Option to use Ox as a SAX handler

Gemfile

@@ -7,4 +7,5 @@ group :development, :test do
   gem 'guard-rspec'
   gem 'simplecov', require: false, platforms: :mri
   gem 'activerecord', '~> 4.1'
+  gem 'ox', '>= 2.1.2'
 end

README.md

@@ -4,7 +4,7 @@
 
 ## Description
 
-A declarative SAX parsing library backed by Nokogiri
+A declarative SAX parsing library backed by Nokogiri or Ox
 
 ## Usage
 ```ruby
@@ -104,4 +104,4 @@ MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
 IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
 CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
 TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
-SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

lib/sax-machine.rb

@@ -1,8 +1,18 @@
 require "sax-machine/version"
 require "sax-machine/sax_document"
 require "sax-machine/sax_configure"
-require "sax-machine/sax_handler"
 require "sax-machine/sax_config"
+require "sax-machine/handlers/sax_abstract_handler"
+require "sax-machine/handlers/sax_nokogiri_handler"
 
 module SAXMachine
-end
+  @@handler = :nokogiri
+
+  def self.handler
+    @@handler
+  end
+
+  def self.handler=(handler)
+    @@handler = handler
+  end
+end

lib/sax-machine/sax_handler.rb → lib/sax-machine/handlers/sax_abstract_handler.rb

@@ -1,8 +1,7 @@
-require "nokogiri"
 require "time"
 
 module SAXMachine
-  class SAXHandler < Nokogiri::XML::SAX::Document
+  module SAXAbstractHandler
     NO_BUFFER = :no_buffer
 
     class StackNode < Struct.new(:object, :config, :buffer)
@@ -13,14 +12,14 @@ def initialize(object, config = nil, buffer = NO_BUFFER)
       end
     end
 
-    def initialize(object, on_error = nil, on_warning = nil)
+    def _initialize(object, on_error = nil, on_warning = nil)
       @stack = [ StackNode.new(object) ]
       @parsed_configs = {}
       @on_error = on_error
       @on_warning = on_warning
     end
 
-    def characters(data)
+    def _characters(data)
       node = stack.last
 
       if node.buffer == NO_BUFFER
@@ -29,10 +28,8 @@ def characters(data)
         node.buffer << data
       end
     end
-    alias cdata_block characters
-
-    def start_element(name, attrs = [])
 
+    def _start_element(name, attrs = [])
       name   = normalize_name(name)
       node   = stack.last
       object = node.object
@@ -76,7 +73,7 @@ def start_element(name, attrs = [])
       end
     end
 
-    def end_element(name)
+    def _end_element(name)
       name = normalize_name(name)
 
       start_tag = stack[-2]
@@ -134,30 +131,29 @@ def end_element(name)
       stack.pop
     end
 
-    private
-
-    def mark_as_parsed(object, element_config)
-      unless element_config.collection?
-        @parsed_configs[[object.object_id, element_config.object_id]] = true
+    def _error(string)
+      if @on_error
+        @on_error.call(string)
       end
     end
 
-    def parsed_config?(object, element_config)
-      @parsed_configs[[object.object_id, element_config.object_id]]
-    end
-
-    def warning(string)
+    def _warning(string)
       if @on_warning
         @on_warning.call(string)
       end
     end
 
-    def error(string)
-      if @on_error
-        @on_error.call(string)
+    private
+
+    def mark_as_parsed(object, element_config)
+      unless element_config.collection?
+        @parsed_configs[[object.object_id, element_config.object_id]] = true
       end
     end
 
+    def parsed_config?(object, element_config)
+      @parsed_configs[[object.object_id, element_config.object_id]]
+    end
 
     def sax_config_for(object)
       if object.class.respond_to?(:sax_config)

lib/sax-machine/handlers/sax_nokogiri_handler.rb

@@ -0,0 +1,15 @@
+require "nokogiri"
+
+module SAXMachine
+  class SAXNokogiriHandler < Nokogiri::XML::SAX::Document
+    include SAXAbstractHandler
+
+    alias_method :initialize, :_initialize
+    alias_method :characters, :_characters
+    alias_method :cdata_block, :_characters
+    alias_method :start_element, :_start_element
+    alias_method :end_element, :_end_element
+    alias_method :error, :_error
+    alias_method :warning, :_warning
+  end
+end

lib/sax-machine/handlers/sax_ox_handler.rb

@@ -0,0 +1,40 @@
+require "ox"
+
+module SAXMachine
+  class SAXOxHandler < Ox::Sax
+    include SAXAbstractHandler
+
+    def initialize(*args)
+      _initialize(*args)
+      _reset_element
+    end
+
+    def attr(name, str)
+      @attrs[name] = str
+    end
+
+    def attrs_done
+      _start_element(@element, @attrs)
+      _reset_element
+    end
+
+    def start_element(name)
+      @element = name
+    end
+
+    def error(message, line, column)
+      _error("#{message} on line #{line} column #{column}")
+    end
+
+    alias_method :text, :_characters
+    alias_method :cdata, :_characters
+    alias_method :end_element, :_end_element
+
+    private
+
+    def _reset_element
+      @attrs = {}
+      @element = ""
+    end
+  end
+end

lib/sax-machine/sax_document.rb

@@ -8,11 +8,24 @@ def self.included(base)
   end
 
   def parse(xml_text, on_error = nil, on_warning = nil)
-    sax_handler = SAXHandler.new(self, on_error, on_warning)
-    parser = Nokogiri::XML::SAX::Parser.new(sax_handler)
-    parser.parse(xml_text) do |ctx|
-      ctx.replace_entities = true
+    if SAXMachine.handler == :ox
+      Ox.sax_parse(
+        SAXOxHandler.new(self, on_error, on_warning),
+        StringIO.new(xml_text),
+        {
+          symbolize: false,
+          convert_special: true,
+          skip: :skip_return,
+        }
+      )
+    else
+      handler = SAXNokogiriHandler.new(self, on_error, on_warning)
+      parser = Nokogiri::XML::SAX::Parser.new(handler)
+      parser.parse(xml_text) do |ctx|
+        ctx.replace_entities = true
+      end
     end
+
     self
   end
 

sax-machine.gemspec

@@ -9,7 +9,7 @@ Gem::Specification.new do |s|
   s.email     = %q{paul@pauldix.net}
   s.homepage  = %q{http://github.com/pauldix/sax-machine}
 
-  s.summary   = %q{Declarative SAX Parsing with Nokogiri}
+  s.summary   = %q{Declarative SAX Parsing with Nokogiri or Ox}
 
   s.license   = %q{MIT}
 

spec/benchmarks/benchmark.rb

@@ -26,7 +26,7 @@ class Atom
     elements :entry, :as => :entries, :class => AtomEntry
   end
 end
-feed_text = File.read("spec/sax-machine/atom.xml")
+feed_text = File.read("spec/fixtures/atom.xml")
 
 benchmark do |t|
   t.report("feedzirra") do
@@ -65,7 +65,7 @@ class Atom
 #   element :title, String
 #   has_many :entry, Entry
 # end
-# feed_text = File.read("spec/sax-machine/atom.xml")
+# feed_text = File.read("spec/fixtures/atom.xml")
 # 
 # benchmark do |t|
 #   t.report("sax-machine") do

spec/fixtures/atom-content.html

@@ -0,0 +1,15 @@
+
+<div xmlns="http://www.w3.org/1999/xhtml"><p>In my previous <a href="http://www.pauldix.net/2008/08/serializing-dat.html">post about the speed of serializing data</a>, I concluded that Marshal was the quickest way to get things done. So I set about using Marshal to store some data in an ActiveRecord object. Things worked great at first, but on some test data I got this error: marshal data too short. Luckily, <a href="http://www.brynary.com/">Bryan Helmkamp</a> had helpfully pointed out that there were sometimes problems with storing marshaled data in the database. He said it was best to base64 encode the marshal dump before storing.</p>
+
+<p>I was curious why it was working on some things and not others. It turns out that some types of data being marshaled were causing the error to pop up. Here's the test data I used in my specs:</p>
+<pre>{ :foo =&gt; 3, :bar =&gt; 2 } # hash with symbols for keys and integer values<br />[3, 2.1, 4, 8]&nbsp; &nbsp;&nbsp; &nbsp;&nbsp; &nbsp;&nbsp; # array with integer and float values</pre>
+<p>Everything worked when I switched the array values to all integers so it seems that floats were causing the problem. However, in the interest of keeping everything working regardless of data types, I base64 encoded before going into the database and decoded on the way out.</p>
+
+<p>I also ran the benchmarks again to determine what impact this would have on speed. Here are the results for 100 iterations on a 10k element array and a 10k element hash with and without base64 encode/decode:</p>
+<pre>&nbsp; &nbsp;&nbsp; &nbsp;&nbsp; &nbsp;&nbsp; &nbsp;&nbsp; &nbsp; user&nbsp; &nbsp;&nbsp; &nbsp; system&nbsp; &nbsp;&nbsp; total&nbsp; &nbsp;&nbsp; &nbsp; real<br />array marshal&nbsp; 0.200000&nbsp; &nbsp;0.010000&nbsp; &nbsp;0.210000 (&nbsp; 0.214018) (without Base64)<br />array marshal&nbsp; 0.220000&nbsp; &nbsp;0.010000&nbsp; &nbsp;0.230000 (&nbsp; 0.250260)<br /><br />hash marshal&nbsp; &nbsp;1.830000&nbsp; &nbsp;0.040000&nbsp; &nbsp;1.870000 (&nbsp; 1.892874) (without Base64)<br />hash marshal&nbsp; &nbsp;2.040000&nbsp; &nbsp;0.100000&nbsp; &nbsp;2.140000 (&nbsp; 2.170405)</pre>
+<p>As you can see the difference in speed is pretty negligible. I assume that the error has to do with AR cleaning the stuff that gets inserted into the database, but I'm not really sure. In the end it's just easier to use Base64.encode64 when serializing data into a text field in ActiveRecord using Marshal.</p>
+
+<p>I've also read people posting about this error when using the database session store. I can only assume that it's because they were trying to store either way too much data in their session (too much for a regular text field) or they were storing float values or some other data type that would cause this to pop up. Hopefully this helps.</p></div>
+<div class="feedflare">
+<a href="http://feeds.feedburner.com/~f/PaulDixExplainsNothing?a=rWfWO"><img src="http://feeds.feedburner.com/~f/PaulDixExplainsNothing?i=rWfWO" border="0"></img></a> <a href="http://feeds.feedburner.com/~f/PaulDixExplainsNothing?a=RaCqo"><img src="http://feeds.feedburner.com/~f/PaulDixExplainsNothing?i=RaCqo" border="0"></img></a> <a href="http://feeds.feedburner.com/~f/PaulDixExplainsNothing?a=1CBLo"><img src="http://feeds.feedburner.com/~f/PaulDixExplainsNothing?i=1CBLo" border="0"></img></a>
+</div><img src="http://feeds.feedburner.com/~r/PaulDixExplainsNothing/~4/383536354" height="1" width="1"/>

spec/sax-machine/sax_document_spec.rb

@@ -64,7 +64,7 @@ def title=(val)
         it "should be available" do
           @klass.data_class(:date).should == DateTime
         end
-        
+
         it "should handle an integer class" do
           @klass = Class.new do
             include SAXMachine
@@ -73,15 +73,15 @@ def title=(val)
           document = @klass.parse("<number>5</number>")
           document.number.should == 5
         end
-        
+
         it "should handle an float class" do
           @klass = Class.new do
             include SAXMachine
             element :number, :class => Float
           end
           document = @klass.parse("<number>5.5</number>")
           document.number.should == 5.5
-        end        
+        end
 
         it "should handle an string class" do
           @klass = Class.new do
@@ -91,7 +91,7 @@ def title=(val)
           document = @klass.parse("<number>5.5</number>")
           document.number.should == "5.5"
         end
-        
+
         it "should handle a time class" do
           @klass = Class.new do
             include SAXMachine
@@ -100,7 +100,7 @@ def title=(val)
           document = @klass.parse("<time>1994-02-04T06:20:00Z</time>")
           document.time.should == Time.utc(1994, 2, 4, 6, 20, 0, 0)
         end
-        
+
       end
       describe "the required attribute" do
         it "should be available" do
@@ -466,7 +466,7 @@ class Item
           elements :item, :as => :items, :with => {:type => /Foo/}, :class => Foo
         end
       end
-      
+
       it "should cast into the correct class" do
         document = @klass.parse("<items><item type=\"Bar\"><title>Bar title</title></item><item type=\"Foo\"><title>Foo title</title></item></items>")
         document.items.size.should == 2
@@ -536,22 +536,22 @@ class Foo
       end
     end
   end
-  
+
   describe "when dealing with element names containing dashes" do
-    it 'should automatically convert dashes to underscores' do      
+    it 'should automatically convert dashes to underscores' do
       class Dashes
         include SAXMachine
         element :dashed_element
       end
-      
+
       parsed = Dashes.parse('<dashed-element>Text</dashed-element>')
       parsed.dashed_element.should eq "Text"
     end
   end
 
   describe "full example" do
     before :each do
-      @xml = File.read('spec/sax-machine/atom.xml')
+      @xml = File.read('spec/fixtures/atom.xml')
       class AtomEntry
         include SAXMachine
         element :title
@@ -570,17 +570,21 @@ class Atom
         element :link, :value => :href, :as => :feed_url, :with => {:type => "application/atom+xml"}
         elements :entry, :as => :entries, :class => AtomEntry
       end
+
+      @feed = Atom.parse(@xml)
     end # before
 
     it "should parse the url" do
-      f = Atom.parse(@xml)
-      f.url.should == "http://www.pauldix.net/"
+      @feed.url.should == "http://www.pauldix.net/"
     end
 
     it "should parse entry url" do
-      f = Atom.parse(@xml)
-      f.entries.first.url.should == "http://www.pauldix.net/2008/09/marshal-data-to.html?param1=1&param2=2"
-      f.entries.first.alternate.should == "http://feeds.feedburner.com/~r/PaulDixExplainsNothing/~3/383536354/marshal-data-to.html?param1=1&param2=2"
+      @feed.entries.first.url.should == "http://www.pauldix.net/2008/09/marshal-data-to.html?param1=1&param2=2"
+      @feed.entries.first.alternate.should == "http://feeds.feedburner.com/~r/PaulDixExplainsNothing/~3/383536354/marshal-data-to.html?param1=1&param2=2"
+    end
+
+    it "should parse content" do
+      @feed.entries.first.content.should == File.read('spec/fixtures/atom-content.html')
     end
   end
 
@@ -835,4 +839,25 @@ def title=(blah)
       @item.authors.last.role.should == 'artist'
     end
   end
+
+  describe "with error handling" do
+    before do
+      @xml = %[
+        <item id="1">
+          <title>sweet</title>
+      ]
+
+      class ItemElement5
+        include SAXMachine
+        element :title
+      end
+
+      @errors = []
+      @item = ItemElement5.parse(@xml, ->(x) { @errors << x })
+    end
+
+    it 'should have error' do
+      @errors.uniq.size.should == 1
+    end
+  end
 end