onix 0.8.5 → 0.9.0

data/CHANGELOG

@@ -1,6 +1,12 @@
-v0.8.5 (21st December 2010)
+v0.9.0 (14th April 2011)
 - switch back to the vanilla roxml gem. Ben is maintaining it again and
   he has merged in my bug fixes
+- clarify comments explaining encoding behaviour
+- Add options hash to ONIX::Reader. Only option at this stage is :encoding,
+  which allows the user to override the assumed encoding of the input XML
+- API change, so new minor version
+
+v0.8.5 (21st December 2010)
 - update packaging - use bundler and rspec 2.x
 - support normalising short tag files that include HTML tags
 

data/README.markdown

@@ -21,6 +21,10 @@ ONIX::Normaliser to convert any short tag files to reference tags.
 
 ONIX::Writer only generates reference tag ONIX files.
 
+It baffles me why anyone thought designing two parallel versions of the ONIX
+spec was a good idea. Use reference tags my friends, and let short tags fade
+away into irrelevant obscurity.
+
 ## DTD Loading
 
 To correctly handle named entities when reading an ONIX file, this gem attempts

data/lib/onix.rb

@@ -9,8 +9,8 @@ require 'andand'
 module ONIX
   module Version #:nodoc:
     Major = 0
-    Minor = 8
-    Tiny  = 5
+    Minor = 9
+    Tiny  = 0
 
     String = [Major, Minor, Tiny].join('.')
   end

data/lib/onix/reader.rb

@@ -25,7 +25,7 @@ module ONIX
   # in a shim that provides simple accessor access to common attributes, pass the
   # shim class as a second argument
   #
-  #   reader = ONIX::Reader.new("somefile.xml", ONIX::APAProduct)
+  #   reader = ONIX::Reader.new("somefile.xml", :product_class => ONIX::APAProduct)
   #
   #   puts reader.header.inspect
   #
@@ -39,7 +39,7 @@ module ONIX
   # As well as accessing the file header, there are handful of other read only
   # attributes that might be useful
   #
-  #   reader = ONIX::Reader.new("somefile.xml", ONIX::APAProduct)
+  #   reader = ONIX::Reader.new("somefile.xml")
   #
   #   puts reader.version
   #   puts reader.xml_lang
@@ -50,23 +50,45 @@ module ONIX
   # ONIX spec, and you may need to handle the file differently based on what
   # version it is.
   #
+  # == File Encoding
+  #
+  # ONIX::Reader returns all strings as UTF-8. Source file encoding is detected by
+  # the encoding declaration at the top of the file, like so:
+  #
+  #   <?xml version="1.0" encoding="iso-8859-1"?>
+  #
+  # If the encoding declaration is missing the file is assumed to be UTF-8.
+  #
+  # If the encoding declaration is missing or wrong and the file isn't UTF-8,
+  # you can manually set or override it like so:
+  #
+  #   reader = ONIX::Reader.new("somefile.xml", :encoding => "iso-8859-1")
+  #
+  # If the file contains invalid bytes for the source encoding an exception will
+  # be raised. This isn't ideal, but I'm still looking for ways to make this
+  # behaviour configurable.
+  #
   class Reader
     include Enumerable
 
     attr_reader :header, :release
 
-    def initialize(input, product_klass = ::ONIX::Product)
+    def initialize(input, *args)
+      opts = args.last.kind_of?(Hash) ? args.pop : {}
+      if args.size > 0
+        ActiveSupport::Deprecation.warn("Passing a klass as ONIX::Reader's second argument is deprecated, use the :product_class option instead", caller)
+      end
+      @product_klass = opts[:product_class] || args.pop || ::ONIX::Product
+
       if input.kind_of?(String)
         @file   = File.open(input, "r")
-        @reader = Nokogiri::XML::Reader(@file) { |cfg| cfg.dtdload.noent }
+        @reader = Nokogiri::XML::Reader(@file, nil, opts[:encoding]) { |cfg| cfg.dtdload.noent }
       elsif input.kind_of?(IO)
-        @reader = Nokogiri::XML::Reader(input) { |cfg| cfg.dtdload.noent }
+        @reader = Nokogiri::XML::Reader(input, nil, opts[:encoding]) { |cfg| cfg.dtdload.noent }
       else
         raise ArgumentError, "Unable to read from file or IO stream"
       end
 
-      @product_klass = product_klass
-
       @release = find_release
       @header = find_header
 

data/spec/reader_spec.rb

@@ -12,6 +12,7 @@ describe ONIX::Reader do
     @entity_file = File.join(@data_path, "entities.xml")
     @utf_16_file = File.join(@data_path, "utf_16.xml")
     @iso_8859_1_file = File.join(@data_path, "iso_8859_1.xml")
+    @no_encoding_decl_file = File.join(@data_path, "aau.xml")
   end
 
   it "should initialize with a filename" do
@@ -90,19 +91,36 @@ describe ONIX::Reader do
 
   it "should transparently convert a iso-8859-1 file to utf-8" do
     reader = ONIX::Reader.new(@iso_8859_1_file)
-    product = nil
-    reader.each do |p|
-      product = p
-    end
+    reader.each do |product|
+      if RUBY_VERSION >= "1.9"
+        utf8 = Encoding.find("utf-8")
+        product.contributors[0].person_name_inverted.encoding.should eql(utf8)
+      end
 
-    # ROXML appears to munge the string encodings
-    if RUBY_VERSION >= "1.9"
-      utf8 = Encoding.find("utf-8")
-      product.contributors[0].person_name_inverted.encoding.should eql(utf8)
+      product.contributors[0].person_name_inverted.should eql("Küng, Hans")
     end
+  end
 
-    product.contributors[0].person_name_inverted.should eql("Küng, Hans")
+  # This isn't ideal behaviour, but i'm somewhat hamstrung by nokogiri API. It'd
+  # be nice to have the option to replace unrecognised bytes with a valid char.
+  it "should raise an exception when an iso-8859-1 file isn't declared as such" do
+    reader = ONIX::Reader.new(@no_encoding_decl_file)
+    lambda {
+      reader.each do |product|
+      end
+    }.should raise_error(Nokogiri::XML::SyntaxError)
+  end
+
+  it "should transparently convert an iso-8859-1 file to utf-8 when there's no declaration but the user manually specifies iso-8859-1" do
+    reader = ONIX::Reader.new(@no_encoding_decl_file, :encoding => "iso-8859-1")
+    reader.each do |product|
+      if RUBY_VERSION >= "1.9"
+        utf8 = Encoding.find("utf-8")
+        product.contributors[0].person_name_inverted.encoding.should eql(utf8)
+      end
 
+      product.contributors[0].person_name_inverted.should eql("Melo,Patr¡cia")
+    end
   end
 
   it "should transparently convert a utf-16 file to utf-8" do
@@ -119,6 +137,19 @@ describe ONIX::Reader do
     end
 
     product.contributors[0].person_name_inverted.should eql("Küng, Hans")
+  end
 
+  it "should support returning an APAProduct using deprecated API" do
+    reader = ONIX::Reader.new(@file1, ONIX::APAProduct)
+    reader.each do |product|
+      product.should be_a_kind_of(ONIX::APAProduct)
+    end
+  end
+
+  it "should support returning an APAProduct using new API" do
+    reader = ONIX::Reader.new(@file1, :product_class => ONIX::APAProduct)
+    reader.each do |product|
+      product.should be_a_kind_of(ONIX::APAProduct)
+    end
   end
 end

metadata

@@ -1,13 +1,12 @@
 --- !ruby/object:Gem::Specification 
 name: onix
 version: !ruby/object:Gem::Version 
-  hash: 53
   prerelease: false
   segments: 
   - 0
-  - 8
-  - 5
-  version: 0.8.5
+  - 9
+  - 0
+  version: 0.9.0
 platform: ruby
 authors: 
 - James Healy
@@ -15,18 +14,17 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-12-21 00:00:00 +11:00
+date: 2011-04-14 00:00:00 +10:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
-  name: yob-roxml
+  name: roxml
   prerelease: false
   requirement: &id001 !ruby/object:Gem::Requirement 
     none: false
     requirements: 
-    - - ">="
+    - - ~>
       - !ruby/object:Gem::Version 
-        hash: 15
         segments: 
         - 3
         - 1
@@ -35,77 +33,87 @@ dependencies:
   type: :runtime
   version_requirements: *id001
 - !ruby/object:Gem::Dependency 
-  name: i18n
+  name: activesupport
   prerelease: false
   requirement: &id002 !ruby/object:Gem::Requirement 
     none: false
     requirements: 
-    - - ">="
+    - - ~>
       - !ruby/object:Gem::Version 
-        hash: 3
         segments: 
+        - 3
         - 0
-        version: "0"
+        - 5
+        version: 3.0.5
   type: :runtime
   version_requirements: *id002
 - !ruby/object:Gem::Dependency 
-  name: andand
+  name: i18n
   prerelease: false
   requirement: &id003 !ruby/object:Gem::Requirement 
     none: false
     requirements: 
     - - ">="
       - !ruby/object:Gem::Version 
-        hash: 3
         segments: 
         - 0
         version: "0"
   type: :runtime
   version_requirements: *id003
 - !ruby/object:Gem::Dependency 
-  name: nokogiri
+  name: andand
   prerelease: false
   requirement: &id004 !ruby/object:Gem::Requirement 
     none: false
     requirements: 
     - - ">="
       - !ruby/object:Gem::Version 
-        hash: 7
+        segments: 
+        - 0
+        version: "0"
+  type: :runtime
+  version_requirements: *id004
+- !ruby/object:Gem::Dependency 
+  name: nokogiri
+  prerelease: false
+  requirement: &id005 !ruby/object:Gem::Requirement 
+    none: false
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
         segments: 
         - 1
         - 4
         version: "1.4"
   type: :runtime
-  version_requirements: *id004
+  version_requirements: *id005
 - !ruby/object:Gem::Dependency 
   name: rake
   prerelease: false
-  requirement: &id005 !ruby/object:Gem::Requirement 
+  requirement: &id006 !ruby/object:Gem::Requirement 
     none: false
     requirements: 
     - - ">="
       - !ruby/object:Gem::Version 
-        hash: 3
         segments: 
         - 0
         version: "0"
   type: :development
-  version_requirements: *id005
+  version_requirements: *id006
 - !ruby/object:Gem::Dependency 
   name: rspec
   prerelease: false
-  requirement: &id006 !ruby/object:Gem::Requirement 
+  requirement: &id007 !ruby/object:Gem::Requirement 
     none: false
     requirements: 
     - - ~>
       - !ruby/object:Gem::Version 
-        hash: 1
         segments: 
         - 2
         - 1
         version: "2.1"
   type: :development
-  version_requirements: *id006
+  version_requirements: *id007
 description: A convient mapping between ruby objects and the ONIX XML specification
 email: 
 - jimmy@deefa.com
@@ -367,7 +375,6 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
-      hash: 3
       segments: 
       - 0
       version: "0"
@@ -376,7 +383,6 @@ required_rubygems_version: !ruby/object:Gem::Requirement
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
-      hash: 3
       segments: 
       - 0
       version: "0"