storexplore 0.3.0 → 0.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 62ed201012136d3411fcb8d36033e956cbd257ee
-  data.tar.gz: 5cabaaaf2165ec4513fa74114114328ac219b019
+  metadata.gz: 10fe4669a7509cd1b7ff06f47fd12fc3269894ff
+  data.tar.gz: 2d3f18b217666c80219922f49de9488fad98b03a
 SHA512:
-  metadata.gz: bc4456ffb7df629729938ddd041491682758f8d14c1d31f396e7d5a4971da9c9d989c0a8d9d84bd8069b7a8a0b29c9acbc9c99cd8b97ec1935adfbaa216f9536
-  data.tar.gz: b0cf18167eb27abee8ff815dae79e10f82904c085dec9ceb4ca1e3b99a42f01226f0c84f638b364bfac2588f077ae7fb9f29a9a6d702052279493225ed3cd71d
+  metadata.gz: 39223a9ac42ce39c6e35394723464b87a192235f92b9d1e398022426b92a5c8a93a0e1ea955003d1cef71edc9f117f7bfdc7a1062c729ee9de2f1be6cf04e8a1
+  data.tar.gz: 93e31befe06656e2ff8f71590b78b06780eaba677b963bc86adff83bc963480502a7441576dffda5a48a3453742f3f6ddd77dd96710696af3ab179f8fb81e240

data/lib/storexplore/api.rb

@@ -19,6 +19,8 @@
 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
 # MA 02110-1301  USA
 
+require 'mechanize'
+
 module Storexplore
 
   # Main entry point to the library
@@ -37,7 +39,14 @@ module Storexplore
     # included in the url of the store.
     # Returns a Storexplore::Walker for the home page of the store
     def self.browse(store_url)
-      builder(store_url).new_walker(WalkerPage.open(store_url))
+      agent = Mechanize.new do |it|
+          # NOTE: by default Mechanize has infinite history, and causes memory leaks
+          #it.history.max_size = 0
+        end
+
+      builder = builder(store_url)
+      builder.configure_agent(agent)
+      builder.new_walker(WalkerPage.open(agent,store_url))
     end
 
     # Forgets the previously defined store API by its name. Mainly useful while

data/lib/storexplore/dsl.rb

@@ -36,11 +36,20 @@ module Storexplore
     # Initializes a new instance with no special categories, items or
     # attributes definition. (Internal usage)
     def initialize
+      @configure_agent_block = lambda do |_| {} end
       @scrap_attributes_block = lambda do |_| {} end
       @categories_digger = NullDigger.new
       @items_digger = NullDigger.new
     end
 
+    # Registers a block that can customize the Mechanize Agent that will be
+    # used throughout the the store digging. This can be useful to setup
+    # custom cookies for example. Ignored anywhere except on the top level
+    # store definition.
+    def agent(&block)
+      @configure_agent_block = block
+    end
+
     # Registers the block to be used to extract attributes from a store page.
     # Block will be evaluated within the context of a Storexplore::WalkerPage
     def attributes(&block)
@@ -60,6 +69,13 @@ module Storexplore
       @items_digger = Digger.new(selector, Dsl.walker_builder(&block))
     end
 
+    # Initializes the mechanize agent with the given setup
+    # * agent : the mechanize agent that will be used throughout the walking
+    # (Internal usage)
+    def configure_agent(agent)
+      @configure_agent_block.call(agent)
+    end
+
     # Initializes a new Storexplore::Walker instance based on specified custom
     # definitions from the instance.
     # * page_getter : proxy to the page that we want to explore

data/lib/storexplore/version.rb

@@ -20,5 +20,5 @@
 # MA 02110-1301  USA
 
 module Storexplore
-  VERSION = "0.3.0"
+  VERSION = "0.4.0"
 end

data/lib/storexplore/walker_page.rb

@@ -19,8 +19,6 @@
 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
 # MA 02110-1301  USA
 
-require 'mechanize'
-
 module Storexplore
 
   # Wrapper around Mechanize::Page providing strict one liners to select
@@ -30,8 +28,8 @@ module Storexplore
 
     # A new lazy proxy on the page.
     # (Internal Usage)
-    def self.open(uri)
-      Getter.new(uri)
+    def self.open(agent, uri)
+      Getter.new(agent, uri)
     end
 
     # Uri of the page
@@ -99,7 +97,8 @@ module Storexplore
     class Getter
       attr_reader :uri
 
-      def initialize(uri)
+      def initialize(agent, uri)
+        @agent = agent
         @uri = uri
       end
       def get
@@ -113,12 +112,7 @@ module Storexplore
       private
 
       def get_page
-        agent = Mechanize.new do |it|
-          # NOTE: by default Mechanize has infinite history, and causes memory leaks
-          it.history.max_size = 0
-        end
-
-        WalkerPage.new(agent.get(@uri))
+        WalkerPage.new(@agent.get(@uri))
       end
     end
 

data/spec/lib/storexplore/dsl_spec.rb

@@ -23,6 +23,12 @@ require "spec_helper"
 
 module Storexplore
 
+  class DslSpecAgent
+    class << self
+      attr_accessor :instance
+    end
+  end
+
   describe Dsl do
 
     def browse
@@ -34,6 +40,9 @@ module Storexplore
     end
 
     context 'a simple store' do
+
+      attr_accessor :actual_agent
+
       before :each do
         FakeWeb.register_uri(:get, "http://www.cats-surplus.com", content_type: 'text/html', body: <<-eos)
           <html>
@@ -72,6 +81,10 @@ module Storexplore
         FakeWeb.register_uri(:get, "http://www.cats-surplus.com/legal.html", content_type: 'text/html', body: "")
 
         Storexplore::Api.define 'cats' do
+          agent do |it|
+            it.history.max_size = 1
+            DslSpecAgent.instance = it
+          end
           items 'a.item' do
             attributes do
               { page: page,
@@ -141,6 +154,12 @@ module Storexplore
         expect(@walker.items.first.attributes[:page]).to be_instance_of(WalkerPage)
       end
 
+      it "enables customization of the mechanize agent" do
+        @walker.items
+
+        expect(DslSpecAgent.instance.page.uri).to eq URI("http://www.cats-surplus.com")
+      end
+
 
       context "when troubleshooting" do
 
@@ -211,5 +230,6 @@ module Storexplore
         expect(@walker.attributes).to be_empty
       end
     end
+
   end
 end

data/spec/lib/storexplore/walker_page_spec.rb

@@ -20,6 +20,7 @@
 # MA 02110-1301  USA
 
 require 'spec_helper'
+require 'mechanize'
 
 module Storexplore
 
@@ -27,12 +28,12 @@ module Storexplore
 
     before :each do
       @uri = URI.parse("file://" + File.expand_path(File.join(File.dirname(__FILE__), 'store_walker_page_spec_fixture.html')))
-      @page_getter = WalkerPage.open(@uri)
+      @page_getter = WalkerPage.open(Mechanize.new, @uri)
     end
 
     context "before actually getting the page" do
       it "nothing should throw if the uri is invalid" do
-        expect(lambda { WalkerPage.open("http://impossible.file.name") }).not_to raise_error
+        expect(lambda { WalkerPage.open(Mechanize.new, "http://impossible.file.name") }).not_to raise_error
       end
 
       it "knows the uri of the page" do

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: storexplore
 version: !ruby/object:Gem::Version
-  version: 0.3.0
+  version: 0.4.0
 platform: ruby
 authors:
 - Philou
@@ -29,7 +29,7 @@ cert_chain:
   LKjGZUoe+A3/FKYHWLPZAArCXrXLhpjAfopLBNmOAju0e30ObcjPQDWcj7z5jfJl
   ILiFgo6hdHrp/tFaDv14PiMRm6sZaQ==
   -----END CERTIFICATE-----
-date: 2014-11-04 00:00:00.000000000 Z
+date: 2014-11-05 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: mechanize