libxml-jruby 1.0.0

data/lib/libxml-jruby/xml/ns.rb

@@ -0,0 +1,19 @@
+module LibXMLJRuby
+  module XML
+    class NS
+      class << self
+        def from_java(java_obj)
+          
+        end
+      end
+      
+      def initialize(node, href, prefix)
+        @node, @href, @prefix = node, href, prefix
+      end
+      
+      attr_reader :href, :prefix
+      def href?; !!href; end
+      def prefix?; !!prefix; end
+    end
+  end
+end

data/lib/libxml-jruby/xml/parser.rb

@@ -0,0 +1,121 @@
+class LibXMLJRuby::XML::Parser
+  class ParseError < RuntimeError; end
+
+  VERNUM = 0
+  VERSION = '0'
+  
+  @default_tree_indent_string = '  '
+  @check_lib_versions = ''
+  @features = []
+  
+  class << self        
+    attr_accessor :debug_entities, :default_compression, 
+                  :default_keep_blanks, :default_line_numbers,
+                  :default_substitute_entities, 
+                  :default_tree_indent_string,
+                  :default_validity_checking, :default_warnings,
+                  :features, :indent_tree_output,
+                  :check_lib_versions
+    
+    def register_error_handler(handler = nil, &block)
+      
+    end
+
+    def file(filename)
+      p = new
+      p.filename = filename
+      p
+    end
+    
+    def string(string)
+      p = new
+      p.string = string
+      p
+    end
+    
+    def io(io)
+      p = new
+      p.io = io
+      p
+    end
+  end
+  
+  def initialize
+    @source_type = nil
+    @parsed = false
+  end
+  
+  attr_reader :string, :filename, :io
+  
+  def string=(string = nil)
+    raise TypeError if string.nil?
+    @parsed = false
+    @source_type = String
+    @string = string
+  end
+  
+  def filename=(filename)
+    @source_type = File
+    @filename = filename
+  end
+
+  def io=(io)
+    raise TypeError if StringIO === io
+    @io = io
+    self.string = @io.read
+  end
+  
+  def parse
+    raise ParseError, "You cannot parse twice" if @parsed
+    raise ParseError, "No input specified, please specify an input" unless @source_type
+    doc = send("parse_#{@source_type.to_s.downcase}")
+    @parsed = true
+    doc
+  end
+  
+  def document_builder_factory
+    @dbf ||= DocumentBuilderFactory.new_instance
+  end
+  
+  def document_builder
+    document_builder_factory.namespace_aware = true
+    document_builder_factory.new_document_builder
+  end
+  
+  def string_reader
+    StringReader.new(@string)
+  end
+  
+  def input_source
+    InputSource.new(string_reader)
+  end
+  
+  def parse_string
+    raise ParseError if @string.empty?
+    builder = document_builder
+    
+    begin
+      jdoc = builder.parse(input_source)
+    rescue NativeException
+      raise ParseError
+    end
+
+    XML::Document.from_java(jdoc)
+  end
+  
+  def parse_file
+    builder = document_builder
+    
+    begin
+      jdoc = builder.parse(@filename)
+    rescue
+      raise ParseError
+    end
+    
+    XML::Document.from_java(jdoc)
+  end
+  
+  def parse_io
+    parse_string
+  end
+end

data/lib/libxml-jruby/xml/xpath.rb

@@ -0,0 +1,98 @@
+module LibXMLJRuby
+  module XML
+    module XPath
+      NODESET = nil
+      
+      class Object
+        include Enumerable
+        
+        def initialize(expr, document, nslist = nil)
+          @expr, @document = expr, document
+          @nslist = nslist          
+        end
+        
+        def each(&block)
+          set.each(&block)
+        end
+        
+        def length
+          set.length
+        end
+        
+        def first
+          set.first
+        end
+        
+        def xpath_type
+          XML::XPath::NODESET
+        end
+        
+        def set
+          @set ||= XML::Node::Set.from_java(evaluate_expression)
+        end
+        
+        def [](index)
+          set[index]
+        end
+        
+        private
+        def xpath_factory
+          @xpath_factory ||= XPathFactory.new_instance
+        end
+        
+        def xpath
+          @xpath ||= xpath_factory.newXPath
+        end
+        
+        def resolver
+          PrefixResolverDefault.new(doc_element)
+        end
+        
+        def doc_element
+          if document.respond_to?(:getDocumentElement)
+            document.getDocumentElement
+          elsif document.respond_to?(:getOwnerDocument)
+            document.getOwnerDocument
+          else
+            nil
+          end
+        end
+        
+        def namespace_context          
+          CustomNamespaceContext.new(resolver)
+        end
+        
+        def evaluate_expression
+          xpath.setNamespaceContext(namespace_context)
+          xpath.evaluate(@expr, document, XPathConstants::NODESET)
+        end
+        
+        def document
+          @document.respond_to?(:java_obj) ? @document.java_obj : @document
+        end
+
+        class CustomNamespaceContext
+          include javax.xml.namespace.NamespaceContext
+
+          attr_reader :resolver
+
+          def initialize(resolver)
+            @resolver = resolver
+          end
+
+        	def getNamespaceURI(prefix)
+        		resolver.getNamespaceForPrefix(prefix)
+        	end
+
+        	def getPrefixes(val)
+        		nil
+        	end
+
+        	def getPrefix(uri)
+            nil
+        	end
+        end
+      end
+    end
+  end
+end

data/lib/libxml.rb

@@ -0,0 +1 @@
+require 'libxml-jruby'

data/lib/xml.rb

@@ -0,0 +1,13 @@
+# This file loads libxml and adds the LibXML namespace
+# to the toplevel for conveneience. The end result
+# is to have XML:: universally exposed.
+#
+# It is recommend that you only load this file for libs
+# that do not have their own namespace, eg. administrative
+# scripts, personal programs, etc. For other applications
+# require 'libxml' instead and include LibXML into your 
+# app/libs namespace.
+
+require 'libxml'
+
+include LibXML

data/lib/xml/libxml.rb

@@ -0,0 +1,8 @@
+# This is here for backward compatibility.
+#
+# TODO: DEPRECATE!
+
+require 'libxml.rb'
+
+include LibXML
+

data/test/etc_doc_to_s.rb

@@ -0,0 +1,19 @@
+require 'xml'
+
+1.times do |count|
+
+  xml_doc = XML::Document.new()
+  xml_doc.encoding = "UTF-8"
+  xml_doc.root = XML::Node.new("Request")
+
+  1000.times do |index|
+    xml_doc.root << node = XML::Node.new("row")
+    node["user_id"] = index.to_s
+    node << "600445"
+  end
+
+  xml_str = xml_doc.to_s
+  print "\r#{count}"
+  $stdout.flush
+end
+puts "\n"

data/test/ets_copy_bug.rb

@@ -0,0 +1,21 @@
+require 'xml'
+
+def test( doc2 )
+  doc = XML::Document.new('1.0')
+  doc.root = XML::Node.new("ccc")
+  doc.root['aaa'] = 'aaa'
+  doc.root.child_add(doc2.root.copy(true)) # BUG!
+  doc.root << doc2.root.copy(true)
+  return doc
+end
+
+def test2
+  doc2 = XML::Document.new('1.0')
+  doc2.root = XML::Node.new("aaa")
+  test( doc2 )
+end
+
+1000.times { |i|
+  print "\r#{i}"; $stdout.flush
+  test2
+}

data/test/ets_copy_bug3.rb

@@ -0,0 +1,38 @@
+require 'xml'
+
+str = <<-STR
+  <html>
+     <body>
+         <div class="textarea" id="t1"
+  style="STATIC">werwerwerwerwer                 </div>
+         <div class="textarea" id="t2" style="STATIC">
+             Quisque et diam dapibus nisi bibendum blandit.
+         </div>
+         <div class="textarea" id="t3" style="STATIC">
+             <p>aaaaaaaaa</p>
+         </div>
+     </body>
+  </html>
+STR
+
+XML::Parser.default_keep_blanks = false
+xp = XML::Parser.new
+xp.string = str
+doc = xp.parse
+
+xpath = "//div[@id='t1']"
+div1 = doc.find(xpath).to_a[0]
+printf "xxx div1: #{div1}\n"
+
+xpath = "//div[@id='t2']"
+div2 = doc.find(xpath).to_a[0]
+printf "xxx div2: #{div2}\n"
+
+
+div2.each do |child|
+   #c = child.clone
+   c = child.copy(false)
+   div1.child_add(c)
+end
+
+printf "xxx root: #{doc.root}\n"

data/test/ets_doc_file.rb

@@ -0,0 +1,15 @@
+require 'xml'
+
+# This is related to bug 8337, complaint is on amd64/fbsd
+# unknown if it happens on other amd64/os combos
+
+Process.setrlimit(Process::RLIMIT_NOFILE,10005)
+
+(1..10000).each{|time|
+  XML::Document.file(File.join(File.dirname(__FILE__),'ets_test.xml'))
+  if time % 100 == 0
+    print "\r#{time}"  
+    $stdout.flush
+  end
+}
+puts "\n"

data/test/ets_doc_to_s.rb

@@ -0,0 +1,21 @@
+require 'xml'
+
+100.times do |count|
+
+  xml_doc = XML::Document.new()
+  xml_doc.encoding = "UTF-8"
+  xml_doc.root = XML::Node.new("Request")
+
+  1000.times do |index|
+
+    xml_doc.root << node = XML::Node.new("row")
+    node["user_id"] = index.to_s
+    node << "600445"
+
+  end
+
+  xml_str = xml_doc.to_s
+  print "\r#{count}"
+  $stdout.flush
+end
+puts "\n"

data/test/ets_gpx.rb

@@ -0,0 +1,26 @@
+require 'libxml'
+
+100.times do
+  doc = XML::Document.new
+  doc.encoding = 'UTF-8'
+
+  root = XML::Node.new 'gpx'
+  root['version'] = '1.0'
+  root['creator'] = 'OpenStreetMap.org'
+  root['xmlns'] = "http://www.topografix.com/GPX/1/0/"
+
+  doc.root = root
+
+  track = XML::Node.new 'trk'
+  doc.root << track
+
+  trkseg = XML::Node.new 'trkseg'
+  track << trkseg
+
+  1.upto(1000) do |n|
+    trkpt = XML::Node.new 'trkpt'
+    trkpt['lat'] = n.to_s
+    trkpt['lon'] = n.to_s
+    trkseg << trkpt
+  end
+end

data/test/ets_node_gc.rb

@@ -0,0 +1,21 @@
+require 'xml'
+
+# test of bug 13310, clears MEM2
+
+include GC
+
+inner = XML::Node.new('inner')
+save = nil
+1.times do
+  outer = XML::Node.new('outer')
+  outer.child = inner
+  1.times do
+    doc = XML::Document.new
+    doc.root = outer
+    # Uncomment the following line and it won't crash
+    save = doc
+  end
+  garbage_collect
+end
+garbage_collect
+puts inner

data/test/ets_tsr.rb

@@ -0,0 +1,9 @@
+require 'xml'
+
+10_000.times {|n|
+j=XML::Node.new2(nil,"happy#{n}")
+  10.times {|r|
+    j1=XML::Node.new("happy#{r}","farts")
+    j.child=j1
+  }
+}