mms2r 2.4.1 → 3.0.0

data/.gitignore

@@ -3,4 +3,5 @@ coverage/
 pkg/
 email.txt
 coverage.info
+mkmf.log
 doc/

data/History.txt

@@ -1,3 +1,8 @@
+### 3.0.0 / 2010-02-24 (General Crozier - Chairman of the Joint Chiefs of Staff)
+
+* 1 new feature
+  * dependence upon Mail gem rather than TMail gem
+
 ### 2.4.1 / 2010-02-07 (Vater Orlaag - political and spiritual specialist)
 
 * 3 minor enhancements

data/Manifest.txt

@@ -44,7 +44,7 @@ init.rb
 lib/mms2r.rb
 lib/mms2r/media.rb
 lib/mms2r/media/sprint.rb
-lib/tmail_ext.rb
+lib/mail_ext.rb
 mms2r.gemspec
 test/fixtures/1nbox-2images-01.mail
 test/fixtures/1nbox-2images-02.mail

data/README.txt

@@ -2,8 +2,8 @@
 
   http://mms2r.rubyforge.org/
   by Mike Mondragon
-  http://rubyforge.org/tracker/?group_id=3065
   http://github.com/monde/mms2r
+  http://github.com/monde/mms2r/issues
   http://peepcode.com/products/mms2r-pdf
 
 == DESCRIPTION
@@ -60,6 +60,12 @@ Corpus of carriers currently processed by MMS2R:
 * Virgin Mobile of Canada: vmobile.ca
 * Vodacom: mms.vodacom4me.co.za
 
+Corpus of smart phones known to MMS2R:
+* Blackberry variants
+* Droid variants
+* HTC variants (T-Mobile Dash, Sprint HERO)
+* Apple iPhone variants
+
 == FEATURES
 
 * #default_media and #default_text methods return a File that can be used in
@@ -67,8 +73,8 @@ Corpus of carriers currently processed by MMS2R:
 * #process supports blocks to for enumerating over the content of the MMS
 * #process can be made lazy when :process => :lazy is passed to new
 * logging is enabled when :logger => your_logger is passed to new
-* an mms instance acts like a tmail object, any methods not defined on the
-  instance are delegated to its underlying tmail object
+* an mms instance acts like a Mail object, any methods not defined on the
+  instance are delegated to its underlying Mail object
 * #device_type? returns a symbol representing a device or smartphone type
   Known smartphones thus far: iPhone, BlackBerry, T-Mobile Dash, Droid
 
@@ -83,12 +89,9 @@ http://peepcode.com/products/mms2r-pdf
   require 'mms2r'
 
   # required for the example
-  require 'tmail'
   require 'fileutils'
 
-  mail = MMS2R.parse mail
-  # mail = MMS2R.parse File.read('some_saved_mail.file')
-
+  mail = MMS2R.new(Mail.read('some_saved_mail.file'))
   puts "mail has default carrier subject" if mail.subject.empty?
 
   # access the sender's phone number
@@ -119,7 +122,7 @@ http://peepcode.com/products/mms2r-pdf
 
   puts "does the mail have quicktime video? #{!mail.media['video/quicktime'].nil?}"
 
-  puts "plus run anything that TMail provides, e.g. #{mail.to.inspect}"
+  puts "plus run anything that Mail provides, e.g. #{mail.to.inspect}"
 
   # check if the mail is from a mobile phone
   puts "mail is from a mobile phone #{mail.is_mobile?}"
@@ -149,7 +152,7 @@ http://peepcode.com/products/mms2r-pdf
 
 == REQUIREMENTS
 
-* TMail
+* Mail
 * Nokogiri (for mms from Sprint)
 * UUIDTools
 

data/Rakefile

@@ -11,17 +11,18 @@ end
 $LOAD_PATH.unshift File.join(File.dirname(__FILE__), "lib")
 require 'mms2r'
 
-Hoe.new('mms2r', MMS2R::Media::VERSION) do |p|
+Hoe.spec('mms2r') do |p|
+  p.version        = MMS2R::Media::VERSION
   p.rubyforge_name = 'mms2r'
-  p.author = ['Mike Mondragon']
-  p.email = ['mikemondragon@gmail.com']
-  p.summary = 'Extract user media from MMS (and not carrier cruft)'
-  p.description = p.paragraphs_of('README.txt', 2..5).join("\n\n")
-  p.url = p.paragraphs_of('README.txt', 1).first.strip
-  p.changes = p.paragraphs_of('History.txt', 0..2).join("\n\n")
-  p.extra_deps << ['nokogiri']
-  p.extra_deps << ['tmail']
-  p.extra_deps << ['uuidtools']
+  p.author         = ['Mike Mondragon']
+  p.email          = ['mikemondragon@gmail.com']
+  p.summary        = 'Extract user media from MMS (and not carrier cruft)'
+  p.description    = p.paragraphs_of('README.txt', 2..5).join("\n\n")
+  p.url            = p.paragraphs_of('README.txt', 1).first.strip
+  p.changes        = p.paragraphs_of('History.txt', 0..1).join("\n\n")
+  p.extra_deps << ['nokogiri', '>= 1.4.0']
+  p.extra_deps << ['mail', '>= 2.1.0']
+  p.extra_deps << ['uuidtools', '>= 2.1.0']
   p.clean_globs << 'coverage'
 end
 

data/conf/mobile.indosat.net.id.yml

@@ -9,6 +9,6 @@ ignore:
    - cp__newBanner.jpg
 number:
   - return-path
-  - /^<([^@]+)@.*/
+  - /^([^@]+)@.*/
   - "\1"
 

data/conf/waw.plspictures.com.yml

@@ -5,7 +5,7 @@ transform:
     - ""
 number:
   - return-path
-  - /^<([^@]+)@.*/
+  - /^([^@]+)@.*/
   - "\1"
 ignore:
   text/html: 

data/lib/mail_ext.rb

@@ -0,0 +1,37 @@
+#--
+# Copyright (c) 2010 by Mike Mondragon (mikemondragon@gmail.com)
+#
+# Please see the LICENSE file for licensing information.
+#++
+
+class Mail::Message
+
+    ##
+    # Generically determines the mime-type of a message used in mms2r processing.
+    # Guarantees a type is returned.
+
+    def part_type?
+      if self.content_type
+        self.content_type.split(';').first.downcase
+      else
+        'text/plain'
+      end
+    end
+
+    ##
+    # override #filename to account for the true filename in the content_type
+    # returns foo.jpg #content_type is 'image/jpeg; filename="foo.jpg"; name="foo.jpg"'
+    # returns foo.jpg #content_type is 'image/jpeg;Name=foo.jpg'
+
+    def filename
+      if self.content_type && names = Hash[self.content_type.split(';').map{|t| t.strip.split('=')}]
+        if name = names.detect{|key,val| key.downcase == 'filename'} || names.detect{|key,val| key.downcase == 'name'}
+          return (name.last.match(/^"?(.+?)"?$/))[1]
+        end
+      end
+
+      find_attachment
+    end
+
+
+end

data/lib/mms2r.rb

@@ -39,29 +39,31 @@ module MMS2R
     ##
     # MMS2R library version
 
-    VERSION = '2.4.1'
+    VERSION = '3.0.0'
 
   end
 
   # Simple convenience function to make it a one-liner:
   # MMS2R.parse raw_mail or MMS2R.parse File.load(raw_mail)
-  # Combined w/ the method_missing delegation, this should behave as an enhanced TMail object, more or less.
+  # Combined w/ the method_missing delegation, this should behave as an enhanced Mail object, more or less.
   def self.parse raw_mail
-    mail = TMail::Mail.parse raw_mail
+    mail = Mail.new raw_mail
     MMS2R::Media.new(mail)
   end
 
 end
 
 require 'rubygems'
-require 'tmail/mail'
+require 'mail'
+gem 'mail', '>= 2.1.2'
 require 'fileutils'
 require 'pathname'
 require 'tmpdir'
 require 'yaml'
 require 'uuidtools'
+require 'iconv'
 
-require File.join(File.dirname(__FILE__), 'tmail_ext')
+require File.join(File.dirname(__FILE__), 'mail_ext')
 require File.join(File.dirname(__FILE__), 'mms2r', 'media')
 require File.join(File.dirname(__FILE__), 'mms2r', 'media', 'sprint')
 MMS2R.register('pm.sprint.com', MMS2R::Media::Sprint)

data/lib/mms2r/media.rb

@@ -23,7 +23,7 @@
 #
 #  require 'rubygems'
 #  require 'mms2r'
-#  mail = TMail::Mail.parse(IO.readlines("sample-MMS.file").join)
+#  mail = Mail.read("sample-MMS.file")
 #  mms = MMS2R::Media.new(mail)
 #  subject = mms.subject
 #  number = mms.number
@@ -114,14 +114,14 @@ module MMS2R
       end
     end
 
-    # Pass off everything we don't do to the TMail object
+    # Pass off everything we don't do to the Mail object
     # TODO: refactor to explicit addition a la http://blog.jayfields.com/2008/02/ruby-replace-methodmissing-with-dynamic.html
     def method_missing method, *args, &block
       mail.send method, *args, &block
     end
 
     ##
-    # TMail object that the media files were derived from.
+    # Mail object that the media files were derived from.
 
     attr_reader :mail
 
@@ -143,14 +143,9 @@ module MMS2R
 
     attr_reader :media_dir
 
-    ##
-    # Various multi-parts that are bundled into mail
-
-    MULTIPARTS_TO_SPLIT = [ 'multipart/related', 'multipart/alternative', 'multipart/mixed', 'multipart/appledouble' ]
-
     ##
     # Factory method that creates MMS2R::Media products based on the domain
-    # name of the carrier from which the MMS originated.  mail is a TMail
+    # name of the carrier from which the MMS originated.  mail is a Mail
     # object.
 
     def self.create(mail)
@@ -171,9 +166,8 @@ module MMS2R
 
     def self.domain(mail)
       return_path = case
-        when mail.header['return-path']
-          matched = /^<.+@([^@]+)>$/.match(mail.header['return-path'].to_s.strip)
-          matched ? matched[1] : ''
+        when mail.return_path
+          mail.return_path ? mail.return_path.split('@').last : ''
         else
           ''
         end
@@ -248,11 +242,15 @@ module MMS2R
     def number
       unless @number
         params = config['number']
-        if params && (header = mail.header[params[0]])
+        if params && params.any? && (header = mail.header[params[0]])
           @number = header.to_s.gsub(eval(params[1]), params[2])
         end
+        if @number.nil? || @number.blank?
+          @number = mail.from.first.split(/@|\//).first rescue ""
+        end
       end
-      @number ||= mail.from.first.split('@').first rescue ""
+
+      @number
     end
 
     ##
@@ -344,7 +342,7 @@ module MMS2R
         for i in 1..2
           flat = []
           parts.each do |p|
-            if MULTIPARTS_TO_SPLIT.include?(p.part_type?)
+            if p.multipart?
               p.parts.each {|mp| flat << mp }
             else
               flat << p
@@ -380,12 +378,12 @@ module MMS2R
     # true for media such as images that are advertising, carrier logos, etc.
     # See the ignore section in the discussion of the built-in configuration.
 
-    def ignore_media?(type,part)
+    def ignore_media?(type, part)
       ignores = config['ignore'][type] || []
-      ignore = ignores.detect{|test| filename?(part) == test}
-      ignore ||= ignores.detect{|test| filename?(part) =~ eval(test) if test.index('/') == 0 }
-      ignore ||= ignores.detect{|test| part.body.strip =~ eval(test) if test.index('/') == 0 }
-      ignore ||= (part.body.strip.size == 0 ? true : nil)
+      ignore   = ignores.detect{ |test| filename?(part) == test}
+      ignore ||= ignores.detect{ |test| filename?(part) =~ eval(test) if test.index('/') == 0 }
+      ignore ||= ignores.detect{ |test| part.body.decoded.strip =~ eval(test) if test.index('/') == 0 }
+      ignore ||= (part.body.decoded.strip.size == 0 ? true : nil)
       ignore.nil? ? false : true
     end
 
@@ -402,20 +400,20 @@ module MMS2R
     # Returns a tuple of content type, file path
 
     def process_media(part)
-      # TMail body auto-magically decodes quoted
+      # Mail body auto-magically decodes quoted
       # printable for text/html type.
       file = temp_file(part)
-      if part.main_type('text') == 'text' ||
-         part.content_type == 'application/smil'
+      if part.part_type? =~ /^text\// ||
+         part.part_type? == 'application/smil'
         type, content = transform_text_part(part)
         mode = 'w'
       else
-        if part.content_type == 'application/octet-stream'
+        if part.part_type? == 'application/octet-stream'
           type = type_from_filename(filename?(part))
         else
-          type = part.content_type
+          type = part.part_type?
         end
-        content = part.body
+        content = part.body.decoded
         mode = 'wb' # open with binary bit for Windows for non text
       end
       return type, nil if content.nil? || content.empty?
@@ -456,7 +454,7 @@ module MMS2R
 
     def transform_text_part(part)
       type = part.part_type?
-      text = part.body.strip
+      text = part.body.decoded.strip
       transform_text(type, text)
     end
 
@@ -506,11 +504,9 @@ module MMS2R
     # returns a filename declared for a part, or a default if its not defined
 
     def filename?(part)
-      name = part.sub_header("content-type", "name") ||
-        part.sub_header("content-disposition", "filename") ||
-        (part['content-location'] && part['content-location'].to_s.strip)
+      name = part.filename
       if (name.nil? || name.empty?)
-        if part['content-id'] && part['content-id'].real_body.strip =~ /^<(.+)>$/
+        if part.content_id && part.content_id.strip =~ /^<(.+)>$/
           name = $1
         else
           name = "#{Time.now.to_f}.#{self.default_ext(part.part_type?)}"
@@ -570,11 +566,9 @@ module MMS2R
       headers.keys.each do |header|
         if mail.header[header.downcase]
           # headers[header] refers to a hash of smart phone types with regex values
-          # that if they match the header signals the type should be returned
+          # that if they match, the header signals the type should be returned
           headers[header].each do |type, regex|
-            # HACK to get at the full value of the header before TMail parses according to spec
-            # see @body in and ensure_parsed in lib/tmail/header.rb
-            return type if mail.header[header.downcase].instance_variable_get(:@body) =~ regex
+            return type if mail.header[header.downcase].decoded =~ regex
           end
         end
       end

data/lib/mms2r/media/sprint.rb

@@ -15,12 +15,12 @@ module MMS2R
     ##
     # Sprint version of MMS2R::Media
     #
-    # Sprint is an annoying carrier because they don't actually transmit user 
-    # generated content (like images or videos) directly in the MMS message.  
-    # Instead, they hijack the media that is sent from the cellular subscriber 
-    # and place that content on a content server.  In place of the media 
-    # the recipient receives a HTML message with unsolicited Sprint 
-    # advertising and links back to their content server.  The recipient has 
+    # Sprint is an annoying carrier because they don't actually transmit user
+    # generated content (like images or videos) directly in the MMS message.
+    # Instead, they hijack the media that is sent from the cellular subscriber
+    # and place that content on a content server.  In place of the media
+    # the recipient receives a HTML message with unsolicited Sprint
+    # advertising and links back to their content server.  The recipient has
     # to click through Sprint more pages to view the content.
     #
     # The default subject on these messages from the
@@ -29,11 +29,11 @@ module MMS2R
     class Sprint < MMS2R::Media
 
       ##
-      # Override process() because Sprint doesn't attach media (images, video, 
-      # etc.) to its MMS.  Media such as images and videos are hosted on a 
-      # Sprint content server.   MMS2R::Media::Sprint has to pick apart an 
-      # HTML attachment to find the URL to the media on Sprint's content 
-      # server and download each piece of content.  Any text message part of 
+      # Override process() because Sprint doesn't attach media (images, video,
+      # etc.) to its MMS.  Media such as images and videos are hosted on a
+      # Sprint content server.   MMS2R::Media::Sprint has to pick apart an
+      # HTML attachment to find the URL to the media on Sprint's content
+      # server and download each piece of content.  Any text message part of
       # the MMS if it exists is embedded in the html.
 
       def process
@@ -46,21 +46,21 @@ module MMS2R
           doc = nil
           parts.each do |p|
             next unless p.part_type? == 'text/html'
-            d = Nokogiri(p.body)
+            d = Nokogiri(p.body.decoded)
             title = d.at('title').inner_html
             if title =~ /You have new Picture Mail!/
               doc = d
-              @is_video = (p.body =~ /type=&quot;VIDEO&quot;&gt;/m ? true : false)
+              @is_video = (p.body.decoded =~ /type=&quot;VIDEO&quot;&gt;/m ? true : false)
             end
           end
           return if doc.nil? # it was a dud
           @is_video ||= false
-  
+
           # break it down
           sprint_phone_number(doc)
           sprint_process_text(doc)
           sprint_process_media(doc)
-        
+
           @was_processed = true
         end
 
@@ -90,8 +90,8 @@ module MMS2R
 
       def sprint_process_text(doc)
         # there is at least one <pre> with MMS text if text has been included by
-        # the user.  (note) we'll have to verify that if they attach multiple texts 
-        # to the MMS then Sprint stacks it up in multiple <pre>'s.  The only <pre> 
+        # the user.  (note) we'll have to verify that if they attach multiple texts
+        # to the MMS then Sprint stacks it up in multiple <pre>'s.  The only <pre>
         # tag in the document is for text from the user.
         doc.search("/html/body//pre").each do |pre|
           type = 'text/plain'
@@ -110,7 +110,7 @@ module MMS2R
         srcs = Array.new
         # collect all the images in the document, even though
         # they are <img> tag some might actually refer to video.
-        # To know the link refers to vide one must look at the 
+        # To know the link refers to vide one must look at the
         # content type on the http GET response.
         imgs = doc.search("/html/body//img")
         imgs.each do |i|
@@ -130,7 +130,7 @@ module MMS2R
         cnt = 0
         srcs.each do |src|
           begin
-            
+
             url = URI.parse(CGI.unescapeHTML(src))
             unless @is_video
               query={}