Introduction
Mail is an internet library for Ruby that is designed to handle emails generation, parsing and sending in a simple, rubyesque manner.
The purpose of this library is to provide a single point of access to handle all email functions, including sending and receiving emails. All network type actions are done through proxy methods to Net::SMTP, Net::POP3 etc.
Built from my experience with TMail, it is designed to be a pure ruby implementation that makes generating, sending and parsing emails a no brainer.
It is also designed form the ground up to work with Ruby 1.9. This is because Ruby 1.9 handles text encodings much more magically than Ruby 1.8.x and so these features have been taken full advantage of in this library allowing Mail to handle a lot more messages more cleanly than TMail. Mail does run on Ruby 1.8.x... it's just not as fun to code.
Finally, Mail has been designed with a very simple object oriented system that really opens up the email messages you are parsing, if you know what you are doing, you can fiddle with every last bit of your email directly.
Compatibility
Mail is tested and works on the following platforms:
- jruby-1.5.2 [ x86_64-java ]
- ree-1.8.7-2010.02 [ x86_64 ]
- ruby-1.8.6-p399 [ x86_64 ]
- ruby-1.8.7-p302 [ x86_64 ]
- ruby-1.9.2-p0 [ x86_64 ]
Discussion
If you want to discuss mail with like minded individuals, please subscribe to the Google Group.
Current Capabilities of Mail
- RFC2822 Support, Reading and Writing
- RFC2045-2049 Support for multipart emails
- Support for creating multipart alternate emails
- Support for reading multipart/report emails & getting details from such
- Support for multibyte emails - needs quite a lot of work and testing
- Wrappers for File, Net/POP3, Net/SMTP
- Auto encoding of non US-ASCII header fields
- Auto encoding of non US-ASCII bodies
Mail is RFC2822 compliant now, that is, it can parse and generate valid US-ASCII emails. There are a few obsoleted syntax emails that it will have problems with, but it also is quite robust, meaning, if it finds something it doesn't understand it will not crash, instead, it will skip the problem and keep parsing. In the case of a header it doesn't understand, it will initialise the header as an optional unstructured field and continue parsing.
This means Mail won't (ever) crunch your data (I think).
You can also create MIME emails. There are helper methods for making a multipart/alternate email for text/plain and text/html (the most common pair) and you can manually create any other type of MIME email.
Roadmap
Next TODO:
- Improve MIME support for character sets in headers, currently works, mostly, needs refinement.
Testing Policy
Basically... we do BDD on Mail. No method gets written in Mail without a corresponding or covering spec. We expect as a minimum 100% coverage measured by RCov. While this is not perfect by any measure, it is pretty good. Additionally, all functional tests from TMail are to be passing before the gem gets released.
It also means you can be sure Mail will behave correctly.
API Policy
No API removals within a single point release. All removals to be depreciated with warnings for at least one MINOR point release before removal.
Also, all private or protected methods to be declared as such - though this is still I/P.
Installation
Installation is fairly simple, I host mail on rubygems, so you can just do:
# gem install mail
Encodings
If you didn't know, handling encodings in Emails is not as straight forward as you would hope.
I have tried to simplify it some:
All objects that can render into an email, have an
#encoded
method. Encoded will return the object as a complete string ready to send in the mail system, that is, it will include the header field and value and CRLF at the end and wrapped as needed.All objects that can render into an email, have a :decoded method. Decoded will return the object's "value" only as a string. This means it will not include the header fields (like 'To:' or 'Subject:').
By default, calling
#to_s
on a container object will call its encoded method, while#to_s
on a field object will call it's decoded method. So calling#to_s
on a Mail object will return the mail, all encoded ready to send, while calling#to_s
on the From field or the body will return the decoded value of the object. The header object of Mail is considered a container. If you are in doubt, call#encoded
, or#decoded
explicitly, this is safer if you are not sure.Structured fields that have parameter values that can be encoded (e.g. Content-Type) will provide decoded parameter values when you call the parameter names as methods against the object.
Structured fields that have parameter values that can be encoded (e.g. Content-Type) will provide encoded parameter values when you call the parameter names through the
object.parameters['
method call.']
Contributing
Please do! Contributing is easy in Mail:
- Check the Reference RFCs, they are in the References directory, so no excuses.
- Open a ticket on GitHub, maybe someone else has the problem too
- Make a fork of my GitHub repository
- Make a spec driven change to the code base
- Make sure it works and all specs pass, on Ruby versions 1.8.6, 1.8.7 and 1.9
- Update the README if needed to reflect your change / addition
- With all specs passing push your changes back to your fork
- Send me a pull request
Usage
All major mail functions should be able to happen from the Mail module.
So, you should be able to just require 'mail'
to get started.
Making an email
mail = Mail.new do
from '[email protected]'
to '[email protected]'
subject 'This is a test email'
body File.read('body.txt')
end
mail.to_s #=> "From: [email protected]\r\nTo: you@...
Making an email, have it your way:
mail = Mail.new do
body File.read('body.txt')
end
mail['from'] = '[email protected]'
mail[:to] = '[email protected]'
mail.subject = 'This is a test email'
mail.to_s #=> "From: [email protected]\r\nTo: you@...
Don't Worry About Message IDs:
mail = Mail.new do
to '[email protected]'
body 'Some simple body'
end
mail.to_s =~ /Message\-ID: <[\d\w_]+@.+.mail/ #=> 27
Mail will automatically add a Message-ID field if it is missing and give it a unique, random Message-ID along the lines of:
<[email protected]>
Or do worry about Message-IDs:
mail = Mail.new do
to '[email protected]'
'<[email protected]>'
body 'Some simple body'
end
mail.to_s =~ /Message\-ID: <[email protected]>/ #=> 27
Mail will take the message_id you assign to it trusting that you know what you are doing.
Sending an email:
Mail defaults to sending via SMTP to local host port 25. If you have a sendmail or postfix daemon running on on this port, sending email is as easy as:
Mail.deliver do
from '[email protected]'
to '[email protected]'
subject 'Here is the image you wanted'
body File.read('body.txt')
add_file '/full/path/to/somefile.png'
end
or
mail = Mail.new do
from '[email protected]'
to '[email protected]'
subject 'Here is the image you wanted'
body File.read('body.txt')
add_file :filename => 'somefile.png', :content => File.read('/somefile.png')
end
mail.deliver!
Sending via sendmail can be done like so:
mail = Mail.new do
from '[email protected]'
to '[email protected]'
subject 'Here is the image you wanted'
body File.read('body.txt')
add_file :filename => 'somefile.png', :content => File.read('/somefile.png')
end
mail.delivery_method :sendmail
mail.deliver
Getting emails from a pop server:
You can configure Mail to receive email using retriever_method
within Mail.defaults
:
Mail.defaults do
retriever_method :pop3, :address => "pop.gmail.com",
:port => 995,
:user_name => '<username>',
:password => '<password>',
:enable_ssl => true
end
You can access incoming email in a number of ways.
The most recent email:
Mail.all #=> Returns an array of all emails
Mail.first #=> Returns the first unread email
Mail.last #=> Returns the first unread email
The first 10 emails sorted by date in ascending order:
emails = Mail.find(:what => :first, :count => 10, :order => :asc)
emails.length #=> 10
Or even all emails:
emails = Mail.all
emails.length #=> LOTS!
Reading an Email
mail = Mail.read('/path/to/message.eml')
mail.envelope.from #=> '[email protected]'
mail.from.addresses #=> ['[email protected]', '[email protected]']
mail.sender.address #=> '[email protected]'
mail.to #=> '[email protected]'
mail.cc #=> '[email protected]'
mail.subject #=> "This is the subject"
mail.date.to_s #=> '21 Nov 1997 09:55:06 -0600'
mail. #=> '<[email protected]>'
mail.body.decoded #=> 'This is the body of the email...
Many more methods available.
Reading a Multipart Email
mail = Mail.read('multipart_email')
mail.multipart? #=> true
mail.parts.length #=> 2
mail.preamble #=> "Text before the first part"
mail.epilogue #=> "Text after the last part"
mail.parts.map { |p| p.content_type } #=> ['text/plain', 'application/pdf']
mail.parts.map { |p| p.class } #=> [Mail::Message, Mail::Message]
mail.parts[0].content_type_parameters #=> {'charset' => 'ISO-8859-1'}
mail.parts[1].content_type_parameters #=> {'name' => 'my.pdf'}
Mail generates a tree of parts. Each message has many or no parts. Each part is another message which can have many or no parts.
A message will only have parts if it is a multipart/mixed or related/mixed content type and has a boundary defined.
Writing and sending a multipart/alternative (html and text) email
Mail makes some basic assumptions and makes doing the common thing as simple as possible.... (asking a lot from a mail library)
mail = Mail.deliver do
to '[email protected]'
from 'Mikel Lindsaar <[email protected]>'
subject 'First multipart email sent with Mail'
text_part do
body 'This is plain text'
end
html_part do
content_type 'text/html; charset=UTF-8'
body '<h1>This is HTML</h1>'
end
end
Mail then delivers the email at the end of the block and returns the resulting Mail::Message object, which you can then inspect if you so desire...
puts mail.to_s #=>
To: [email protected]
From: Mikel Lindsaar <[email protected]>
Subject: First multipart email sent with Mail
Content-Type: multipart/alternative;
boundary=--==_mimepart_4a914f0c911be_6f0f1ab8026659
Message-ID: <[email protected]>
Date: Mon, 24 Aug 2009 00:15:46 +1000
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
----==_mimepart_4a914f0c911be_6f0f1ab8026659
Content-ID: <[email protected]>
Date: Mon, 24 Aug 2009 00:15:46 +1000
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
This is plain text
----==_mimepart_4a914f0c911be_6f0f1ab8026659
Content-Type: text/html; charset=UTF-8
Content-ID: <[email protected]>
Date: Mon, 24 Aug 2009 00:15:46 +1000
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
<h1>This is HTML</h1>
----==_mimepart_4a914f0c911be_6f0f1ab8026659--
Mail inserts the content transfer encoding, the mime version, the content-id's and handles the content-type and boundary.
Mail assumes that if your text in the body is only us-ascii, that your transfer encoding is 7bit and it is text/plain. You can override this by explicitly declaring it.
Making Multipart/Alternate, without a block
You don't have to use a block with the text and html part included, you can just do it declaratively. However, you need to add Mail::Parts to an email, not Mail::Messages.
mail = Mail.new do
to '[email protected]'
from 'Mikel Lindsaar <[email protected]>'
subject 'First multipart email sent with Mail'
end
text_part = Mail::Part.new do
body 'This is plain text'
end
html_part = Mail::Part.new do
content_type 'text/html; charset=UTF-8'
body '<h1>This is HTML</h1>'
end
mail.text_part = text_part
mail.html_part = html_part
Results in the same email as done using the block form
Getting error reports from an email:
@mail = Mail.read('/path/to/bounce_message.eml')
@mail.bounced? #=> true
@mail.final_recipient #=> rfc822;[email protected]
@mail.action #=> failed
@mail.error_status #=> 5.5.0
@mail.diagnostic_code #=> smtp;550 Requested action not taken: mailbox unavailable
@mail.retryable? #=> false
Attaching and Detaching Files
You can just read the file off an absolute path, Mail will try to guess the mime_type and will encode the file in Base64 for you.
@mail = Mail.new
@mail.add_file("/path/to/file.jpg")
@mail.parts.first. #=> true
@mail.parts.first.content_transfer_encoding.to_s #=> 'base64'
@mail..first.mime_type #=> 'image/jpg'
@mail..first.filename #=> 'file.jpg'
@mail..first.decoded == File.read('/path/to/file.jpg') #=> true
Or You can pass in file_data and give it a filename, again, mail will try and guess the mime_type for you.
@mail = Mail.new
@mail.['myfile.pdf'] = File.read('path/to/myfile.pdf')
@mail.parts.first. #=> true
@mail..first.mime_type #=> 'application/pdf'
@mail..first.decoded == File.read('path/to/myfile.pdf') #=> true
You can also override the guessed MIME media type if you really know better than mail (this should be rarely needed)
@mail = Mail.new
file_data = File.read('path/to/myfile.pdf')
@mail.['myfile.pdf'] = { :mime_type => 'application/x-pdf',
:content => File.read('path/to/myfile.pdf') }
@mail.parts.first.mime_type #=> 'application/x-pdf'
Of course... Mail will round trip an attachment as well
@mail = Mail.new do
to '[email protected]'
from 'Mikel Lindsaar <[email protected]>'
subject 'First multipart email sent with Mail'
text_part do
body 'Here is the attachment you wanted'
end
html_part do
content_type 'text/html; charset=UTF-8'
body '<h1>Funky Title</h1><p>Here is the attachment you wanted</p>'
end
add_file '/path/to/myfile.pdf'
end
@round_tripped_mail = Mail.new(@mail.encoded)
@round_tripped_mail..length #=> 1
@round_tripped_mail..first.filename #=> 'myfile.pdf'
Using Mail with Testing or Spec'ing Libraries
If mail is part of your system, you'll need a way to test it without actually sending emails, the TestMailer can do this for you.
require 'mail'
=> true
Mail.defaults do
delivery_method :test
end
=> #<Mail::Configuration:0x19345a8 @delivery_method=Mail::TestMailer>
Mail::TestMailer.deliveries
=> []
Mail.deliver do
to '[email protected]'
from '[email protected]'
subject 'testing'
body 'hello'
end
=> #<Mail::Message:0x19284ec ...
Mail::TestMailer.deliveries.length
=> 1
Mail::TestMailer.deliveries.first
=> #<Mail::Message:0x19284ec ...
Mail::TestMailer.deliveries.clear
=> []
Excerpts from TREC Spam Corpus 2005
The spec fixture files in spec/fixtures/emails/from_trec_2005 are from the 2005 TREC Public Spam Corpus. They remain copyrighted under the terms of that project and license agreement. They are used in this project to verify and describe the development of this email parser implementation.
http://plg.uwaterloo.ca/~gvcormac/treccorpus/
They are used as allowed by 'Permitted Uses, Clause 3':
"Small excerpts of the information may be displayed to others
or published in a scientific or technical context, solely for
the purpose of describing the research and development and
related issues."
-- http://plg.uwaterloo.ca/~gvcormac/treccorpus/
License
(The MIT License)
Copyright (c) 2009, 2010, 2011
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the 'Software'), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.