premailer 1.5.2 → 1.5.3

data/CHANGELOG.rdoc

@@ -1,5 +1,8 @@
 = Premailer CHANGELOG
 
+== Version 1.5.3
+ * improved plaintext conversion
+
 == Version 1.5.2
  * released to GitHub
  * fixed handling of mailto links

data/README.rdoc

@@ -7,9 +7,9 @@ huge pain and a simple newsletter becomes un-managable very quickly. This
 script is my solution.
 
 * CSS styles are converted to inline style attributes
-  Checks style and link[rel=stylesheet] tags and preserves existing inline attributes
+  Checks <tt>style</tt> and <tt>link[rel=stylesheet]</tt> tags and preserves existing inline attributes
 * Relative paths are converted to absolute paths
-  Checks links in href, src and CSS url('')
+  Checks links in <tt>href</tt>, <tt>src</tt> and CSS <tt>url('')</tt>
 * CSS properties are checked against e-mail client capabilities
   Based on the Email Standards Project's guides
 * A plain text version is created
@@ -59,7 +59,7 @@ A few areas that are particularly in need of love:
 
 Premailer is written in Ruby.  
 
-The web interface can be found at http://premailer.dialect.ca/ .
+The web interface can be found at http://premailer.dialect.ca .
 
 The source code can be found at http://github.com/alexdunae/premailer .
 

data/init.rb

@@ -0,0 +1 @@
+require 'premailer'

data/lib/premailer/html_to_plain_text.rb

@@ -8,51 +8,67 @@ module HtmlToPlainText
   #
   # TODO:
   #  - add support for DL, OL
-  def convert_to_text(html, line_length, from_charset = 'UTF-8')
+  def convert_to_text(html, line_length = 65, from_charset = 'UTF-8')
     r = Text::Reform.new(:trim => true, 
                          :squeeze => false, 
                          :break => Text::Reform.break_wrap)
 
     txt = html
     
-    he = HTMLEntities.new                                 # decode HTML entities
-
+    # decode HTML entities
+    he = HTMLEntities.new
     txt = he.decode(txt)
 
-    txt.gsub!(/<h([0-9]+)[^>]*>(.*)<\/h[0-9]+>/i) do |s|  # handle headings
+    # handle headings (H1-H6)
+    txt.gsub!(/[ \t]*<h([0-9]+)[^>]*>(.*)<\/h[0-9]+>/i) do |s|
       hlevel = $1.to_i
-      htext = $2.gsub(/<\/?[^>]*>/i, '')                  # remove tags inside headings
+      # cleanup text inside of headings
+      htext = $2.gsub(/<\/?[^>]*>/i, '').strip
       hlength = (htext.length > line_length ? 
                   line_length : 
                   htext.length)
 
       case hlevel
-        when 1                                            # H1
+        when 1   # H1, asterisks above and below
           ('*' * hlength) + "\n" + htext + "\n" + ('*' * hlength) + "\n"
-        when 2                                            # H2
+        when 2   # H1, dashes above and below
           ('-' * hlength) + "\n" + htext + "\n" + ('-' * hlength) + "\n"
-        else                                              # H3-H6 are styled the same
+        else     # H3-H6, dashes below
           htext + "\n" + ('-' * htext.length) + "\n"
       end
     end
 
-    txt.gsub!(/<a.*href=\"([^\"]*)\"[^>]*>(.*)<\/a>/i) do |s|   # links
-      $2 + ' [' + $1 + ']'
+    # links
+    txt.gsub!(/<a.*href=\"([^\"]*)\"[^>]*>(.*)<\/a>/i) do |s|
+      $2.strip + ' ( ' + $1.strip + ' )'
     end
 
-    txt.gsub!(/(<li[\s]+[^>]*>|<li>)/i, '  * ')                     # unordered LIsts
-    txt.gsub!(/<\/p>/i, "\n\n")                           # paragraphs
+    # lists -- TODO: should handle ordered lists
+    txt.gsub!(/[\s]*(<li[^>]*>)[\s]*/i, '* ')
+    # list not followed by a newline
+    txt.gsub!(/<\/li>[\s]*(?![\n])/i, "\n")
+    
+    # paragraphs and line breaks
+    txt.gsub!(/<\/p>/i, "\n\n")
+    txt.gsub!(/<br[\/ ]*>/i, "\n")
+    
+    # strip remaining tags
+    txt.gsub!(/<\/?[^>]*>/, '')
+
+    # wrap text
+    txt = r.format(('[' * line_length), txt)
     
-    txt.gsub!(/<\/?[^>]*>/, '')                           # strip remaining tags
-    txt.gsub!(/\A[\s]+|[\s]+\Z|^[ \t]+/m, '')             # strip extra spaces
-    txt.gsub!(/[\n]{3,}/m, "\n\n")                        # tighten line breaks
-
-    txt = r.format(('[' * line_length), txt)   # wrap text
-    txt.gsub!(/^[\*][\s]/m, '  * ')                        # add spaces back to lists
-  
-    txt.gsub!(/^\s+$/, "\n")                    # \r\n and \r -> \n
-    txt.gsub!(/\r\n?/, "\n")                    # \r\n and \r -> \n
-    txt.gsub!(/[\n]{3,}/, "\n")
-    txt
+    # remove linefeeds (\r\n and \r -> \n)
+    txt.gsub!(/\r\n?/, "\n")
+    
+    # strip extra spaces
+    txt.gsub!(/\302\240+/, " ") # non-breaking spaces -> spaces
+    txt.gsub!(/\n[ \t]+/, "\n") # space at start of lines
+    txt.gsub!(/[ \t]+\n/, "\n") # space at end of lines
+
+    # no more than two consecutive newlines
+    txt.gsub!(/[\n]{3,}/, "\n\n")
+
+    txt.strip
   end
 end

data/lib/premailer/premailer.rb

@@ -33,7 +33,7 @@ class Premailer
   include HtmlToPlainText
   include CssParser
 
-  VERSION = '1.5.2'
+  VERSION = '1.5.3'
 
   CLIENT_SUPPORT_FILE = File.dirname(__FILE__) + '/../../misc/client_support.yaml'
 
@@ -42,7 +42,9 @@ class Premailer
   # should also exclude :first-letter, etc...
 
   # URI of the HTML file used
-  attr_reader   :html_file            
+  attr_reader   :html_file
+  
+  attr_reader   :processed_doc
 
   module Warnings
     NONE = 0
@@ -175,6 +177,8 @@ class Premailer
 
     doc.search('*').remove_class if @options[:remove_classes]  
 
+    @processed_doc = doc
+
     doc.to_html
   end
 
@@ -265,7 +269,6 @@ protected
       next if tags.empty?
 
       tags.each do |tag|
-
         # skip links that look like they have merge tags
         # and mailto, ftp, etc...
         if tag.attributes[attribute] =~ /^(\{|\[|<|\#|mailto:|ftp:|gopher:)/i
@@ -329,15 +332,15 @@ protected
 
   # from http://www.ruby-forum.com/topic/140101
   def self.canonicalize(uri) # :nodoc:
-     u = uri.kind_of?(URI) ? uri : URI.parse(uri.to_s)
-     u.normalize!
-     newpath = u.path
-     while newpath.gsub!(%r{([^/]+)/\.\./?}) { |match|
-                $1 == '..' ? match : ''
-              } do end
-     newpath = newpath.gsub(%r{/\./}, '/').sub(%r{/\.\z}, '/')
-     u.path = newpath
-     u.to_s
+    u = uri.kind_of?(URI) ? uri : URI.parse(uri.to_s)
+    u.normalize!
+    newpath = u.path
+    while newpath.gsub!(%r{([^/]+)/\.\./?}) { |match|
+      $1 == '..' ? match : ''
+    } do end
+    newpath = newpath.gsub(%r{/\./}, '/').sub(%r{/\.\z}, '/')
+    u.path = newpath
+    u.to_s
   end
 
   # Check <tt>CLIENT_SUPPORT_FILE</tt> for any CSS warnings

data/rakefile.rb

@@ -0,0 +1,42 @@
+require 'rake'
+require 'fileutils'
+require 'lib/premailer'
+
+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)
+  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

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: premailer
 version: !ruby/object:Gem::Version 
-  version: 1.5.2
+  version: 1.5.3
 platform: ruby
 authors: 
 - Alex Dunae
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2009-11-27 00:00:00 -08:00
+date: 2009-12-03 00:00:00 -08:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -51,12 +51,14 @@ extensions: []
 extra_rdoc_files: []
 
 files: 
-- README.rdoc
-- CHANGELOG.rdoc
-- LICENSE.rdoc
+- init.rb
+- rakefile.rb
 - lib/premailer.rb
-- lib/premailer/premailer.rb
 - lib/premailer/html_to_plain_text.rb
+- lib/premailer/premailer.rb
+- CHANGELOG.rdoc
+- LICENSE.rdoc
+- README.rdoc
 - misc/client_support.yaml
 has_rdoc: true
 homepage: http://premailer.dialect.ca/