saxerator 0.9.5 → 0.9.8

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 8a9e778a67df70446a761b0c44b299ec71d17767
-  data.tar.gz: 4dd1010e6d76ec961dfaaac08b03e5c4e9179658
+  metadata.gz: e54e504ee27201c8e10d314e067250c1dd779514
+  data.tar.gz: 3a7338b78ab7e05f316c365903c3ab625e81d62d
 SHA512:
-  metadata.gz: f575598f8e4eb2a50a828457107084615cecb12dfc4c1c94bb0b673cf7a6cc0c75771400c38d8b9eb403bc95da1be5f9bfcd7d1ff5be85f171090ef7961a6c30
-  data.tar.gz: b33b5a6cd650a36d77ddf69ce33380aad95cc88d9c25d9000a86d1bd88377e4d07f0c7451a592b3b3c8d377c88de2ece5a72dda3c5be4dfdecceb487ee4768b0
+  metadata.gz: 350eb22db29091954e33bdf251a4e5e8133c3397221f7318d66ab878366862c614096d8b55a63de7d8300070b4ef8c434e0584769e5bbd6da7bb062218a2224e
+  data.tar.gz: 46145d2ff6a104304e0abdf330ecca528d0e1b14bbe5ce789f1016aeb7fe51f77888cce1c101233e75ce37bc3b3af67cfd5469682697a4ec37c3e00d8a446e7e

data/.gitignore

@@ -8,3 +8,5 @@ coverage/
 .rvmrc
 .ruby-version
 .ruby-gemset
+
+/spec/examples.txt

data/.travis.yml

@@ -1,14 +1,11 @@
-script: "rspec spec"
+before_install:
+  - gem install bundler -v 1.12.5
+script: "bundle exec rake spec:adapters"
 language: ruby
-bundler_args: --without coverage
+bundler_args: --without coverage --binstubs
+sudo: false
+cache: bundler
 rvm:
-  - 2.1.0
-  - 2.0.0
-  - 1.9.2
-  - 1.9.3
-  - jruby-19mode
-  - rbx-2.2.1
-matrix:
-  allow_failures:
-    - rvm: jruby-19mode
-    - rvm: rbx-2.2.1
+  - 2.3.1
+  - 2.2.2
+  - 2.1.6

data/Gemfile

@@ -1,8 +1,10 @@
-source "http://rubygems.org"
+source 'http://rubygems.org'
 
 # Specify your gem's dependencies in saxerator.gemspec
 gemspec
 
 group :coverage do
   gem 'simplecov'
-end
+end
+
+gem 'pry', platforms: [:ruby_21, :ruby_22, :ruby_23]

data/README.md

@@ -13,6 +13,15 @@ You can parse any valid xml in 3 simple steps.
 1. Perform your work in an `each` block, or using any [Enumerable](http://apidock.com/ruby/Enumerable)
 method
 
+Installation
+------------
+1. `gem install saxerator`
+1. Choose an xml parser
+    * (default) Use ruby's built-in REXML parser - no other dependencies necessary
+    * `gem install nokogiri`
+    * `gem install ox`
+1. If not using the default, specify your adapter in the [Saxerator configuration](#configuration)
+
 The DSL
 -------
 The DSL consists of predicates that may be combined to describe which elements the parser should enumerate over.
@@ -90,7 +99,8 @@ end
 
 | Setting           | Default | Values          | Description
 |:------------------|:--------|-----------------|------------
-| `output_type`     | `:hash` | `:hash`, `:xml` | The type of object generated by Saxerator's parsing. `:hash` should be self-explanatory, `:xml` generates a `Nokogiri::XML::Document`
+| `adapter`         | `:nokogiri` | `:nokogiri`, `:ox`, `:rexml` | The XML parser used by Saxerator |
+| `output_type`     | `:hash` | `:hash`, `:xml` | The type of object generated by Saxerator's parsing. `:hash` generates a Ruby Hash, `:xml` generates a `REXML::Document`
 | `symbolize_keys!` | n/a     | n/a             | Call this method if you want the hash keys to be symbols rather than strings
 | `ignore_namespaces!`| n/a   | n/a             | Call this method if you want to treat the XML document as if it has no namespace information. It differs slightly from `strip_namespaces!` since it deals with how the XML is processed rather than how it is output
 | `strip_namespaces!`| n/a     | user-specified  | Called with no arguments this strips all namespaces, or you may specify an arbitrary number of namespaces to strip, i.e. `config.strip_namespaces! :rss, :soapenv`
@@ -101,6 +111,12 @@ Known Issues
 * JRuby closes the file stream at the end of parsing, therefor to perform multiple operations
   which parse a file you will need to instantiate a new parser with a new File object.
 
+Other Documentation
+-------------------
+* [REXML](http://www.germane-software.com/software/rexml/) ([api docs](http://ruby-doc.org/stdlib-2.4.0/libdoc/rexml/rdoc/REXML/Document.html))
+* [Nokogiri](http://www.nokogiri.org/) ([api docs](http://www.rubydoc.info/github/sparklemotion/nokogiri))
+* [Ox](https://github.com/ohler55/ox) ([api docs](http://www.ohler.com/ox/))
+
 FAQ
 ---
 Why the name 'Saxerator'?
@@ -129,9 +145,13 @@ When I fetch a tag that has one or more elements, sometimes I get an `Array`, an
   > occasionally comes up is for elements that are sometimes-empty. Empty elements behave mostly like an
   > empty `Hash`, however you may convert it to a more `String`-like object via `#to_s`
 
+###  Contribution ###
+
+For running tests for all parsers run `rake spec:adapters`
+
 ### Acknowledgements ###
-Saxerator was inspired by - but not affiliated with - [nori](https://github.com/savonrb/nori) and [Gregory Brown](http://majesticseacreature.com/)'s
+Saxerator was inspired by - but not affiliated with - [nori](https://github.com/savonrb/nori) and Gregory Brown's
 [Practicing Ruby](http://practicingruby.com/)
 
 #### Legal Stuff ####
-Copyright © Bradley Schaefer. MIT License (see LICENSE file).
+Copyright © 2012-2017 Bradley Schaefer. MIT License (see LICENSE file).

data/Rakefile

@@ -4,4 +4,15 @@ require 'rspec/core/rake_task'
 
 RSpec::Core::RakeTask.new(:spec)
 
-task :default => :spec
+task default: :spec
+
+namespace :spec do
+  desc 'Run specs against all available adapters'
+  task :adapters do |_|
+    %w(nokogiri ox rexml).each do |adapter|
+      ENV['SAXERATOR_ADAPTER'] = adapter
+      Rake::Task['spec'].invoke
+      ::Rake.application['spec'].reenable
+    end
+  end
+end

data/benchmark/benchmark.rb

@@ -1,23 +1,65 @@
-$:.push File.expand_path('../../lib', __FILE__)
+$LOAD_PATH.push File.expand_path('../../lib', __FILE__)
 require 'saxerator'
 require 'benchmark'
 
 file = ARGV.shift
-if !File.exists?(file)
+unless File.exist?(file)
   puts "Cannot find file #{file}"
   exit 1
 end
 file = File.new(file)
 
-count = count2 = count3 = count4 = 0
-Benchmark.bm do |x|
-  x.report('for_tag') { Saxerator.parser(file).for_tag(:artist).each { count = count + 1 } }
-  x.report('at_depth') { Saxerator.parser(file).at_depth(2).each { count2 = count2 + 1 } }
-  x.report('within') { Saxerator.parser(file).within(:artists).each { count3 = count3 + 1 } }
-  x.report('composite') { Saxerator.parser(file).for_tag(:name).within(:artist).at_depth(3).each { count4 = count4 + 1} }
+ADAPTERS = [:nokogiri, :ox].freeze
+
+class SaxeratorBenchmark
+  def initialize(file)
+    @file = file
+  end
+
+  def with_adapter(adapter) # rubocop:disable Metrics/MethodLength
+    puts '@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@'
+    puts
+    puts "Benchmark with :#{adapter} parser"
+    puts
+    puts '@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@'
+    puts
+
+    count = count2 = count3 = count4 = 0
+
+    Benchmark.bm do |x|
+      x.report('for_tag') do
+        Saxerator.parser(@file) { |confing| confing.adapter = adapter }
+                 .for_tag(:artist).each { count += 1 }
+      end
+
+      x.report('at_depth') do
+        Saxerator.parser(@file) { |confing| confing.adapter = adapter }
+                 .at_depth(2).each { count2 += 1 }
+      end
+
+      x.report('within') do
+        Saxerator.parser(@file) { |confing| confing.adapter = adapter }
+                 .within(:artists).each { count3 += 1 }
+      end
+
+      x.report('composite') do
+        Saxerator.parser(@file) { |confing| confing.adapter = adapter }
+                 .for_tag(:name)
+                 .within(:artist).at_depth(3).each { count4 += 1 }
+      end
+    end
+
+    puts
+    puts '##########################################################'
+    puts
+    puts "for_tag:   #{count} artist elements parsed"
+    puts "at_depth:  #{count2} elements parsed"
+    puts "within:    #{count3} artists children parsed"
+    puts "composite: #{count4} names within artist nested 3 tags deep parsed"
+    puts
+  end
 end
 
-puts "for_tag:   #{count} artist elements parsed"
-puts "at_depth:  #{count2} elements parsed"
-puts "within:    #{count3} artists children parsed"
-puts "composite: #{count4} names within artist nested 3 tags deep parsed"
+saxerator_benchmark = SaxeratorBenchmark.new(file)
+
+ADAPTERS.each { |adapter| saxerator_benchmark.with_adapter(adapter) }

data/benchmark/generate_sample_file.rb

@@ -28,4 +28,4 @@ File.open(filename, 'w') do |f|
   end
   f.puts '</artists>'
 end
-puts "DONE!"
+puts 'DONE!'

data/lib/saxerator.rb

@@ -1,7 +1,6 @@
 require 'saxerator/version'
 
-require 'nokogiri'
-
+require 'saxerator/sax_handler'
 require 'saxerator/dsl'
 require 'saxerator/full_document'
 require 'saxerator/document_fragment'
@@ -31,6 +30,6 @@ module Saxerator
     config = Configuration.new
     yield(config) if block_given?
 
-    Saxerator::FullDocument.new(xml, config)
+    FullDocument.new(xml, config)
   end
 end

data/lib/saxerator/adapters/nokogiri.rb

@@ -0,0 +1,39 @@
+require 'forwardable'
+require 'nokogiri'
+
+module Saxerator
+  module Adapters
+    class Nokogiri < ::Nokogiri::XML::SAX::Document
+      extend Forwardable
+
+      def self.parse(source, reader)
+        parser = ::Nokogiri::XML::SAX::Parser.new(new(reader))
+        parser.parse(source)
+      end
+
+      def initialize(reader)
+        @reader = reader
+        @ignore_namespaces = reader.ignore_namespaces?
+      end
+
+      def_delegators :@reader, :start_element, :end_element, :characters
+      def_delegator :@reader, :characters, :cdata_block
+
+      def start_element_namespace(name, attrs = [], _prefix = nil, _uri = nil, _ns = [])
+        return super unless @ignore_namespaces
+        start_element(name, strip_namespace(attrs))
+      end
+
+      def end_element_namespace(name, _prefix = nil, _uri = nil)
+        return super unless @ignore_namespaces
+        end_element(name)
+      end
+
+      private
+
+      def strip_namespace(attrs)
+        attrs.map { |attr| [attr.localname, attr.value] }
+      end
+    end
+  end
+end

data/lib/saxerator/adapters/ox.rb

@@ -0,0 +1,69 @@
+require 'forwardable'
+require 'ox'
+
+module Saxerator
+  module Adapters
+    class Ox # < ::Ox::Sax
+      extend Forwardable
+
+      def self.parse(source, reader)
+        handler = new(reader)
+        ::Ox.sax_parse(handler, source)
+      end
+
+      attr_accessor :name
+      attr_reader :attributes
+      attr_reader :reader
+
+      def initialize(reader)
+        @reader = reader
+
+        @attributes = {}
+        @name = ''
+      end
+
+      def guard!
+        reader.start_element(name, attributes.to_a) unless name.empty?
+        reset!
+      end
+
+      def attr(name, value)
+        attributes[name.to_s] = value
+      end
+
+      def start_element(name)
+        guard!
+
+        name = name.to_s
+        name = strip_namespace(name) if reader.ignore_namespaces?
+        self.name = name
+      end
+
+      def end_element(name)
+        guard!
+
+        name = name.to_s
+        name = strip_namespace(name) if reader.ignore_namespaces?
+        reader.end_element(name)
+      end
+
+      def text(str)
+        guard!
+        reader.characters(str)
+      end
+
+      alias cdata text
+
+      private
+
+      def reset!
+        @attributes.clear
+        @name = ''
+      end
+
+      def strip_namespace(name)
+        name.split(':').last
+      end
+    end
+  end
+end

data/lib/saxerator/adapters/rexml.rb

@@ -0,0 +1,42 @@
+require 'forwardable'
+require 'rexml/document'
+require 'rexml/streamlistener'
+
+module Saxerator
+  module Adapters
+    class Rexml
+      extend Forwardable
+      include REXML::StreamListener
+
+      def self.parse(source, reader)
+        handler = new(reader)
+        REXML::Document.parse_stream(source, handler)
+      end
+
+      def initialize(reader)
+        @reader = reader
+        @ignore_namespaces = reader.ignore_namespaces?
+      end
+
+      def_delegator :@reader, :characters, :text
+      def_delegator :@reader, :characters, :cdata
+
+      def tag_start(name, attrs)
+        name = strip_namespace(name) if @ignore_namespaces
+        @reader.start_element(name, attrs)
+      end
+
+      def tag_end(name)
+        name = strip_namespace(name) if @ignore_namespaces
+        @reader.end_element(name)
+      end
+
+      private
+
+      def strip_namespace(name)
+        name.split(':').last
+      end
+
+    end
+  end
+end

data/lib/saxerator/builder.rb

@@ -13,7 +13,7 @@ module Saxerator
     def camel_case(str)
       str = str.to_s
       return str if str !~ /_/ && str =~ /[A-Z]+.*/
-      str.split('_').map{|e| e.capitalize}.join
+      str.split('_').map(&:capitalize).join
     end
   end
-end
+end

data/lib/saxerator/builder/array_element.rb

@@ -1,9 +1,15 @@
+require 'delegate'
+
 module Saxerator
   module Builder
-    class ArrayElement < Array
+    class ArrayElement < DelegateClass(Array)
       attr_accessor :name
 
-      def to_ary; self end
+      def initialize(arr = [], name = nil)
+        @name = name
+        super(arr)
+      end
+
       def to_a; self end
     end
   end

data/lib/saxerator/builder/empty_element.rb

@@ -4,6 +4,7 @@ module Saxerator
   module Builder
     class EmptyElement < HashElement
       def nil?; true end
+
       def !; true end
 
       def to_s
@@ -15,9 +16,7 @@ module Saxerator
       end
 
       def to_a
-        array = ArrayElement.new
-        array.name = name
-        array
+        ArrayElement.new([], name)
       end
     end
   end

data/lib/saxerator/builder/hash_builder.rb

@@ -45,12 +45,11 @@ module Saxerator
         hash
       end
 
-      def add_to_hash_element( hash, name, element)
+      def add_to_hash_element(hash, name, element)
         name = generate_key(name)
-        if hash[name]
-          unless hash[name].is_a?(Array)
-            hash[name] = ArrayElement[hash[name]]
-            hash[name].name = name
+        if hash.key? name
+          unless hash[name].is_a?(ArrayElement)
+            hash[name] = ArrayElement.new([hash[name]], name)
           end
           hash[name] << element
         else
@@ -60,12 +59,14 @@ module Saxerator
 
       def block_variable
         return to_s if @text
-        return to_hash if @children.count > 0 || (@attributes.count > 0 && @config.put_attributes_in_hash?)
+        if @children.count > 0 || (@attributes.count > 0 && @config.put_attributes_in_hash?)
+          return to_hash
+        end
         to_empty_element
       end
 
       def normalize_attributes(attributes)
-        Hash[attributes.map {|key, value| [generate_key(key), value] }]
+        Hash[attributes.map { |key, value| [generate_key(key), value] }]
       end
 
       def generate_key(name)