premailer 1.7.1 → 1.7.3

data/lib/premailer/premailer.rb

@@ -1,5 +1,3 @@
-#!/usr/bin/ruby
-#
 # Premailer by Alex Dunae (dunae.ca, e-mail 'code' at the same domain), 2008-10
 #
 # Premailer processes HTML and CSS to improve e-mail deliverability.
@@ -33,11 +31,12 @@ class Premailer
   include HtmlToPlainText
   include CssParser
 
-  VERSION = '1.7.1'
+  VERSION = '1.7.3'
 
   CLIENT_SUPPORT_FILE = File.dirname(__FILE__) + '/../../misc/client_support.yaml'
 
   RE_UNMERGABLE_SELECTORS = /(\:(visited|active|hover|focus|after|before|selection|target|first\-(line|letter))|^\@)/i
+  RE_RESET_SELECTORS = /^(\:\#outlook|body.*|\.ReadMsgBody|\.ExternalClass|img|\#backgroundTable)$/
 
   # list of CSS attributes that can be rendered as HTML attributes
   #
@@ -54,10 +53,33 @@ class Premailer
     'div' => {'text-align' => 'align'},
     'blockquote' => {'text-align' => 'align'},
     'body' => {'background-color' => 'bgcolor'},
-    'table' => {'background-color' => 'bgcolor'},
-    'tr' => {'text-align' => 'align', 'background-color' => 'bgcolor'},
-    'th' => {'text-align' => 'align', 'background-color' => 'bgcolor', 'vertical-align' => 'valign'},
-    'td' => {'text-align' => 'align', 'background-color' => 'bgcolor', 'vertical-align' => 'valign'},
+    'table' => {
+      'background-color' => 'bgcolor',
+      '-premailer-width' => 'width',
+      '-premailer-height' => 'height',
+      '-premailer-cellpadding' => 'cellpadding',
+      '-premailer-cellspacing' => 'cellspacing',
+    },
+    'tr' => {
+      'text-align' => 'align',
+      'background-color' => 'bgcolor',
+      '-premailer-height' => 'height'
+    },
+    'th' => {
+      'text-align' => 'align',
+      'background-color' => 'bgcolor',
+      'vertical-align' => 'valign',
+      '-premailer-width' => 'width',
+      '-premailer-height' => 'height'
+    },
+    'td' => {
+      'text-align' => 'align',
+      'background-color' => 'bgcolor',
+      'vertical-align' => 'valign',
+      '-premailer-width' => 'width',
+      '-premailer-height' => 'height',
+      '-premailer-colspan' => 'colspan'
+    },
     'img' => {'float' => 'align'}
   }
 
@@ -100,13 +122,14 @@ class Premailer
   # [+warn_level+] What level of CSS compatibility warnings to show (see Warnings).
   # [+link_query_string+] A string to append to every <tt>a href=""</tt> link. Do not include the initial <tt>?</tt>.
   # [+base_url+] Used to calculate absolute URLs for local files.
-  # [+css+] Manually specify a CSS stylesheet.
+  # [+css+] Manually specify CSS stylesheets.
   # [+css_to_attributes+] Copy related CSS attributes into HTML attributes (e.g. +background-color+ to +bgcolor+)
   # [+css_string+] Pass CSS as a string
   # [+remove_ids+] Remove ID attributes whenever possible and convert IDs used as anchors to hashed to avoid collisions in webmail programs.  Default is +false+.
   # [+remove_classes+] Remove class attributes. Default is +false+.
   # [+remove_comments+] Remove html comments. Default is +false+.
   # [+preserve_styles+] Whether to preserve any <tt>link rel=stylesheet</tt> and <tt>style</tt> elements.  Default is +false+.
+  # [+preserve_reset+] Whether to preserve styles associated with the MailChimp reset code
   # [+with_html_string+] Whether the +html+ param should be treated as a raw string.
   # [+verbose+] Whether to print errors and warnings to <tt>$stderr</tt>.  Default is +false+.
   # [+adapter+] Which HTML parser to use, either <tt>:nokogiri</tt> or <tt>:hpricot</tt>.  Default is <tt>:hpricot</tt>.
@@ -123,6 +146,7 @@ class Premailer
                 :with_html_string => false,
                 :css_string => nil,
                 :preserve_styles => false,
+                :preserve_reset => true,
                 :verbose => false,
                 :debug => false,
                 :io_exceptions => false,
@@ -131,7 +155,7 @@ class Premailer
     @html_file = html
     @is_local_file = @options[:with_html_string] || Premailer.local_data?(html)
 
-    @css_files = @options[:css]
+    @css_files = [@options[:css]].flatten
 
     @css_warnings = []
 
@@ -207,9 +231,19 @@ protected
   def load_css_from_html! # :nodoc:
     if tags = @doc.search("link[@rel='stylesheet'], style")
       tags.each do |tag|
-        if tag.to_s.strip =~ /^\<link/i and tag.attributes['href'] and media_type_ok?(tag.attributes['media'])
-
-          link_uri = Premailer.resolve_link(tag.attributes['href'].to_s, @html_file)
+        if tag.to_s.strip =~ /^\<link/i && tag.attributes['href'] && media_type_ok?(tag.attributes['media'])
+          # A user might want to <link /> to a local css file that is also mirrored on the site
+          # but the local one is different (e.g. newer) than the live file, premailer will now choose the local file
+          
+          if tag.attributes['href'].to_s.include? @base_url.to_s and @html_file.kind_of?(String)
+            link_uri = File.join(File.dirname(@html_file), tag.attributes['href'].to_s.sub!(@base_url.to_s, ''))
+          end
+          
+          # if the file does not exist locally, try to grab the remote reference
+          if link_uri.nil? or not File.exists?(link_uri)
+            link_uri = Premailer.resolve_link(tag.attributes['href'].to_s, @html_file)
+          end
+          
           if Premailer.local_data?(link_uri)
             $stderr.puts "Loading css from local file: " + link_uri if @options[:verbose]
             load_css_from_local_file!(link_uri)
@@ -239,9 +273,9 @@ public
 
   def media_type_ok?(media_types) # :nodoc:
     return true if media_types.nil? or media_types.empty?
-    return media_types.split(/[\s]+|,/).any? { |media_type| media_type.strip =~ /screen|handheld|all/i }
+    media_types.split(/[\s]+|,/).any? { |media_type| media_type.strip =~ /screen|handheld|all/i }
   rescue
-    return true
+    true
   end
 
   def append_query_string(doc, qs)
@@ -261,7 +295,8 @@ public
     doc.search('a').each do|el|
       href = el.attributes['href'].to_s.strip
       next if href.nil? or href.empty?
-      next if href[0,1] == '#' # don't bother with anchors
+      
+      next if href[0,1] =~ /[\#\{\[\<\%]/ # don't bother with anchors or special-looking links
 
       begin
         href = URI.parse(href)
@@ -294,9 +329,8 @@ public
 
     # Check for an XHTML doctype
   def is_xhtml?
-    intro = @doc.to_s.strip.split("\n")[0..2].join(' ')
-    is_xhtml = (intro =~ /w3c\/\/[\s]*dtd[\s]+xhtml/i)
-    is_xhtml = is_xhtml ? true : false
+    intro = @doc.to_html.strip.split("\n")[0..2].join(' ')
+    is_xhtml = !!(intro =~ /w3c\/\/[\s]*dtd[\s]+xhtml/i)
     $stderr.puts "Is XHTML? #{is_xhtml.inspect}\nChecked:\n#{intro}" if @options[:debug]
     is_xhtml
   end
@@ -322,7 +356,7 @@ public
       tags.each do |tag|
         # skip links that look like they have merge tags
         # and mailto, ftp, etc...
-        if tag.attributes[attribute].to_s =~ /^(\{|\[|<|\#|data:|tel:|file:|sms:|callto:|facetime:|mailto:|ftp:|gopher:)/i
+        if tag.attributes[attribute].to_s =~ /^([\%\<\{\#\[]|data:|tel:|file:|sms:|callto:|facetime:|mailto:|ftp:|gopher:)/i
           next
         end
 
@@ -362,17 +396,16 @@ public
     resolved = nil
     if path =~ /(http[s]?|ftp):\/\//i
       resolved = path
-      return Premailer.canonicalize(resolved)
+      Premailer.canonicalize(resolved)
     elsif base_path.kind_of?(URI)
       resolved = base_path.merge(path)
-      return Premailer.canonicalize(resolved)
+      Premailer.canonicalize(resolved)
     elsif base_path.kind_of?(String) and base_path =~ /^(http[s]?|ftp):\/\//i
       resolved = URI.parse(base_path)
       resolved = resolved.merge(path)
-      return Premailer.canonicalize(resolved)
+      Premailer.canonicalize(resolved)
     else
-
-      return File.expand_path(path, File.dirname(base_path))
+      File.expand_path(path, File.dirname(base_path))
     end
   end
 
@@ -380,13 +413,9 @@ public
   #
   # IO objects return true, as do strings that look like URLs.
   def self.local_data?(data)
-    if data.is_a?(IO) || data.is_a?(StringIO)
-      return true
-    elsif data =~ /^(http|https|ftp)\:\/\//i
-      return false
-    else
-      return true
-    end
+    return true if data.is_a?(IO) || data.is_a?(StringIO)
+    return false if data =~ /^(http|https|ftp)\:\/\//i
+    true
   end
 
   # from http://www.ruby-forum.com/topic/140101
@@ -404,7 +433,7 @@ public
 
   # Check <tt>CLIENT_SUPPORT_FILE</tt> for any CSS warnings
   def check_client_support # :nodoc:
-    @client_support = @client_support ||= YAML::load(File.open(CLIENT_SUPPORT_FILE))
+    @client_support ||= YAML::load(File.open(CLIENT_SUPPORT_FILE))
 
     warnings = []
     properties = []
@@ -450,3 +479,4 @@ public
     return warnings
   end
 end
+

data/local-premailer

@@ -0,0 +1,9 @@
+#!/usr/bin/env ruby
+
+# This binary used only in development environment
+
+require 'rubygems'
+$LOAD_PATH.unshift(File.expand_path('./lib', File.dirname(__FILE__)))
+
+require 'premailer/executor'
+

data/premailer.gemspec

@@ -0,0 +1,22 @@
+Gem::Specification.new do |s|
+  s.name     = "premailer"
+  s.version  = "1.7.3"
+  s.date = Time.now.strftime('%Y-%m-%d')
+  s.summary  = "Preflight for HTML e-mail."
+  s.email    = "code@dunae.ca"
+  s.homepage = "http://premailer.dialect.ca/"
+  s.description = "Improve the rendering of HTML emails by making CSS inline, converting links and warning about unsupported code."
+  s.has_rdoc = true
+  s.author  = "Alex Dunae"
+  s.rdoc_options << '--all' << '--inline-source' << '--line-numbers' << '--charset' << 'utf-8'
+  s.extra_rdoc_files = ["README.rdoc"]
+  s.files            = `git ls-files`.split("\n")
+  s.test_files       = `git ls-files -- {test,spec,features}/*`.split("\n")
+  s.executables      = `git ls-files -- bin/*`.split("\n").map { |f| File.basename(f) }
+  s.add_dependency('css_parser', '>= 1.1.9')
+  s.add_dependency('htmlentities', '>= 4.0.0')
+  s.add_development_dependency('hpricot', '>= 0.8.3')
+  s.add_development_dependency('nokogiri', '>= 1.4.4')
+  s.add_development_dependency('rake', ['~> 0.8',  '!= 0.9.0'])
+  s.add_development_dependency('rdoc', '>= 2.4.2')
+end

data/rakefile.rb

@@ -0,0 +1,69 @@
+$:.unshift File.expand_path('../lib', __FILE__)
+
+require 'rake'
+require 'rake/testtask'
+require 'rdoc/task'
+require 'rubygems/package_task'
+require 'fileutils'
+require 'premailer'
+
+def gemspec
+ @gemspec ||= begin
+   file = File.expand_path('../premailer.gemspec', __FILE__)
+   eval(File.read(file), binding, file)
+ end
+end
+
+Gem::PackageTask.new(gemspec) do |pkg|
+  pkg.need_tar = true
+end
+
+desc 'Default: parse a URL.'
+task :default => [:inline]
+
+desc 'Parse a URL and write out the output.'
+task :inline do
+  url = ENV['url']
+  output = ENV['output']
+  
+  if !url or url.empty? or !output or output.empty?
+    puts 'Usage: rake inline url=http://example.com/ output=output.html'
+    exit
+  end
+
+  premailer = Premailer.new(url, :warn_level => Premailer::Warnings::SAFE, :verbose => true, :adapter => :nokogiri)
+  fout = File.open(output, "w")
+  fout.puts premailer.to_inline_css
+  fout.close
+
+  puts "Succesfully parsed '#{url}' into '#{output}'"
+  puts premailer.warnings.length.to_s + ' CSS warnings were found'
+end
+
+task :text do
+  url = ENV['url']
+  output = ENV['output']
+  
+  if !url or url.empty? or !output or output.empty?
+    puts 'Usage: rake text url=http://example.com/ output=output.txt'
+    exit
+  end
+
+  premailer = Premailer.new(url, :warn_level => Premailer::Warnings::SAFE)
+  fout = File.open(output, "w")
+  fout.puts premailer.to_plain_text
+  fout.close
+  
+  puts "Succesfully parsed '#{url}' into '#{output}'"
+end
+
+Rake::TestTask.new do |t|
+  t.test_files = FileList['test/test_*.rb']
+  t.verbose = false
+end
+
+RDoc::Task.new do |rd|
+  rd.main = "README.rdoc"
+  rd.rdoc_files.include("README.rdoc", "LICENSE.rdoc", "lib/**/*.rb")
+  rd.title = 'Premailer Documentation'
+end

data/test/files/base.html

@@ -0,0 +1,142 @@
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<!--
+
+  You can read this newsletter online at
+  [webversion]
+
+-->
+<html>
+<head>
+	<meta http-equiv="Content-type" content="text/html; charset=utf-8">
+	<title>Premailer Test</title>
+	<link rel="stylesheet" type="text/css" href="styles.css">
+	<style type="text/css">
+		@import "import.css" screen, handheld;
+	</style>
+	<style type="text/css">
+		@import "noimport.css" print;
+	</style>
+	<style type="text/css">
+	@media only screen and (max-device-width: 480px) {
+	  #iphone { display: block; }
+	}
+	</style>
+</head>
+<body>
+<div id="wrapper">
+<p class="hide" id="hide01">This line should be hidden.</p>
+<p class="hide" id="iphone">This is an iPhone style.</p>
+<table width="646" class="container" cellspacing="0" cellpadding="0">
+<tr><td id="webversion" colspan="6">Having trouble reading this newsletter? <webversion>Click here to see it in your browser</webversion></td></tr>
+
+<tr><td height="13" colspan="6" class="frame">&#x00a0;</td></tr>
+
+
+
+<table width="646" class="container" cellspacing="0" cellpadding="0">
+
+<tr>
+	<td class="frame" width="13">&#x00a0;</td>
+	<td class="gutter" width="60">&#x00a0;</td>
+
+	<td class="content" colspan="2" width="500">
+
+<h1><span>Premailer Test</span></h1>
+
+<table width="500" cellpadding="0" cellspacing="0">
+<tr>
+	<td width="20">&#x00a0;</td>
+	<td colspan="2" width="460">
+		<h2>Lorem ipsum dolor</h2>
+		<h3>Suspendisse id velit vitae ligula volutpat condimentum</h3>
+		<p class="dt">Morbi commodo, ipsum sed</p>
+
+
+		<p class="unaligned"><img src="2009-placeholder.png" alt="Image" align="right" class="right">Lorem&nbsp;ipsum&#x00a0;dolor sit amet, consectetuer adipiscing elit. Morbi commodo, ipsum sed pharetra gravida, orci magna rhoncus neque, id pulvinar odio lorem non turpis. Nullam sit amet enim. Suspendisse id velit vitae ligula volutpat condimentum. Aliquam erat volutpat. Sed quis velit. <a href="http://premailer.dialect.ca/">Nulla facilisi</a>. Nulla libero.</p>
+
+		<p attr="another quote">Here&rsquo;s a quote.  Here’s a quote.  &#x201c;Here’s a quote in quotes”.</p>
+
+		<p>Nullam sit amet enim. Suspendisse id velit vitae ligula volutpat condimentum. Aliquam erat volutpat. Sed quis velit. Nulla facilisi.</p>
+		<p>Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Morbi commodo, ipsum sed pharetra gravida, orci magna rhoncus neque, id pulvinar odio lorem non turpis. Nullam sit amet enim. Suspendisse id velit vitae ligula volutpat condimentum. Aliquam erat volutpat. Sed quis velit. Nulla facilisi. Nulla libero.</p>
+		<blockquote><p>“Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Morbi commodo, ipsum sed pharetra gravida, orci magna rhoncus neque, id pulvinar odio lorem non turpis.”</p></blockquote>
+		<p>Aliquam erat volutpat. Sed quis velit. Nulla facilisi. Nulla libero.</p>
+
+		<h3>Link tests</h3>
+		<ul>
+			<li><a id="l01" href="/">Relative path to root</a></li>
+			<li><a id="l02" href="http://premailer.dialect.ca/">Absolute path to root</a></li>
+			<li><a id="l03" href="http://example.com/">Different domain</a></li>
+			<li><a id="l04" href="images/">Relative path to sub-directory</a></li>
+			<li><a id="l05" href="#relative">Link is not converted</a></li>
+			<li><a id="l06" href="http://example.com/test.html?cn=tf&amp;c=20&amp;ord=%%RANDOM%%">Funky ASP URL</a></li>
+			<li><a id="l07" href="?query=string">Appends tracking query string</a></li>
+			<li><a id="l08" href="{DONOTCONVERT}">Link is not converted</a></li>
+			<li><a id="l09" href="[DONOTCONVERT]">Link is not converted</a></li>
+			<li><a id="l10" href="<DONOTCONVERT>">Link is not converted</a></li>
+			<li><a id="l11" href="mailto:premailer@example.com">mailto link</a></li>
+			<li><a id="l12" href="ftp://example.com">FTP link</a></li>
+			<li><a id="l13" href="gopher://gopher.floodgap.com/1/fun/twitpher">Gopher link</a></li>
+		</ul>
+
+
+
+		<p>Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Morbi commodo, ipsum sed pharetra gravida, orci magna rhoncus neque, id pulvinar odio lorem non turpis.</p>
+
+		<p>&nbsp;</p>
+
+
+	</td>
+	<td width="20">&#x00a0;</td>
+</tr>
+</table>
+
+<p class="section"><img src="dots_end.png" alt="---" width="499" height="75"></p>
+	</td><!-- /#content -->
+	<td class="gutter" width="60">&#x00a0;</td>
+
+	<td class="frame" width="13">&#x00a0;</td>
+</tr>
+
+<tr>
+	<td class="frame" width="13">&#x00a0;</td>
+	<td colspan="4" width="620">
+		<table summary="Contact information" cellspacing="0" cellpadding="0" width="620">
+			<tr><td height="4" class="hairline">&#x00a0;</td></tr>
+			<tr><td height="34"class="contact">&#x00a0;</td></tr>
+			<tr>
+
+				<td align="center" class="contact" id="contact_info">
+					<p id="address">Premailer Test<br>
+					<a href="http://dialect.ca/?utm_source=Premailer&utm_medium=Test+Suite&utm_campaign=Premailer">by Dialect</a><br>
+					Vancouver Island, British Columbia<br>
+					250 555.2222</p>
+				</td>
+
+			</tr>
+			<tr><td height="34"class="contact">&#x00a0;</td></tr>
+			<tr><td height="4" class="hairline">&#x00a0;</td></tr>
+		</table>
+	</td>
+	<td class="frame" width="13">&#x00a0;</td>
+</tr>
+
+<tr>
+	<td class="frame" width="13">&#x00a0;</td>
+
+	<td colspan="4" class="content" height="60">&#x00a0;</td>
+	<td class="frame" width="13">&#x00a0;</td>
+</tr>
+
+
+
+
+
+<tr><td height="13" colspan="6" class="frame">&#x00a0;</td></tr>
+<tr><td height="22" colspan="6" >&#x00a0;</td></tr>
+<tr><td id="credit" colspan="6">Newsletter communications by<br><a href="http://dialect.ca/dialogue/?utm_source=Dialogue&utm_medium=Credit&utm_campaign=South+Hollow"><img src="inc/dialect.png" alt="Dialect" width="60" height="35" border="0"></a><br><unsubscribe>Click here to unsubscribe</unsubscribe></td></tr>
+
+</table>
+
+</div>
+</body>
+</html>