webshooter 0.0.3a → 0.0.4a

data/Gemfile.lock

@@ -1,11 +1,13 @@
 PATH
   remote: .
   specs:
-    webshooter (0.0.3a)
+    webshooter (0.0.4a)
+      nokogiri (~> 1.4.4)
 
 GEM
   remote: http://rubygems.org/
   specs:
+    nokogiri (1.4.4)
 
 PLATFORMS
   ruby

data/README.txt

@@ -11,10 +11,10 @@ gem install webshooter --pre
 
 ## From within ruby
 require 'webshooter'
-Webshooter.capture('http://www.jedi.be','jedi.png','1024x768')
+Webshooter.capture('http://www.jedi.be',{ :output => 'jedi.png', :width => '1024' , :height => '768' , :delay => '2')
 
 ## As a commandline tool
-webshooter 'http://www.jedi.be' 'jedi.png' '1024x768'
+webshooter 'http://www.jedi.be' --width=1024 --height=786 --delay=2 --output=jedi.png
 
 # Limitations
 - does not handle redirects currently
@@ -27,6 +27,10 @@ webshooter 'http://www.jedi.be' 'jedi.png' '1024x768'
 This library is a compilation of various parts I found on the web
 The research was done a few years ago, so unfortunatly I don't have all the references anymore
 
+- webkit2png - http://www.paulhammond.org/webkit2png/
+- http://cocoadevblog.com/webkit-screenshots-cocoa-objective-c
+
+
 - Darkroom -  Copyright (c) 2007 Justin Palmer.
   - https://gist.github.com/34824 
   - https://gist.github.com/86435

data/bin/webshooter

@@ -0,0 +1,61 @@
+#!/usr/bin/env ruby
+
+
+require 'rubygems'
+require 'webshooter'
+require 'optparse'
+
+module Webshooter
+  class CLI
+    #Ruby CLI tool - http://rubylearning.com/blog/2011/01/03/how-do-i-make-a-command-line-tool-in-ruby/
+    def self.execute(url,options)
+      webshotr=Webshooter.capture(url,options)
+    end
+  end
+end
+
+options = {}
+parser = OptionParser.new do |opts|
+  opts.banner = "Usage: #$0 [options] URL"
+
+  opts.on('--width=[WIDTH]', Integer, 'Force width of the screenshot') do |v|
+    options[:width] = v
+  end
+
+  opts.on('--delay=[DELAY]', Integer, "Delay") do |v|
+    options[:delay] = v
+  end
+
+  opts.on('--height=[HEIGHT]', Integer, 'Force height of screenshot') do |v|
+    options[:height] = v
+  end
+
+  opts.on('--output=[FILENAME]', String, 'Specify filename for saving') do |v|
+    options[:output] = v
+  end
+
+  opts.on_tail('--help', 'Display this message and exit') do
+    puts opts
+    exit
+  end
+
+  opts.on_tail('-v', "--version", "Show version") do
+    puts "webshooter v"+Webshooter::VERSION
+    exit
+  end
+
+  opts.parse!(ARGV)  
+
+end
+
+if ARGV.size < 1
+  puts "You need to specify the URL as a minimum"
+else
+  uri=ARGV.first
+  if (uri =~ URI::regexp).nil?  
+    puts "You specified an invalid URL"
+  else
+    Webshooter::CLI.execute(uri, options)  
+  end
+
+end

data/lib/webshooter/redirectfollower.rb

@@ -3,6 +3,7 @@
 
 require 'net/http'
 require 'net/https'
+require 'nokogiri'
 
 module Webshooter
   
@@ -27,11 +28,16 @@ class RedirectFollower
       #http.verify_mode = OpenSSL::SSL::VERIFY_NONE # Not setting this explicitly will result in an error and the value being set anyway
     end
 
+
     if (uri.path=="")
-      request = Net::HTTP::Get.new("/")      
+      request = Net::HTTP::Get.new("#{ '/' + (uri.query ? ('?' + uri.query) : '')}")      
     else
-      request = Net::HTTP::Get.new(uri.path)
+    
+      # http://intertwingly.net/blog/2006/08/19/Quack-Squared
+      request = Net::HTTP::Get.new(uri.path + (uri.query ? ('?' + uri.query) : '') )
     end
+    
+    
     self.response = http.start {|http| http.request(request) }
 
     if response.kind_of?(Net::HTTPRedirection)      
@@ -40,11 +46,34 @@ class RedirectFollower
 
       puts "redirect found, headed to #{url}"
       resolve
-    end   
+    end 
+  
+    meta_link=meta_parse
+    if !meta_link.nil?
+      self.url = meta_link
+      self.redirect_limit -= 1
+      puts "metalink found, headed to #{url}"
+      resolve
+    end
+      
     return url
     
   end
 
+  def meta_parse
+    # http://stackoverflow.com/questions/5003367/mechanize-how-to-follow-or-click-meta-refreshes-in-rails/5012684#5012684
+    html=response.body.to_s.downcase
+    #puts html
+    doc = Nokogiri::HTML(html)
+    meta_tag=doc.at('meta[http-equiv="refresh"]')
+    if !meta_tag.nil?
+      meta_link = meta_tag['content'][/url=(.+)/, 1]
+    else  
+      meta_link=nil
+    end
+    meta_link # => "http://www.example.com/"
+  end
+  
   def redirect_url
     if response['location'].nil?
       response.body.match(/<a href=\"([^>]+)\">/i)[1]

data/lib/webshooter/version.rb

@@ -1,3 +1,3 @@
 module Webshooter
-	VERSION = "0.0.3a"
+	VERSION = "0.0.4a"
 end

data/lib/webshooter/webshotprocessor.rb

@@ -41,7 +41,7 @@ class WebShotProcessor
       @webView.preferences.setAutosaves(false)
       # Set some useful options.
       @webView.preferences.setShouldPrintBackgrounds(true)
-      @webView.preferences.setJavaScriptCanOpenWindowsAutomatically(false)
+      @webView.preferences.setJavaScriptCanOpenWindowsAutomatically(true)
       @webView.preferences.setAllowsAnimatedImages(false)
       # Make sure we don't get a scroll bar.
       @webView.mainFrame.frameView.setAllowsScrolling(false)          
@@ -49,6 +49,7 @@ class WebShotProcessor
       # This @delegate will get a message when the load completes
       @delegate = SimpleLoadDelegate.alloc.init
       @delegate.webshooter = self
+
       @webView.setFrameLoadDelegate(@delegate)
 
       # Replace the window's content @webView with the web @webView
@@ -56,15 +57,23 @@ class WebShotProcessor
       @webView.release
     end
 
-    def capture(uri, path, dimensions = "1024x768" )
+    def capture(uri, options )
+      
+      options[:width] ||= 1024
+      options[:height] ||= 768
+      options[:output] ||= "webshot.png" 
+      options[:delay] ||= 2 
       
-      snapshot_dimension=dimensions.split('x')
+      @delegate.options = options
+           
+      snapshot_dimension=[ options[:width] , options[:height]]
       # Tell the frame to load the URL we want
       @webView.window.setContentSize(snapshot_dimension) 
       @webView.setFrameSize(snapshot_dimension)
     
       final_link = RedirectFollower.new(uri).resolve
       
+      #puts "final link = #{final_link}"
       myURI = URI.parse(final_link)
        
       #Allow all https certificates
@@ -73,8 +82,6 @@ class WebShotProcessor
       #puts "Getting ready for the loadRequest"+uri
       @webView.mainFrame.loadRequest(NewNSURLRequest.requestWithURL(OSX::NSURL.URLWithString(final_link)))
 
-      #Wait for some pages to terminate
-      sleep 2
       #
       # Run the main event loop until the frame loads
       @timeout=false
@@ -93,7 +100,6 @@ class WebShotProcessor
         view.window.orderFront(nil)
         #view.window.display
   
-        #puts "-------------------------------------------------"
         #puts "We got success"
         @docview=view.mainFrame.frameView.documentView
 
@@ -125,7 +131,8 @@ class WebShotProcessor
         if view.bounds.size.height < 300000
           view.lockFocus
           bitmap = OSX::NSBitmapImageRep.alloc.initWithFocusedViewRect(view.bounds)
-          bitmap.representationUsingType_properties(OSX::NSPNGFileType, nil).writeToFile_atomically(path, true)
+          bitmap.representationUsingType_properties(OSX::NSPNGFileType, nil).writeToFile_atomically(options[:output], true)
+          logger.info( "Webshot for #{final_link} => '#{options[:output]}' ")
           bitmap.release
           view.unlockFocus
 
@@ -136,7 +143,7 @@ class WebShotProcessor
       end
 
       upon_failure do |error, logger|
-        logger.warn("Unable to load URI: #{uri} (#{error})")
+        logger.warn("Unable to load URI: #{final_link} (#{error})")
       end
       
       
@@ -166,7 +173,7 @@ class WebShotProcessor
 
     class SimpleLoadDelegate < OSX::NSObject
 
-      attr_accessor :webshooter
+      attr_accessor :webshooter, :options
 
       def stopLoop
         mainLoop=OSX.CFRunLoopGetMain
@@ -176,10 +183,15 @@ class WebShotProcessor
       end
 
       def webView_didFinishLoadForFrame(sender, frame)
+  
         #This did the trick, we have to wait for the right frame to load, not other frames
         if (frame == sender.mainFrame)
           then
           #puts "Finish Load For Frame"
+          #sleep 10
+          #puts "#{  @options[:delay]}"
+          sleep   @options[:delay]
+          #puts "we got a finish"
           @webshooter.load_success = true; 
 
           stopLoop
@@ -189,8 +201,15 @@ class WebShotProcessor
         end
       end
 
+      # keeping track of all content being loaded
+      # http://www.opensubscriber.com/message/webkitsdk-dev@lists.apple.com/2978556.html
+      def webView_didCommitLoadForFrame(sender,frame)
+        #puts "we got a commit"
+      end
+      
       def webView_didFailLoadWithError_forFrame(webview, load_error, frame)
 
+        #puts "we got a failed"
         #This is trick # 2
         #We have to catch this stupid error
         if (load_error.code == OSX::NSURLErrorCancelled) 
@@ -208,6 +227,7 @@ class WebShotProcessor
       end
 
       def webView_didFailProvisionalLoadWithError_forFrame(webview, load_error, frame)
+        puts "we got a provisional load"
         if (load_error.code == OSX::NSURLErrorCancelled) 
           then
           #pp "WARN: did Fail PROVISIONAL LOAD WITH ERROR For Frame"

data/lib/webshooter.rb

@@ -5,9 +5,9 @@ require 'webshooter/webshotprocessor'
 
 module Webshooter
 	class Webshooter
-	  def self.capture(uri, path, dimensions = "1024x768" )
+	  def self.capture(uri, options )
 	    webProcessor=WebShotProcessor.new
-	    webProcessor.capture(uri,path,dimensions)
+	    webProcessor.capture(uri,options)
 	  end
 	end
 end

data/webshooter.gemspec

@@ -16,6 +16,7 @@ Gem::Specification.new do |s|
 
   s.add_development_dependency "bundler", ">= 1.0.0"
   #s.add_dependency "responsalizr", "~>1.0.2"
+  s.add_dependency "nokogiri", "~>1.4.4"
 
   s.files        = `git ls-files`.split("\n")
   s.executables  = `git ls-files`.split("\n").map{|f| f =~ /^bin\/(.*)/ ? $1 : nil}.compact

metadata

@@ -1,12 +1,14 @@
 --- !ruby/object:Gem::Specification 
 name: webshooter
 version: !ruby/object:Gem::Version 
-  prerelease: true
+  hash: 58
+  prerelease: 5
   segments: 
   - 0
   - 0
-  - 3a
-  version: 0.0.3a
+  - 4
+  - a
+  version: 0.0.4a
 platform: ruby
 authors: 
 - Patrick Debois
@@ -18,24 +20,42 @@ date: 2011-02-20 00:00:00 +01:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
-  version_requirements: &id001 !ruby/object:Gem::Requirement 
+  prerelease: false
+  type: :development
+  requirement: &id001 !ruby/object:Gem::Requirement 
+    none: false
     requirements: 
     - - ">="
       - !ruby/object:Gem::Version 
+        hash: 23
         segments: 
         - 1
         - 0
         - 0
         version: 1.0.0
-  requirement: *id001
   name: bundler
+  version_requirements: *id001
+- !ruby/object:Gem::Dependency 
   prerelease: false
-  type: :development
+  type: :runtime
+  requirement: &id002 !ruby/object:Gem::Requirement 
+    none: false
+    requirements: 
+    - - ~>
+      - !ruby/object:Gem::Version 
+        hash: 15
+        segments: 
+        - 1
+        - 4
+        - 4
+        version: 1.4.4
+  name: nokogiri
+  version_requirements: *id002
 description: This library allows you to create webshots using webkit on MacOSX. A webshot is a screenshot taken inside the browser. The advantage of this library is that it is headless and gives you the real view not a parsed view
 email: 
 - patrick.debois@jedi.be
 executables: 
-- webshooter.rb
+- webshooter
 extensions: []
 
 extra_rdoc_files: []
@@ -46,7 +66,7 @@ files:
 - Gemfile.lock
 - README.txt
 - Rakefile
-- bin/webshooter.rb
+- bin/webshooter
 - lib/webshooter.rb
 - lib/webshooter/newnsurlrequest.rb
 - lib/webshooter/redirectfollower.rb
@@ -68,16 +88,20 @@ rdoc_options: []
 require_paths: 
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement 
+  none: false
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
+      hash: 3
       segments: 
       - 0
       version: "0"
 required_rubygems_version: !ruby/object:Gem::Requirement 
+  none: false
   requirements: 
   - - ">="
     - !ruby/object:Gem::Version 
+      hash: 23
       segments: 
       - 1
       - 3
@@ -86,7 +110,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
 requirements: []
 
 rubyforge_project: webshooter
-rubygems_version: 1.3.6
+rubygems_version: 1.5.2
 signing_key: 
 specification_version: 3
 summary: Create webshot using webkit on MacOSX

data/bin/webshooter.rb

@@ -1,19 +0,0 @@
-#!/usr/bin/env ruby
-
-
-require 'rubygems'
-require 'webshooter'
-
-module Webshooter
-	class CLI
-		#Ruby CLI tool - http://rubylearning.com/blog/2011/01/03/how-do-i-make-a-command-line-tool-in-ruby/
-		def self.execute(args)
-			url=args[0]
-			filename=args[1]
-			size=args[2]
-			webshotr=Webshooter.capture(url, filename, size )
-		end
-	end
-end
-
-Webshooter::CLI.execute(ARGV)