risbn 0.2.0 → 0.3.0

data/README.rdoc

@@ -6,21 +6,29 @@ Supports both isbn-10 and isbn-13.
 
 Provides a simple (barebones) tool for extracting isbns from pdf and chm files.
 
+Provides barebones google data integration for retrieving all sorts of information from the isbn.
+
 == Examples:
 
+  require 'isbn'
   isbn = RISBN.parse_first("Some text with and isbn: ISBN-13: 978-0393317732") # => <RISBN isbn="9780393317732">
   isbn.valid? # => true
 
   require 'risbn/scanner'
   RISBN::Scanner.scan("some/file.pdf")
   RISBN::Scanner.scan("some/file.chm")
-  RISBN::Scanner.scan("some/file.tx")
+  RISBN::Scanner.scan("some/file.txt")
 
-== Notes
+  require 'risbn/gdata'
+  RISBN::GData("978-0393317732").data        # <OpenStruct self_url="http://www.google.com/books/feeds/volumes/kom8GwAACAAJ", ...
+  RISBN::GData("978-0393317732").data.rating # => 3.0
 
-Currently only works on unix like platforms.
+  # ...
+
+== Notes
 
-Requires the following tools for scanning files:
+The chm/pdf extraction only works on unix like platforms currently and requires
+the following tools for scanning files:
 
 * Poppler for pdf (pdftotext utility)
 * Archmage for chm

data/Rakefile

@@ -5,13 +5,13 @@ begin
   require 'jeweler'
   Jeweler::Tasks.new do |gem|
     gem.name = "risbn"
-    gem.summary = %Q{some utils to handle isbn numbers}
-    gem.description = %Q{some utils to handle isbn numbers}
+    gem.summary = %Q{Utils to handle isbns}
+    gem.description = %Q{Utils to handle isbns}
     gem.email = "EmmanuelOga@gmail.com"
     gem.homepage = "http://github.com/EmmanuelOga/risbn"
     gem.authors = ["Emmanuel Oga"]
     gem.add_development_dependency "rspec", ">= 1.2.9"
-    # gem is a Gem::Specification... see http://www.rubygems.org/read/chapter/20 for additional settings
+    gem.add_dependency "nokogiri"
   end
   Jeweler::GemcutterTasks.new
 rescue LoadError

data/VERSION

@@ -1 +1 @@
-0.2.0
+0.3.0

data/lib/risbn.rb

@@ -1,5 +1,7 @@
 class RISBN
-  attr_accessor :isbn
+  class InvalidISBN < ArgumentError; end
+
+  attr_reader :isbn
 
   # Gross, but good to catch anything even remotely similar to a ISBN.
   # Later, we can validate to see if it really is a ISBN.
@@ -29,6 +31,10 @@ class RISBN
     end
   end
 
+  def validate!
+    raise InvalidISBN unless valid?
+  end
+
   def to_s
     isbn
   end
@@ -62,3 +68,7 @@ class RISBN
     result == checksum
   end
 end
+
+def RISBN(text)
+  RISBN.parse_first(text)
+end

data/lib/risbn/gdata.rb

@@ -0,0 +1,88 @@
+require 'risbn/nokogiri_utils'
+require 'open-uri'
+require 'ostruct'
+
+class RISBN
+  class GData
+    attr_reader :isbn
+
+    # Provide an RISBN object or a string to be parsed for an ISBN.
+    def initialize(isbn)
+      risbn = isbn.is_a?(RISBN) ? isbn : RISBN(isbn)
+      risbn.validate!
+      @isbn = risbn.to_s
+    end
+
+    # google does not allow to access the data directly by isbn
+    # so a search is needed first to retrieve the actual google id.
+    def search_url
+      "http://books.google.com/books/feeds/volumes?q=ISBN#{ isbn }"
+    end
+
+    # performs a google search by isbn to retrieve the actual google book id.
+    def entry_url
+      @info ||= ( Nokogiri(open(search_url).read) / "entry" / "id" ).inner_text
+    end
+
+    # returns the original book xml from google.
+    def xml
+      @entry ||= open(entry_url).read
+    rescue => e
+      raise RuntimeError, "coult not open url: #{entry_url}  : \n#{e}"
+    end
+
+    # nokogiri nodes from the xml
+    def xml_nodes
+      @xml_nodes ||= Nokogiri::XML(xml)
+    end
+
+    # hash representation of the google xml
+    def to_hash
+      @to_hash ||= RISBN::NokogiriUtils.hash_from_node(xml_nodes)[:entry]
+    end
+
+    # returns an openstruct with a massaged version of the original xml data.
+    def data
+      return @data if @data
+      h, data = to_hash, OpenStruct.new
+
+      data.creator       = h[:creator]
+      data.date          = h[:date]
+      data.description   = h[:description]
+      data.format        = h[:format]
+      data.id            = h[:id]
+      data.identifier    = h[:identifier]
+      data.language      = h[:language]
+      data.publisher     = h[:publisher]
+      data.rating        = h[:rating][:attributes][:average].to_f
+      data.rating_max    = h[:rating][:attributes][:max].to_f
+      data.rating_min    = h[:rating][:attributes][:min].to_f
+      data.subject       = h[:subject].map { |s| s.split(/ [\-\/] /) }.flatten.map(&:strip).uniq.sort
+      data.title         = h[:title].uniq.join(" ")
+      data.updated       = h[:updated]
+      data.category      = h[:category]
+      data.embeddability = h[:embeddability][:value]
+      data.open_access   = h[:openAccess][:value]
+      data.viewability   = h[:viewability][:value]
+
+      h[:link].each do |link|
+        href = link.is_a?(Hash) ? link[:attributes][:href] : link.last[:href]
+        rel  = link.is_a?(Hash) ? link[:attributes][:rel]  : link.last[:rel]
+        data.send("#{rel[/thumbnail|info|annotation|alternate|self/]}_url=", href)
+      end
+
+      @data = data
+    end
+
+  end
+end
+
+class RISBN
+  def gdata
+    @gdata ||= RISBN::GData.new(self)
+  end
+
+  def self.GData(isbn)
+    RISBN::GData.new(isbn)
+  end
+end

data/lib/risbn/nokogiri_utils.rb

@@ -0,0 +1,40 @@
+require 'nokogiri'
+
+class RISBN
+  module NokogiriUtils
+    extend self
+
+    # http://gist.github.com/370755
+    def hash_from_node(node)
+      { node.root.name.to_sym => xml_node_to_hash(node.root) }
+    end
+
+    def xml_node_to_hash(node)
+      return to_value(node.content.to_s) unless node.element?
+
+      result_hash = {}
+
+      node.attributes.each do |key, attr|
+        ( result_hash[:attributes] ||= Hash.new )[attr.name.to_sym] = to_value(attr.value)
+      end
+
+      node.children.each do |child|
+        result = xml_node_to_hash(child)
+
+        if child.name == "text"
+          return to_value(result) unless child.next_sibling || child.previous_sibling
+        else
+          key, val = child.name.to_sym, to_value(result)
+          result_hash[key] = result_hash.key?(key) ? Array(result_hash[key]).push(val) : val
+        end
+      end
+
+      result_hash
+    end
+
+    def to_value(data)
+      data.is_a?(String) && data =~ /^\d+$/ ? data.to_i : data
+    end
+
+  end
+end

data/lib/risbn/scanner.rb

@@ -1,6 +1,5 @@
 require 'shellwords'
 require 'tmpdir'
-require 'iconv'
 
 class RISBN
 

data/lib/risbn/version.rb

@@ -0,0 +1,3 @@
+class RISBN
+  VERSION = "0.3.0"
+end

data/risbn.gemspec

@@ -0,0 +1,64 @@
+# Generated by jeweler
+# DO NOT EDIT THIS FILE DIRECTLY
+# Instead, edit Jeweler::Tasks in Rakefile, and run the gemspec command
+# -*- encoding: utf-8 -*-
+
+Gem::Specification.new do |s|
+  s.name = %q{risbn}
+  s.version = "0.3.0"
+
+  s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
+  s.authors = ["Emmanuel Oga"]
+  s.date = %q{2010-04-19}
+  s.description = %q{Utils to handle isbns}
+  s.email = %q{EmmanuelOga@gmail.com}
+  s.extra_rdoc_files = [
+    "LICENSE",
+     "README.rdoc"
+  ]
+  s.files = [
+    ".document",
+     ".gitignore",
+     "LICENSE",
+     "README.rdoc",
+     "Rakefile",
+     "VERSION",
+     "lib/risbn.rb",
+     "lib/risbn/gdata.rb",
+     "lib/risbn/nokogiri_utils.rb",
+     "lib/risbn/scanner.rb",
+     "lib/risbn/version.rb",
+     "risbn.gemspec",
+     "spec/fixtures/book.xml",
+     "spec/gdata_spec.rb",
+     "spec/risbn_spec.rb",
+     "spec/spec_helper.rb"
+  ]
+  s.homepage = %q{http://github.com/EmmanuelOga/risbn}
+  s.rdoc_options = ["--charset=UTF-8"]
+  s.require_paths = ["lib"]
+  s.rubygems_version = %q{1.3.6}
+  s.summary = %q{Utils to handle isbns}
+  s.test_files = [
+    "spec/spec_helper.rb",
+     "spec/gdata_spec.rb",
+     "spec/risbn_spec.rb"
+  ]
+
+  if s.respond_to? :specification_version then
+    current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
+    s.specification_version = 3
+
+    if Gem::Version.new(Gem::RubyGemsVersion) >= Gem::Version.new('1.2.0') then
+      s.add_development_dependency(%q<rspec>, [">= 1.2.9"])
+      s.add_runtime_dependency(%q<nokogiri>, [">= 0"])
+    else
+      s.add_dependency(%q<rspec>, [">= 1.2.9"])
+      s.add_dependency(%q<nokogiri>, [">= 0"])
+    end
+  else
+    s.add_dependency(%q<rspec>, [">= 1.2.9"])
+    s.add_dependency(%q<nokogiri>, [">= 0"])
+  end
+end
+

data/spec/fixtures/book.xml

@@ -0,0 +1,52 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<entry xmlns="http://www.w3.org/2005/Atom" xmlns:gbs="http://schemas.google.com/books/2008" xmlns:dc="http://purl.org/dc/terms" xmlns:batch="http://schemas.google.com/gdata/batch" xmlns:gd="http://schemas.google.com/g/2005">
+   <id>http://www.google.com/books/feeds/volumes/VOS6IlCsuU4C</id>
+   <updated>2010-04-19T14:35:30.000Z</updated>
+   <category scheme="http://schemas.google.com/g/2005#kind" term="http://schemas.google.com/books/2008#volume"/>
+   <title type="text">JavaScript</title>
+   <link rel="http://schemas.google.com/books/2008/thumbnail" type="image/x-unknown" href="http://bks8.books.google.com/books?id=VOS6IlCsuU4C&amp;printsec=frontcover&amp;img=1&amp;zoom=5&amp;edge=curl&amp;sig=ACfU3U3gH3x5RUB3-HuSK5bnyrSJqYRtlQ&amp;source=gbs_gdata"/>
+   <link rel="http://schemas.google.com/books/2008/info" type="text/html" href="http://books.google.com/books?id=VOS6IlCsuU4C&amp;ie=ISO-8859-1&amp;source=gbs_gdata"/>
+   <link rel="http://schemas.google.com/books/2008/annotation" type="application/atom+xml" href="http://www.google.com/books/feeds/users/me/volumes"/>
+   <link rel="alternate" type="text/html" href="http://books.google.com/books?id=VOS6IlCsuU4C&amp;ie=ISO-8859-1"/>
+   <link rel="self" type="application/atom+xml" href="http://www.google.com/books/feeds/volumes/VOS6IlCsuU4C"/>
+   <gbs:embeddability value="http://schemas.google.com/books/2008#embeddable"/>
+   <gbs:openAccess value="http://schemas.google.com/books/2008#disabled"/>
+   <gd:rating min="1" max="5" average="4.19"/>
+   <gbs:viewability value="http://schemas.google.com/books/2008#view_partial"/>
+   <dc:creator>David Flanagan</dc:creator>
+   <dc:date>2006</dc:date>
+   <dc:description>This Fifth Edition is completely revised and expanded to cover JavaScript as it is used in today's Web 2.0 applications. This book is both an example-driven programmer's guide and a keep-on-your-desk reference, with new chapters that explain everything you need to know to get the most out of JavaScript, including: Scripted HTTP and Ajax XML processing Client-side graphics using the canvas tag Namespaces in JavaScript--essential when writing complex programs Classes, closures, persistence, Flash, and JavaScript embedded in Java applications Part I explains the core JavaScript language in detail. If you are new to JavaScript, it will teach you the language. If you are already a JavaScript programmer, Part I will sharpen your skills and deepen your understanding of the language. Part II explains the scripting environment provided by web browsers, with a focus on DOM scripting with unobtrusive JavaScript. The broad and deep coverage of client-side JavaScript is illustrated with many sophisticated examples that demonstrate how to: Generate a table of contents for an HTML document Display DHTML animations Automate form validation Draw dynamic pie charts Make HTML elements draggable Define keyboard shortcuts for web applications Create Ajax-enabled tool tips Use XPath and XSLT on XML documents loaded with Ajax And much more Part III is a complete reference for core JavaScript. It documents every class, object, constructor, method, function, property, and constant defined by JavaScript 1.5 and ECMAScript Version 3. Part IV is a reference for client-side JavaScript, covering legacy web browser APIs, the standard Level 2 DOM API, and emerging standards such as the XMLHttpRequestobject and the canvas tag. More than 300,000 JavaScript programmers around the world have made this their indispensable reference book for building JavaScript applications. ""A must-have reference for expert JavaScript programmers...well-organized and detailed."" -- Brendan Eich, creator of JavaScript</dc:description>
+   <dc:format>Dimensions 17.8x24.0x5.2 cm</dc:format>
+   <dc:format>1028 pages</dc:format>
+   <dc:format>book</dc:format>
+   <dc:identifier>VOS6IlCsuU4C</dc:identifier>
+   <dc:identifier>ISBN:0596101996</dc:identifier>
+   <dc:identifier>ISBN:9780596101992</dc:identifier>
+   <dc:language>en</dc:language>
+   <dc:publisher>O'Reilly Media, Inc.</dc:publisher>
+   <dc:subject>JavaScript (Computer program language)</dc:subject>
+   <dc:subject>Web servers</dc:subject>
+   <dc:subject>Object-oriented programming (Computer science)</dc:subject>
+   <dc:subject>Web servers</dc:subject>
+   <dc:subject>JavaScript (Computer program language)</dc:subject>
+   <dc:subject>Object-oriented programming (Computer science)</dc:subject>
+   <dc:subject>Ajax (Web site development technology)</dc:subject>
+   <dc:subject>Scripting languages (Computer science)</dc:subject>
+   <dc:subject>Computers / Programming Languages / JavaScript &amp; VBScript</dc:subject>
+   <dc:subject>Computers / General</dc:subject>
+   <dc:subject>Computers / Interactive &amp; Multimedia</dc:subject>
+   <dc:subject>Computers / Internet / General</dc:subject>
+   <dc:subject>Computers / Internet / Browsers</dc:subject>
+   <dc:subject>Computers / Internet / Hardware</dc:subject>
+   <dc:subject>Computers / Internet / Security</dc:subject>
+   <dc:subject>Computers / Internet / Web Page Design</dc:subject>
+   <dc:subject>Computers / Internet / World Wide Web</dc:subject>
+   <dc:subject>Computers / Programming / Software Development</dc:subject>
+   <dc:subject>Computers / Programming Languages / General</dc:subject>
+   <dc:subject>Computers / Programming Languages / HTML</dc:subject>
+   <dc:subject>Computers / Programming Languages / Java</dc:subject>
+   <dc:subject>Computers / Programming Languages / XML</dc:subject>
+   <dc:subject>Computers / Programming Languages / Java</dc:subject>
+   <dc:title>JavaScript</dc:title>
+   <dc:title>the definitive guide</dc:title>
+</entry>

data/spec/gdata_spec.rb

@@ -0,0 +1,38 @@
+require 'spec_helper'
+
+describe RISBN::GData do
+  it "can create a gdata instance from a string" do
+    RISBN::GData("9780596101992").should be_an_instance_of(RISBN::GData)
+  end
+
+  it "can create a gdata instance from a RISBN" do
+    RISBN("9780596101992").gdata.should be_an_instance_of(RISBN::GData)
+    RISBN::GData(RISBN("9780596101992")).should be_an_instance_of(RISBN::GData)
+  end
+
+  it "raises Invalid ISBN if apropriate" do
+    expect { RISBN::GData("6660596101992")        }.to raise_exception(RISBN::InvalidISBN)
+    expect { RISBN("6660596101992").gdata         }.to raise_exception(RISBN::InvalidISBN)
+    expect { RISBN::GData(RISBN("6660596101992")) }.to raise_exception(RISBN::InvalidISBN)
+  end
+
+  context "retrieving gdata" do
+    before do
+      @isbn = RISBN("9780596101992")
+      @isbn.gdata.stub!(:xml).and_return(File.read(GDATA_FIXTURE_PATH))
+    end
+
+    it "returns an OpenStruct" do
+      @isbn.gdata.data.should be_an_instance_of(OpenStruct)
+    end
+
+    it "maps the attributes of the hash form of google response to an struct" do
+      additional_keys = [:open_access, :rating_max, :rating_min, :thumbnail_url,
+        :info_url, :annotation_url, :alternate_url, :self_url]
+
+      ( @isbn.gdata.to_hash.keys.uniq - [:link, :openAccess] + additional_keys).each do |key|
+        @isbn.gdata.data.should respond_to(key)
+      end
+    end
+  end
+end

data/spec/risbn_spec.rb

@@ -1,4 +1,4 @@
-require File.expand_path(File.dirname(__FILE__) + '/spec_helper')
+require 'spec_helper'
 
 describe RISBN do
   it "parses isbn numbers from a line of text" do

data/spec/spec_helper.rb

@@ -1,9 +1,12 @@
 $LOAD_PATH.unshift(File.dirname(__FILE__))
 $LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
+require 'rubygems'
 require 'risbn'
+require 'risbn/gdata'
 require 'spec'
 require 'spec/autorun'
 
+GDATA_FIXTURE_PATH = File.join(File.dirname(__FILE__), 'fixtures', 'book.xml')
+
 Spec::Runner.configure do |config|
-  
 end

metadata

@@ -1,7 +1,12 @@
 --- !ruby/object:Gem::Specification 
 name: risbn
 version: !ruby/object:Gem::Version 
-  version: 0.2.0
+  prerelease: false
+  segments: 
+  - 0
+  - 3
+  - 0
+  version: 0.3.0
 platform: ruby
 authors: 
 - Emmanuel Oga
@@ -9,20 +14,36 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-02-24 00:00:00 -03:00
+date: 2010-04-19 00:00:00 -03:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
   name: rspec
-  type: :development
-  version_requirement: 
-  version_requirements: !ruby/object:Gem::Requirement 
+  prerelease: false
+  requirement: &id001 !ruby/object:Gem::Requirement 
     requirements: 
     - - ">="
       - !ruby/object:Gem::Version 
+        segments: 
+        - 1
+        - 2
+        - 9
         version: 1.2.9
-    version: 
-description: some utils to handle isbn numbers
+  type: :development
+  version_requirements: *id001
+- !ruby/object:Gem::Dependency 
+  name: nokogiri
+  prerelease: false
+  requirement: &id002 !ruby/object:Gem::Requirement 
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        segments: 
+        - 0
+        version: "0"
+  type: :runtime
+  version_requirements: *id002
+description: Utils to handle isbns
 email: EmmanuelOga@gmail.com
 executables: []
 
@@ -39,7 +60,13 @@ files:
 - Rakefile
 - VERSION
 - lib/risbn.rb
+- lib/risbn/gdata.rb
+- lib/risbn/nokogiri_utils.rb
 - lib/risbn/scanner.rb
+- lib/risbn/version.rb
+- risbn.gemspec
+- spec/fixtures/book.xml
+- spec/gdata_spec.rb
 - spec/risbn_spec.rb
 - spec/spec_helper.rb
 has_rdoc: true
@@ -55,21 +82,24 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
+      segments: 
+      - 0
       version: "0"
-  version: 
 required_rubygems_version: !ruby/object:Gem::Requirement 
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
+      segments: 
+      - 0
       version: "0"
-  version: 
 requirements: []
 
 rubyforge_project: 
-rubygems_version: 1.3.5
+rubygems_version: 1.3.6
 signing_key: 
 specification_version: 3
-summary: some utils to handle isbn numbers
+summary: Utils to handle isbns
 test_files: 
 - spec/spec_helper.rb
+- spec/gdata_spec.rb
 - spec/risbn_spec.rb