royw-imdb 0.0.14 → 0.0.15

data/README

@@ -5,7 +5,6 @@ ImdbMovie Jet Pilot
 - should convert to yaml
 
 ImdbMovie Indiana Jones and the Last Crusade
-- should query IMDB url
 - should get the title
 - should get director(s)
 - should get the poster url
@@ -30,7 +29,6 @@ ImdbMovie Indiana Jones and the Last Crusade
 - should be able to convert to and then from xml
 
 ImdbMovie Han robado una estrella
-- should query IMDB url
 - should get the title
 - should get director(s)
 - should not get the poster
@@ -46,9 +44,6 @@ ImdbMovie Han robado una estrella
 - should get the company
 - should not get any photos
 
-ImdbSearch search that returns multiple movies
-- should query IMDB url
-
 ImdbSearch search that returns multiple movies movies
 - should be a collection of ImdbMovie instances
 - should include 'Indiana Jones and the Last Crusade'
@@ -80,7 +75,6 @@ ImdbSearch searches that match on AKA title "Meltdown" movies
 - should have only one movie from 1995
 
 ImdbMovie Indiana Jones and the Last Crusade
-- should query IMDB url
 - should get the image
 
 String unescape_html
@@ -90,6 +84,6 @@ String unescape_html
 String strip_tags
 - should strip HTML tags
 
-Finished in 4.131553 seconds
+Finished in 3.696984 seconds
 
-65 examples, 0 failures
+61 examples, 0 failures

data/lib/imdb/imdb_image.rb

@@ -1,19 +1,46 @@
 # @imdb_movie.poster.should == 'http://ia.media-imdb.com/images/M/MV5BMTkzODA5ODYwOV5BMl5BanBnXkFtZTcwMjAyNDYyMQ@@._V1._SX216_SY316_.jpg'
 
 class ImdbImage
-  
+
   attr_accessor :url
-  
+
   def initialize(url)
     @url = File.join("http://www.imdb.com/", url)
   end
-  
+
   def image
     document.at("table#principal tr td img")['src'] rescue nil
   end
-  
+
   def document
-    @document ||= Hpricot(open(self.url).read)
+    @document ||= Hpricot(fetch(self.url))
   end
-  
+
+  private
+
+  MAX_ATTEMPTS = 3
+  SECONDS_BETWEEN_RETRIES = 1.0
+
+  def fetch(page)
+    doc = nil
+    attempts = 0
+    begin
+      doc = read_page(page)
+    rescue Exception => e
+      attempts += 1
+      if attempts > MAX_ATTEMPTS
+        raise
+      else
+        sleep SECONDS_BETWEEN_RETRIES
+        retry
+      end
+    end
+    doc
+  end
+
+  def read_page(page)
+    puts "ImdbImage::read_page"
+    open(page).read
+  end
+
 end

data/lib/imdb/imdb_movie.rb

@@ -210,27 +210,23 @@ class ImdbMovie
 #     #document.at("div#tn15title h1").innerHTML.split('<span>').first.unescape_html rescue nil
 #   end
 
+  # Fetch the document with retry to handle the occasional glitches
+  def document
+    if @document.nil?
+      html = fetch(self.url)
+      @document = Hpricot(html)
+    end
+    @document
+  end
+
   MAX_ATTEMPTS = 3
   SECONDS_BETWEEN_RETRIES = 1.0
 
-  # Fetch the document with retry to handle the occasional glitches
-  def document
+  def fetch(page)
+    doc = nil
     attempts = 0
     begin
-      if @document.nil?
-        if ImdbMovie::use_html_cache
-          begin
-            filespec = self.url.gsub(/^http:\//, 'spec/samples').gsub(/\/$/, '.html')
-            html = open(filespec).read
-          rescue Exception
-            html = open(self.url).read
-            cache_html_files(html)
-          end
-        else
-          html = open(self.url).read
-        end
-        @document = Hpricot(html)
-      end
+      doc = read_page(page)
     rescue Exception => e
       attempts += 1
       if attempts > MAX_ATTEMPTS
@@ -240,21 +236,12 @@ class ImdbMovie
         retry
       end
     end
-    @document
+    doc
   end
 
-  # this is used to save imdb pages so they may be used by rspec
-  def cache_html_files(html)
-    begin
-      filespec = self.url.gsub(/^http:\//, 'spec/samples').gsub(/\/$/, '.html')
-      unless File.exist?(filespec)
-        puts "caching #{filespec}"
-        File.mkdirs(File.dirname(filespec))
-        File.open(filespec, 'w') { |f| f.puts html }
-      end
-    rescue Exception => eMsg
-      puts eMsg.to_s
-    end
+  def read_page(page)
+    puts "ImdbMovie::read_page"
+    open(page).read
   end
 
 end

data/lib/imdb/imdb_search.rb

@@ -86,7 +86,32 @@ class ImdbSearch
 
   def document
     filespec = "http://www.imdb.com/find?q=#{CGI::escape(@query)};s=tt"
-    @document ||= Hpricot(open(filespec).read)
+    @document ||= Hpricot(fetch(filespec))
+  end
+
+  MAX_ATTEMPTS = 3
+  SECONDS_BETWEEN_RETRIES = 1.0
+
+  def fetch(page)
+    doc = nil
+    attempts = 0
+    begin
+      doc = read_page(page)
+    rescue Exception => e
+      attempts += 1
+      if attempts > MAX_ATTEMPTS
+        raise
+      else
+        sleep SECONDS_BETWEEN_RETRIES
+        retry
+      end
+    end
+    doc
+  end
+
+  def read_page(page)
+    puts "ImdbSearch::read_page"
+    open(page).read
   end
 
   def parse_movies_from_document

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: royw-imdb
 version: !ruby/object:Gem::Version 
-  version: 0.0.14
+  version: 0.0.15
 platform: ruby
 authors: 
 - Sergio Gil