\-]?)(\d*)(?:\.(\d*)?)?$/ =~ trim(str.to_s)
+ unless Regexp.last_match
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+
+ @sign = $1 || '+'
+ int_part = $2
+ frac_part = $3
+
+ int_part = '0' if int_part.empty?
+ frac_part = frac_part ? frac_part.sub(/0+$/, '') : ''
+ @point = - frac_part.size
+ @number = int_part + frac_part
+
+ # normalize
+ if @sign == '+'
+ @sign = ''
+ elsif @sign == '-'
+ if @number == '0'
+ @sign = ''
+ end
+ end
+
+ @data = _to_s
+ end
+
+ # 0.0 -> 0; right?
+ def _to_s
+ str = @number.dup
+ if @point.nonzero?
+ str[@number.size + @point, 0] = '.'
+ end
+ @sign + str
+ end
+end
+
+class XSDFloat < XSDAnySimpleType
+ Type = QName.new(Namespace, FloatLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def _set(value)
+ # "NaN".to_f => 0 in some environment. libc?
+ if value.is_a?(Float)
+ @data = narrow32bit(value)
+ return
+ end
+
+ str = trim(value.to_s)
+ if str == 'NaN'
+ @data = 0.0/0.0
+ elsif str == 'INF'
+ @data = 1.0/0.0
+ elsif str == '-INF'
+ @data = -1.0/0.0
+ else
+ if /^[+\-\.\deE]+$/ !~ str
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+ # Float("-1.4E") might fail on some system.
+ str << '0' if /e$/i =~ str
+ begin
+ @data = narrow32bit(Float(str))
+ rescue ArgumentError
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+ end
+ end
+
+ # Do I have to convert 0.0 -> 0 and -0.0 -> -0 ?
+ def _to_s
+ if @data.nan?
+ 'NaN'
+ elsif @data.infinite? == 1
+ 'INF'
+ elsif @data.infinite? == -1
+ '-INF'
+ else
+ sprintf("%.10g", @data)
+ end
+ end
+
+ # Convert to single-precision 32-bit floating point value.
+ def narrow32bit(f)
+ if f.nan? || f.infinite?
+ f
+ else
+ packed = [f].pack("f")
+ (/\A\0*\z/ =~ packed)? 0.0 : f
+ end
+ end
+end
+
+# Ruby's Float is double-precision 64-bit floating point value.
+class XSDDouble < XSDAnySimpleType
+ Type = QName.new(Namespace, DoubleLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def _set(value)
+ # "NaN".to_f => 0 in some environment. libc?
+ if value.is_a?(Float)
+ @data = value
+ return
+ end
+
+ str = trim(value.to_s)
+ if str == 'NaN'
+ @data = 0.0/0.0
+ elsif str == 'INF'
+ @data = 1.0/0.0
+ elsif str == '-INF'
+ @data = -1.0/0.0
+ else
+ begin
+ @data = Float(str)
+ rescue ArgumentError
+ # '1.4e' cannot be parsed on some architecture.
+ if /e\z/i =~ str
+ begin
+ @data = Float(str + '0')
+ rescue ArgumentError
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+ else
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+ end
+ end
+ end
+
+ # Do I have to convert 0.0 -> 0 and -0.0 -> -0 ?
+ def _to_s
+ if @data.nan?
+ 'NaN'
+ elsif @data.infinite? == 1
+ 'INF'
+ elsif @data.infinite? == -1
+ '-INF'
+ else
+ sprintf("%.16g", @data)
+ end
+ end
+end
+
+class XSDDuration < XSDAnySimpleType
+ Type = QName.new(Namespace, DurationLiteral)
+
+ attr_accessor :sign
+ attr_accessor :year
+ attr_accessor :month
+ attr_accessor :day
+ attr_accessor :hour
+ attr_accessor :min
+ attr_accessor :sec
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ @sign = nil
+ @year = nil
+ @month = nil
+ @day = nil
+ @hour = nil
+ @min = nil
+ @sec = nil
+ set(value) if value
+ end
+
+private
+
+ def _set(value)
+ /^([+\-]?)P(?:(\d+)Y)?(?:(\d+)M)?(?:(\d+)D)?(T(?:(\d+)H)?(?:(\d+)M)?(?:(\d+(?:\.\d+)?)S)?)?$/ =~ trim(value.to_s)
+ unless Regexp.last_match
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ value }'.")
+ end
+
+ if ($5 and ((!$2 and !$3 and !$4) or (!$6 and !$7 and !$8)))
+ # Should we allow 'PT5S' here?
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ value }'.")
+ end
+
+ @sign = $1
+ @year = $2.to_i
+ @month = $3.to_i
+ @day = $4.to_i
+ @hour = $6.to_i
+ @min = $7.to_i
+ @sec = $8 ? XSDDecimal.new($8) : 0
+ @data = _to_s
+ end
+
+ def _to_s
+ str = ''
+ str << @sign if @sign
+ str << 'P'
+ l = ''
+ l << "#{ @year }Y" if @year.nonzero?
+ l << "#{ @month }M" if @month.nonzero?
+ l << "#{ @day }D" if @day.nonzero?
+ r = ''
+ r << "#{ @hour }H" if @hour.nonzero?
+ r << "#{ @min }M" if @min.nonzero?
+ r << "#{ @sec }S" if @sec.nonzero?
+ str << l
+ if l.empty?
+ str << "0D"
+ end
+ unless r.empty?
+ str << "T" << r
+ end
+ str
+ end
+end
+
+
+require 'rational'
+require 'date'
+unless Object.const_defined?('DateTime')
+ raise LoadError.new('XSD4R requires date2/3.2 or later to be installed. You can download it from http://www.funaba.org/en/ruby.html#date2')
+end
+
+module XSDDateTimeImpl
+ SecInDay = 86400 # 24 * 60 * 60
+
+ def to_time
+ begin
+ if @data.of * SecInDay == Time.now.utc_offset
+ d = @data
+ usec = (d.sec_fraction * SecInDay * 1000000).to_f
+ Time.local(d.year, d.month, d.mday, d.hour, d.min, d.sec, usec)
+ else
+ d = @data.newof
+ usec = (d.sec_fraction * SecInDay * 1000000).to_f
+ Time.gm(d.year, d.month, d.mday, d.hour, d.min, d.sec, usec)
+ end
+ rescue ArgumentError
+ nil
+ end
+ end
+
+ def tz2of(str)
+ /^(?:Z|(?:([+\-])(\d\d):(\d\d))?)$/ =~ str
+ sign = $1
+ hour = $2.to_i
+ min = $3.to_i
+
+ of = case sign
+ when '+'
+ of = +(hour.to_r * 60 + min) / 1440 # 24 * 60
+ when '-'
+ of = -(hour.to_r * 60 + min) / 1440 # 24 * 60
+ else
+ 0
+ end
+ of
+ end
+
+ def of2tz(offset)
+ diffmin = offset * 24 * 60
+ if diffmin.zero?
+ 'Z'
+ else
+ ((diffmin < 0) ? '-' : '+') << format('%02d:%02d',
+ (diffmin.abs / 60.0).to_i, (diffmin.abs % 60.0).to_i)
+ end
+ end
+
+ def _set(t)
+ if (t.is_a?(Date))
+ @data = t
+ elsif (t.is_a?(Time))
+ sec, min, hour, mday, month, year = t.to_a[0..5]
+ diffday = t.usec.to_r / 1000000 / SecInDay
+ of = t.utc_offset.to_r / SecInDay
+ @data = DateTime.civil(year, month, mday, hour, min, sec, of)
+ @data += diffday
+ else
+ set_str(t)
+ end
+ end
+
+ def add_tz(s)
+ s + of2tz(@data.offset)
+ end
+end
+
+class XSDDateTime < XSDAnySimpleType
+ include XSDDateTimeImpl
+ Type = QName.new(Namespace, DateTimeLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def set_str(t)
+ /^([+\-]?\d\d\d\d\d*)-(\d\d)-(\d\d)T(\d\d):(\d\d):(\d\d(?:\.(\d*))?)(Z|(?:[+\-]\d\d:\d\d)?)?$/ =~ trim(t.to_s)
+ unless Regexp.last_match
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ t }'.")
+ end
+ if $1 == '0000'
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ t }'.")
+ end
+
+ year = $1.to_i
+ if year < 0
+ year += 1
+ end
+ mon = $2.to_i
+ mday = $3.to_i
+ hour = $4.to_i
+ min = $5.to_i
+ sec = $6.to_i
+ secfrac = $7
+ zonestr = $8
+
+ @data = DateTime.civil(year, mon, mday, hour, min, sec, tz2of(zonestr))
+
+ if secfrac
+ diffday = secfrac.to_i.to_r / (10 ** secfrac.size) / SecInDay
+ # jd = @data.jd
+ # day_fraction = @data.day_fraction + diffday
+ # @data = DateTime.new0(DateTime.jd_to_rjd(jd, day_fraction,
+ # @data.offset), @data.offset)
+ #
+ # Thanks to Funaba-san, above code can be simply written as below.
+ @data += diffday
+ # FYI: new0 and jd_to_rjd are not necessary to use if you don't have
+ # exceptional reason.
+ end
+ end
+
+ def _to_s
+ year = (@data.year > 0) ? @data.year : @data.year - 1
+ s = format('%.4d-%02d-%02dT%02d:%02d:%02d',
+ year, @data.mon, @data.mday, @data.hour, @data.min, @data.sec)
+ if @data.sec_fraction.nonzero?
+ fr = @data.sec_fraction * SecInDay
+ shiftsize = fr.denominator.to_s.size
+ fr_s = (fr * (10 ** shiftsize)).to_i.to_s
+ s << '.' << '0' * (shiftsize - fr_s.size) << fr_s.sub(/0+$/, '')
+ end
+ add_tz(s)
+ end
+end
+
+class XSDTime < XSDAnySimpleType
+ include XSDDateTimeImpl
+ Type = QName.new(Namespace, TimeLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def set_str(t)
+ /^(\d\d):(\d\d):(\d\d(?:\.(\d*))?)(Z|(?:([+\-])(\d\d):(\d\d))?)?$/ =~ trim(t.to_s)
+ unless Regexp.last_match
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ t }'.")
+ end
+
+ hour = $1.to_i
+ min = $2.to_i
+ sec = $3.to_i
+ secfrac = $4
+ zonestr = $5
+
+ @data = DateTime.civil(1, 1, 1, hour, min, sec, tz2of(zonestr))
+
+ if secfrac
+ @data += secfrac.to_i.to_r / (10 ** secfrac.size) / SecInDay
+ end
+ end
+
+ def _to_s
+ s = format('%02d:%02d:%02d', @data.hour, @data.min, @data.sec)
+ if @data.sec_fraction.nonzero?
+ fr = @data.sec_fraction * SecInDay
+ shiftsize = fr.denominator.to_s.size
+ fr_s = (fr * (10 ** shiftsize)).to_i.to_s
+ s << '.' << '0' * (shiftsize - fr_s.size) << fr_s.sub(/0+$/, '')
+ end
+ add_tz(s)
+ end
+end
+
+class XSDDate < XSDAnySimpleType
+ include XSDDateTimeImpl
+ Type = QName.new(Namespace, DateLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def set_str(t)
+ /^([+\-]?\d\d\d\d\d*)-(\d\d)-(\d\d)(Z|(?:([+\-])(\d\d):(\d\d))?)?$/ =~ trim(t.to_s)
+ unless Regexp.last_match
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ t }'.")
+ end
+
+ year = $1.to_i
+ if year < 0
+ year += 1
+ end
+ mon = $2.to_i
+ mday = $3.to_i
+ zonestr = $4
+
+ @data = DateTime.civil(year, mon, mday, 0, 0, 0, tz2of(zonestr))
+ end
+
+ def _to_s
+ year = (@data.year > 0) ? @data.year : @data.year - 1
+ s = format('%.4d-%02d-%02d', year, @data.mon, @data.mday)
+ add_tz(s)
+ end
+end
+
+class XSDGYearMonth < XSDAnySimpleType
+ include XSDDateTimeImpl
+ Type = QName.new(Namespace, GYearMonthLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def set_str(t)
+ /^([+\-]?\d\d\d\d\d*)-(\d\d)(Z|(?:([+\-])(\d\d):(\d\d))?)?$/ =~ trim(t.to_s)
+ unless Regexp.last_match
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ t }'.")
+ end
+
+ year = $1.to_i
+ if year < 0
+ year += 1
+ end
+ mon = $2.to_i
+ zonestr = $3
+
+ @data = DateTime.civil(year, mon, 1, 0, 0, 0, tz2of(zonestr))
+ end
+
+ def _to_s
+ year = (@data.year > 0) ? @data.year : @data.year - 1
+ s = format('%.4d-%02d', year, @data.mon)
+ add_tz(s)
+ end
+end
+
+class XSDGYear < XSDAnySimpleType
+ include XSDDateTimeImpl
+ Type = QName.new(Namespace, GYearLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def set_str(t)
+ /^([+\-]?\d\d\d\d\d*)(Z|(?:([+\-])(\d\d):(\d\d))?)?$/ =~ trim(t.to_s)
+ unless Regexp.last_match
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ t }'.")
+ end
+
+ year = $1.to_i
+ if year < 0
+ year += 1
+ end
+ zonestr = $2
+
+ @data = DateTime.civil(year, 1, 1, 0, 0, 0, tz2of(zonestr))
+ end
+
+ def _to_s
+ year = (@data.year > 0) ? @data.year : @data.year - 1
+ s = format('%.4d', year)
+ add_tz(s)
+ end
+end
+
+class XSDGMonthDay < XSDAnySimpleType
+ include XSDDateTimeImpl
+ Type = QName.new(Namespace, GMonthDayLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def set_str(t)
+ /^(\d\d)-(\d\d)(Z|(?:[+\-]\d\d:\d\d)?)?$/ =~ trim(t.to_s)
+ unless Regexp.last_match
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ t }'.")
+ end
+
+ mon = $1.to_i
+ mday = $2.to_i
+ zonestr = $3
+
+ @data = DateTime.civil(1, mon, mday, 0, 0, 0, tz2of(zonestr))
+ end
+
+ def _to_s
+ s = format('%02d-%02d', @data.mon, @data.mday)
+ add_tz(s)
+ end
+end
+
+class XSDGDay < XSDAnySimpleType
+ include XSDDateTimeImpl
+ Type = QName.new(Namespace, GDayLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def set_str(t)
+ /^(\d\d)(Z|(?:[+\-]\d\d:\d\d)?)?$/ =~ trim(t.to_s)
+ unless Regexp.last_match
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ t }'.")
+ end
+
+ mday = $1.to_i
+ zonestr = $2
+
+ @data = DateTime.civil(1, 1, mday, 0, 0, 0, tz2of(zonestr))
+ end
+
+ def _to_s
+ s = format('%02d', @data.mday)
+ add_tz(s)
+ end
+end
+
+class XSDGMonth < XSDAnySimpleType
+ include XSDDateTimeImpl
+ Type = QName.new(Namespace, GMonthLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def set_str(t)
+ /^(\d\d)(Z|(?:[+\-]\d\d:\d\d)?)?$/ =~ trim(t.to_s)
+ unless Regexp.last_match
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ t }'.")
+ end
+
+ mon = $1.to_i
+ zonestr = $2
+
+ @data = DateTime.civil(1, mon, 1, 0, 0, 0, tz2of(zonestr))
+ end
+
+ def _to_s
+ s = format('%02d', @data.mon)
+ add_tz(s)
+ end
+end
+
+class XSDHexBinary < XSDAnySimpleType
+ Type = QName.new(Namespace, HexBinaryLiteral)
+
+ # String in Ruby could be a binary.
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+ def set_encoded(value)
+ if /^[0-9a-fA-F]*$/ !~ value
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ value }'.")
+ end
+ @data = trim(String.new(value))
+ @is_nil = false
+ end
+
+ def string
+ [@data].pack("H*")
+ end
+
+private
+
+ def _set(value)
+ @data = value.unpack("H*")[0]
+ @data.tr!('a-f', 'A-F')
+ end
+end
+
+class XSDBase64Binary < XSDAnySimpleType
+ Type = QName.new(Namespace, Base64BinaryLiteral)
+
+ # String in Ruby could be a binary.
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+ def set_encoded(value)
+ if /^[A-Za-z0-9+\/=]*$/ !~ value
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ value }'.")
+ end
+ @data = trim(String.new(value))
+ @is_nil = false
+ end
+
+ def string
+ @data.unpack("m")[0]
+ end
+
+private
+
+ def _set(value)
+ @data = trim([value].pack("m"))
+ end
+end
+
+class XSDAnyURI < XSDAnySimpleType
+ Type = QName.new(Namespace, AnyURILiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def _set(value)
+ begin
+ @data = URI.parse(trim(value.to_s))
+ rescue URI::InvalidURIError
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ value }'.")
+ end
+ end
+end
+
+class XSDQName < XSDAnySimpleType
+ Type = QName.new(Namespace, QNameLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def _set(value)
+ /^(?:([^:]+):)?([^:]+)$/ =~ trim(value.to_s)
+ unless Regexp.last_match
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ value }'.")
+ end
+
+ @prefix = $1
+ @localpart = $2
+ @data = _to_s
+ end
+
+ def _to_s
+ if @prefix
+ "#{ @prefix }:#{ @localpart }"
+ else
+ "#{ @localpart }"
+ end
+ end
+end
+
+
+###
+## Derived types
+#
+class XSDNormalizedString < XSDString
+ Type = QName.new(Namespace, NormalizedStringLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def _set(value)
+ if /[\t\r\n]/ =~ value
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ value }'.")
+ end
+ super
+ end
+end
+
+class XSDInteger < XSDDecimal
+ Type = QName.new(Namespace, IntegerLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def set_str(str)
+ begin
+ @data = Integer(str)
+ rescue ArgumentError
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+ end
+
+ def _to_s()
+ @data.to_s
+ end
+end
+
+class XSDLong < XSDInteger
+ Type = QName.new(Namespace, LongLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def set_str(str)
+ begin
+ @data = Integer(str)
+ rescue ArgumentError
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+ unless validate(@data)
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+ end
+
+ MaxInclusive = +9223372036854775807
+ MinInclusive = -9223372036854775808
+ def validate(v)
+ ((MinInclusive <= v) && (v <= MaxInclusive))
+ end
+end
+
+class XSDInt < XSDLong
+ Type = QName.new(Namespace, IntLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def set_str(str)
+ begin
+ @data = Integer(str)
+ rescue ArgumentError
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+ unless validate(@data)
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+ end
+
+ MaxInclusive = +2147483647
+ MinInclusive = -2147483648
+ def validate(v)
+ ((MinInclusive <= v) && (v <= MaxInclusive))
+ end
+end
+
+class XSDShort < XSDInt
+ Type = QName.new(Namespace, ShortLiteral)
+
+ def initialize(value = nil)
+ super()
+ @type = Type
+ set(value) if value
+ end
+
+private
+
+ def set_str(str)
+ begin
+ @data = Integer(str)
+ rescue ArgumentError
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+ unless validate(@data)
+ raise ValueSpaceError.new("#{ type }: cannot accept '#{ str }'.")
+ end
+ end
+
+ MaxInclusive = +32767
+ MinInclusive = -32768
+ def validate(v)
+ ((MinInclusive <= v) && (v <= MaxInclusive))
+ end
+end
+
+
+end
diff --git a/lib/xsd/datatypes1999.rb b/lib/xsd/datatypes1999.rb
new file mode 100644
index 0000000000..6b6b6be20a
--- /dev/null
+++ b/lib/xsd/datatypes1999.rb
@@ -0,0 +1,31 @@
+=begin
+XSD4R - XML Schema Datatype 1999 support
+Copyright (C) 2001, 2003 NAKAMURA, Hiroshi.
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 2 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PRATICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+this program; if not, write to the Free Software Foundation, Inc., 675 Mass
+Ave, Cambridge, MA 02139, USA.
+=end
+
+
+require 'xsd/datatypes'
+
+
+module XSD
+ Namespace.replace('http://www.w3.org/1999/XMLSchema')
+ InstanceNamespace.replace('http://www.w3.org/1999/XMLSchema-instance')
+ AnyTypeLiteral.replace('ur-type')
+ AnySimpleTypeLiteral.replace('ur-type')
+ NilLiteral.replace('null')
+ NilValue.replace('1')
+ DateTimeLiteral.replace('timeInstant')
+end
diff --git a/lib/xsd/iconvcharset.rb b/lib/xsd/iconvcharset.rb
new file mode 100644
index 0000000000..f607b7db45
--- /dev/null
+++ b/lib/xsd/iconvcharset.rb
@@ -0,0 +1,44 @@
+=begin
+XSD4R - Charset handling with iconv.
+Copyright (C) 2003 NAKAMURA, Hiroshi.
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 2 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PRATICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+this program; if not, write to the Free Software Foundation, Inc., 675 Mass
+Ave, Cambridge, MA 02139, USA.
+=end
+
+
+require 'iconv'
+
+
+module XSD
+
+
+class IconvCharset
+ def self.safe_iconv(to, from, str)
+ iconv = Iconv.new(to, from)
+ out = ""
+ begin
+ out << iconv.iconv(str)
+ rescue Iconv::IllegalSequence => e
+ out << e.success
+ ch, str = e.failed.split(//, 2)
+ out << '?'
+ STDERR.puts("Failed to convert #{ch}")
+ retry
+ end
+ return out
+ end
+end
+
+
+end
diff --git a/lib/xsd/namedelements.rb b/lib/xsd/namedelements.rb
new file mode 100644
index 0000000000..76f958a9e5
--- /dev/null
+++ b/lib/xsd/namedelements.rb
@@ -0,0 +1,86 @@
+=begin
+XSD4R - WSDL named element collection.
+Copyright (C) 2002, 2003 NAKAMURA, Hiroshi.
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 2 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PRATICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+this program; if not, write to the Free Software Foundation, Inc., 675 Mass
+Ave, Cambridge, MA 02139, USA.
+=end
+
+
+module XSD
+
+
+class NamedElements
+ include Enumerable
+
+ def initialize
+ @elements = []
+ @cache = {}
+ end
+
+ def dup
+ o = NamedElements.new
+ o.elements = @elements.dup
+ o
+ end
+
+ def size
+ @elements.size
+ end
+
+ def [](idx)
+ if idx.is_a?(Numeric)
+ @elements[idx]
+ else
+ @cache[idx] ||= @elements.find { |item| item.name == idx }
+ end
+ end
+
+ def find_name(name)
+ @elements.find { |item| item.name.name == name }
+ end
+
+ def each
+ @elements.each do |element|
+ yield(element)
+ end
+ end
+
+ def <<(rhs)
+ @elements << rhs
+ self
+ end
+
+ def +(rhs)
+ o = NamedElements.new
+ o.elements = @elements + rhs.elements
+ o
+ end
+
+ def concat(rhs)
+ @elements.concat(rhs.elements)
+ self
+ end
+
+protected
+
+ def elements=(rhs)
+ @elements = rhs
+ end
+
+ def elements
+ @elements
+ end
+end
+
+end
diff --git a/lib/xsd/ns.rb b/lib/xsd/ns.rb
new file mode 100644
index 0000000000..0767b2c30d
--- /dev/null
+++ b/lib/xsd/ns.rb
@@ -0,0 +1,141 @@
+=begin
+XSD4R - XML Schema Namespace library
+Copyright (C) 2000, 2001, 2002, 2003 NAKAMURA, Hiroshi.
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 2 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PRATICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+this program; if not, write to the Free Software Foundation, Inc., 675 Mass
+Ave, Cambridge, MA 02139, USA.
+=end
+
+
+require 'xsd/datatypes'
+
+
+module XSD
+
+
+class NS
+ class Assigner
+ def initialize
+ @count = 0
+ end
+
+ def assign(ns)
+ @count += 1
+ "n#{ @count }"
+ end
+ end
+
+ attr_reader :default_namespace
+
+ class FormatError < Error; end
+
+public
+
+ def initialize(tag2ns = {})
+ @tag2ns = tag2ns
+ @assigner = nil
+ @ns2tag = {}
+ @tag2ns.each do |tag, ns|
+ @ns2tag[ns] = tag
+ end
+ @default_namespace = nil
+ end
+
+ def assign(ns, tag = nil)
+ if (tag == '')
+ @default_namespace = ns
+ tag
+ else
+ @assigner ||= Assigner.new
+ tag ||= @assigner.assign(ns)
+ @ns2tag[ns] = tag
+ @tag2ns[tag] = ns
+ tag
+ end
+ end
+
+ def assigned?(ns)
+ @ns2tag.key?(ns)
+ end
+
+ def assigned_tag?(tag)
+ @tag2ns.key?(tag)
+ end
+
+ def clone_ns
+ cloned = NS.new(@tag2ns.dup)
+ cloned.assigner = @assigner
+ cloned.assign(@default_namespace, '') if @default_namespace
+ cloned
+ end
+
+ def name(name)
+ if (name.namespace == @default_namespace)
+ name.name
+ elsif @ns2tag.key?(name.namespace)
+ @ns2tag[name.namespace] + ':' << name.name
+ else
+ raise FormatError.new('Namespace: ' << name.namespace << ' not defined yet.')
+ end
+ end
+
+ def compare(ns, name, rhs)
+ if (ns == @default_namespace)
+ return true if (name == rhs)
+ end
+ @tag2ns.each do |assigned_tag, assigned_ns|
+ if assigned_ns == ns && "#{ assigned_tag }:#{ name }" == rhs
+ return true
+ end
+ end
+ false
+ end
+
+ # $1 and $2 are necessary.
+ ParseRegexp = Regexp.new('^([^:]+)(?::(.+))?$')
+
+ def parse(elem)
+ ns = nil
+ name = nil
+ ParseRegexp =~ elem
+ if $2
+ ns = @tag2ns[$1]
+ name = $2
+ if !ns
+ raise FormatError.new('Unknown namespace qualifier: ' << $1)
+ end
+ elsif $1
+ ns = @default_namespace
+ name = $1
+ end
+ if !name
+ raise FormatError.new("Illegal element format: #{ elem }")
+ end
+ XSD::QName.new(ns, name)
+ end
+
+ def each_ns
+ @ns2tag.each do |ns, tag|
+ yield(ns, tag)
+ end
+ end
+
+protected
+
+ def assigner=(assigner)
+ @assigner = assigner
+ end
+end
+
+
+end
diff --git a/lib/xsd/qname.rb b/lib/xsd/qname.rb
new file mode 100644
index 0000000000..150a837c1d
--- /dev/null
+++ b/lib/xsd/qname.rb
@@ -0,0 +1,77 @@
+=begin
+XSD4R - XML QName definition.
+Copyright (C) 2002, 2003 NAKAMURA, Hiroshi.
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 2 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PRATICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+this program; if not, write to the Free Software Foundation, Inc., 675 Mass
+Ave, Cambridge, MA 02139, USA.
+=end
+
+
+module XSD
+
+
+class QName
+ attr_accessor :namespace
+ attr_accessor :name
+
+ def initialize(namespace = nil, name = nil)
+ @namespace = namespace
+ @name = name
+ end
+
+ def dup_name(name)
+ self.class.new(@namespace, name)
+ end
+
+ def match(rhs)
+ unless self.class === rhs
+ return false
+ end
+ if rhs.namespace and (rhs.namespace != @namespace)
+ return false
+ end
+ if rhs.name and (rhs.name != @name)
+ return false
+ end
+ true
+ end
+
+ def ==(rhs)
+ (self.class === rhs && @namespace == rhs.namespace && @name == rhs.name)
+ end
+
+ def ===(rhs)
+ (self == rhs)
+ end
+
+ def eql?(rhs)
+ (self == rhs)
+ end
+
+ def hash
+ @namespace.hash ^ @name.hash
+ end
+
+ def to_s
+ "{#{ namespace }}#{ name }"
+ end
+
+ NormalizedNameRegexp = /^\{([^}]*)\}(.*)$/
+ def parse(str)
+ NormalizedNameRegexp =~ str
+ self.new($1, $2)
+ end
+end
+
+
+end
diff --git a/lib/xsd/xmlparser.rb b/lib/xsd/xmlparser.rb
new file mode 100644
index 0000000000..7d6d389261
--- /dev/null
+++ b/lib/xsd/xmlparser.rb
@@ -0,0 +1,72 @@
+=begin
+XSD4R - XML Instance parser library.
+Copyright (C) 2002, 2003 NAKAMURA, Hiroshi.
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 2 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PRATICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+this program; if not, write to the Free Software Foundation, Inc., 675 Mass
+Ave, Cambridge, MA 02139, USA.
+=end
+
+
+require 'xsd/xmlparser/parser'
+
+
+module XSD
+
+
+module XMLParser
+ def create_parser(host, opt)
+ XSD::XMLParser::Parser.create_parser(host, opt)
+ end
+ module_function :create_parser
+
+ # $1 is necessary.
+ NSParseRegexp = Regexp.new('^xmlns:?(.*)$')
+
+ def filter_ns(ns, attrs)
+ return attrs if attrs.nil? or attrs.empty?
+ newattrs = {}
+ attrs.each do |key, value|
+ if (NSParseRegexp =~ key)
+ # '' means 'default namespace'.
+ tag = $1 || ''
+ ns.assign(value, tag)
+ else
+ newattrs[key] = value
+ end
+ end
+ newattrs
+ end
+ module_function :filter_ns
+end
+
+
+end
+
+
+# Try to load XML processor.
+loaded = false
+[
+ 'xsd/xmlparser/xmlscanner',
+ 'xsd/xmlparser/xmlparser',
+ 'xsd/xmlparser/rexmlparser',
+].each do |lib|
+ begin
+ require lib
+ loaded = true
+ break
+ rescue LoadError
+ end
+end
+unless loaded
+ raise RuntimeError.new("XML processor module not found.")
+end
diff --git a/lib/xsd/xmlparser/parser.rb b/lib/xsd/xmlparser/parser.rb
new file mode 100644
index 0000000000..0c7fd48084
--- /dev/null
+++ b/lib/xsd/xmlparser/parser.rb
@@ -0,0 +1,107 @@
+=begin
+XSD4R - XML Instance parser library.
+Copyright (C) 2002, 2003 NAKAMURA, Hiroshi.
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 2 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PRATICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+this program; if not, write to the Free Software Foundation, Inc., 675 Mass
+Ave, Cambridge, MA 02139, USA.
+=end
+
+
+require 'xsd/qname'
+require 'xsd/ns'
+require 'xsd/charset'
+
+
+module XSD
+module XMLParser
+
+
+class Parser
+ class ParseError < Error; end
+ class FormatDecodeError < ParseError; end
+ class UnknownElementError < FormatDecodeError; end
+ class UnknownAttributeError < FormatDecodeError; end
+ class UnexpectedElementError < FormatDecodeError; end
+ class ElementConstraintError < FormatDecodeError; end
+
+ @@parser_factory = nil
+
+ def self.factory
+ @@parser_factory
+ end
+
+ def self.create_parser(host, opt = {})
+ @@parser_factory.new(host, opt)
+ end
+
+ def self.add_factory(factory)
+ if $DEBUG
+ puts "Set #{ factory } as XML processor."
+ end
+ @@parser_factory = factory
+ end
+
+public
+
+ attr_accessor :charset
+
+ def initialize(host, opt = {})
+ @host = host
+ @charset = opt[:charset] || 'us-ascii'
+ end
+
+ def parse(string_or_readable)
+ @textbuf = ''
+ prologue
+ do_parse(string_or_readable)
+ epilogue
+ end
+
+private
+
+ def do_parse(string_or_readable)
+ raise NotImplementError.new(
+ 'Method do_parse must be defined in derived class.')
+ end
+
+ def start_element(name, attrs)
+ @host.start_element(name, attrs)
+ end
+
+ def characters(text)
+ @host.characters(text)
+ end
+
+ def end_element(name)
+ @host.end_element(name)
+ end
+
+ def prologue
+ end
+
+ def epilogue
+ end
+
+ def xmldecl_encoding=(charset)
+ if @charset.nil?
+ @charset = charset
+ else
+ # Definition in a stream (like HTTP) has a priority.
+ p "encoding definition: #{ charset } is ignored." if $DEBUG
+ end
+ end
+end
+
+
+end
+end
diff --git a/lib/xsd/xmlparser/rexmlparser.rb b/lib/xsd/xmlparser/rexmlparser.rb
new file mode 100644
index 0000000000..2500d432d8
--- /dev/null
+++ b/lib/xsd/xmlparser/rexmlparser.rb
@@ -0,0 +1,65 @@
+=begin
+XSD4R - REXMLParser XML parser library.
+Copyright (C) 2002, 2003 NAKAMURA, Hiroshi.
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 2 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PRATICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+this program; if not, write to the Free Software Foundation, Inc., 675 Mass
+Ave, Cambridge, MA 02139, USA.
+=end
+
+
+require 'xsd/xmlparser'
+require 'rexml/streamlistener'
+require 'rexml/document'
+
+
+module XSD
+module XMLParser
+
+
+class REXMLParser < XSD::XMLParser::Parser
+ include REXML::StreamListener
+
+ def do_parse(string_or_readable)
+ source = nil
+ source = REXML::SourceFactory.create_from(string_or_readable)
+ source.encoding = charset if charset
+ # Listener passes a String in utf-8.
+ @charset = 'utf-8'
+ REXML::Document.parse_stream(source, self)
+ end
+
+ def epilogue
+ end
+
+ def tag_start(name, attrs)
+ start_element(name, attrs)
+ end
+
+ def tag_end(name)
+ end_element(name)
+ end
+
+ def text(text)
+ characters(text)
+ end
+
+ def xmldecl(version, encoding, standalone)
+ # Version should be checked.
+ end
+
+ add_factory(self)
+end
+
+
+end
+end
diff --git a/lib/xsd/xmlparser/xmlparser.rb b/lib/xsd/xmlparser/xmlparser.rb
new file mode 100644
index 0000000000..f555b99b26
--- /dev/null
+++ b/lib/xsd/xmlparser/xmlparser.rb
@@ -0,0 +1,61 @@
+=begin
+XSD4R - XMLParser XML parser library.
+Copyright (C) 2001, 2003 NAKAMURA, Hiroshi.
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 2 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PRATICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+this program; if not, write to the Free Software Foundation, Inc., 675 Mass
+Ave, Cambridge, MA 02139, USA.
+=end
+
+
+require 'xsd/xmlparser'
+require 'xml/parser'
+
+
+module XSD
+module XMLParser
+
+
+class XMLParser < XSD::XMLParser::Parser
+ class Listener < XML::Parser
+ begin
+ require 'xml/encoding-ja'
+ include XML::Encoding_ja
+ rescue LoadError
+ # uconv may not be installed.
+ end
+ end
+
+ def do_parse(string_or_readable)
+ # XMLParser passes a String in utf-8.
+ @charset = 'utf-8'
+ @parser = Listener.new
+ @parser.parse(string_or_readable) do |type, name, data|
+ case type
+ when XML::Parser::START_ELEM
+ start_element(name, data)
+ when XML::Parser::END_ELEM
+ end_element(name)
+ when XML::Parser::CDATA
+ characters(data)
+ else
+ raise FormatDecodeError.new("Unexpected XML: #{ type }/#{ name }/#{ data }.")
+ end
+ end
+ end
+
+ add_factory(self)
+end
+
+
+end
+end
diff --git a/lib/xsd/xmlparser/xmlscanner.rb b/lib/xsd/xmlparser/xmlscanner.rb
new file mode 100644
index 0000000000..c10e275b9e
--- /dev/null
+++ b/lib/xsd/xmlparser/xmlscanner.rb
@@ -0,0 +1,158 @@
+=begin
+XSD4R - XMLScan XML parser library.
+Copyright (C) 2002, 2003 NAKAMURA, Hiroshi.
+
+This program is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free Software
+Foundation; either version 2 of the License, or (at your option) any later
+version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PRATICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+this program; if not, write to the Free Software Foundation, Inc., 675 Mass
+Ave, Cambridge, MA 02139, USA.
+=end
+
+
+require 'xsd/xmlparser'
+require 'xmlscan/scanner'
+
+
+module XSD
+module XMLParser
+
+
+class XMLScanner < XSD::XMLParser::Parser
+ include XMLScan::Visitor
+
+ def do_parse(string_or_readable)
+ @attrs = {}
+ @curattr = nil
+ @scanner = XMLScan::XMLScanner.new(self)
+ @scanner.kcode = ::XSD::Charset.charset_str(charset) if charset
+ @scanner.parse(string_or_readable)
+ end
+
+ def scanner_kcode=(charset)
+ @scanner.kcode = ::XSD::Charset.charset_str(charset) if charset
+ self.xmldecl_encoding = charset
+ end
+
+ ENTITY_REF_MAP = {
+ 'lt' => '<',
+ 'gt' => '>',
+ 'amp' => '&',
+ 'quot' => '"',
+ 'apos' => '\''
+ }
+
+ def parse_error(msg)
+ raise ParseError.new(msg)
+ end
+
+ def wellformed_error(msg)
+ raise NotWellFormedError.new(msg)
+ end
+
+ def valid_error(msg)
+ raise NotValidError.new(msg)
+ end
+
+ def warning(msg)
+ p msg if $DEBUG
+ end
+
+ # def on_xmldecl; end
+
+ def on_xmldecl_version(str)
+ # 1.0 expected.
+ end
+
+ def on_xmldecl_encoding(str)
+ self.scanner_kcode = str
+ end
+
+ # def on_xmldecl_standalone(str); end
+
+ # def on_xmldecl_other(name, value); end
+
+ # def on_xmldecl_end; end
+
+ # def on_doctype(root, pubid, sysid); end
+
+ # def on_prolog_space(str); end
+
+ # def on_comment(str); end
+
+ # def on_pi(target, pi); end
+
+ def on_chardata(str)
+ characters(str)
+ end
+
+ # def on_cdata(str); end
+
+ def on_etag(name)
+ end_element(name)
+ end
+
+ def on_entityref(ref)
+ characters(ENTITY_REF_MAP[ref])
+ end
+
+ def on_charref(code)
+ characters([code].pack('U'))
+ end
+
+ def on_charref_hex(code)
+ on_charref(code)
+ end
+
+ # def on_start_document; end
+
+ # def on_end_document; end
+
+ def on_stag(name)
+ @attrs = {}
+ end
+
+ def on_attribute(name)
+ @attrs[name] = @curattr = ''
+ end
+
+ def on_attr_value(str)
+ @curattr << str
+ end
+
+ def on_attr_entityref(ref)
+ @curattr << ENTITY_REF_MAP[ref]
+ end
+
+ def on_attr_charref(code)
+ @curattr << [code].pack('U')
+ end
+
+ def on_attr_charref_hex(code)
+ on_attr_charref(code)
+ end
+
+ # def on_attribute_end(name); end
+
+ def on_stag_end_empty(name)
+ on_stag_end(name)
+ on_etag(name)
+ end
+
+ def on_stag_end(name)
+ start_element(name, @attrs)
+ end
+
+ add_factory(self)
+end
+
+
+end
+end