Basic pull parsing setup.

This parser extends the regular DOM parser but instead delegates certain nodes to a block instead of building a DOM tree. The API is a bit raw in its current form but I'll extend it and make it a bit more user friendly in the following commits. In particular I want to make it easier to figure out if a certain node is nested inside another node.
2014-04-28 17:22:17 +02:00 · 2014-04-28 17:22:17 +02:00 · 030a0068bd
parent fd5bbbc9a2
commit 030a0068bd
4 changed files with 136 additions and 0 deletions
--- a/lib/oga.rb
+++ b/lib/oga.rb
@ -2,6 +2,8 @@ require 'set'

 require_relative 'oga/xml/lexer'
 require_relative 'oga/xml/parser'
+require_relative 'oga/xml/pull_parser'
+
 require_relative 'oga/xml/node'
 require_relative 'oga/xml/element'
 require_relative 'oga/xml/document'
--- a/lib/oga/xml/pull_parser.rb
+++ b/lib/oga/xml/pull_parser.rb
@ -0,0 +1,86 @@
+module Oga
+  module XML
+    ##
+    # The PullParser class can be used to parse an XML document incrementally
+    # instead of parsing it as a whole. This results in lower memory usage and
+    # potentially faster parsing times. The downside is that pull parsers are
+    # typically more difficult to use compared to DOM parsers.
+    #
+    # Basic parsing using this class works as following:
+    #
+    #     parser = Oga::XML::PullParser.new('... xml here ...')
+    #
+    #     parser.parse do |node|
+    #       if node.is_a?(Oga::XML::PullParser)
+    #
+    #       end
+    #     end
+    #
+    # This parses yields proper XML instances such as {Oga::XML::Element}.
+    # Doctypes and XML declarations are ignored by this parser.
+    #
+    class PullParser < Parser
+      ##
+      # @return [Array]
+      #
+      DISABLED_CALLBACKS = [
+        :on_document,
+        :on_doctype,
+        :on_xml_decl,
+        :on_element_children
+      ]
+
+      ##
+      # @return [Array]
+      #
+      BLOCK_CALLBACKS = [
+        :on_cdata,
+        :on_comment,
+        :on_text,
+        :on_element
+      ]
+
+      ##
+      # @see Oga::XML::Parser#reset
+      #
+      def reset
+        super
+
+        @block = nil
+      end
+
+      ##
+      # Parses the input and yields every node to the supplied block.
+      #
+      # @yieldparam [Oga::XML::Node]
+      #
+      def parse(&block)
+        @block = block
+
+        yyparse(self, :yield_next_token)
+
+        reset
+
+        return
+      end
+
+      # eval is a heck of a lot faster than define_method on both Rubinius and
+      # JRuby.
+      DISABLED_CALLBACKS.each do |method|
+        eval <<-EOF, nil, __FILE__, __LINE__ + 1
+        def #{method}(*_)
+          return
+        end
+        EOF
+      end
+
+      BLOCK_CALLBACKS.each do |method|
+        eval <<-EOF, nil, __FILE__, __LINE__ + 1
+        def #{method}(*args)
+          @block.call(super)
+        end
+        EOF
+      end
+    end # PullParser
+  end # XML
+end # Oga
--- a/spec/oga/xml/pull_parser/doctype_spec.rb
+++ b/spec/oga/xml/pull_parser/doctype_spec.rb
@ -0,0 +1,19 @@
+require 'spec_helper'
+
+describe Oga::XML::PullParser do
+  context 'doctypes' do
+    before :all do
+      @parser = described_class.new('<!DOCTYPE html>')
+    end
+
+    example 'ignore doctypes' do
+      amount = 0
+
+      @parser.parse do
+        amount += 1
+      end
+
+      amount.should == 0
+    end
+  end
+end
--- a/spec/oga/xml/pull_parser/element_spec.rb
+++ b/spec/oga/xml/pull_parser/element_spec.rb
@ -0,0 +1,29 @@
+require 'spec_helper'
+
+describe Oga::XML::PullParser do
+  context 'elements' do
+    before :all do
+      @parser = described_class.new('<person>Alice</person>')
+    end
+
+    example 'parse an element' do
+      name = nil
+
+      @parser.parse do |node|
+        name = node.name if node.is_a?(Oga::XML::Element)
+      end
+
+      name.should == 'person'
+    end
+
+    example 'parse the text of an element' do
+      text = nil
+
+      @parser.parse do |node|
+        text = node.text if node.is_a?(Oga::XML::Text)
+      end
+
+      text.should == 'Alice'
+    end
+  end
+end