mail_address 1.0.0 → 1.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: ad0921e631ebcc827050a1a1c2572df665eca7f3
-  data.tar.gz: 398cf4fe10f1bebc20231b2cbb4d8dcca60a1dc5
+  metadata.gz: 02e966fc9f6878a8d0954ca5e658db596e581d15
+  data.tar.gz: 108c96f94e4b19c9aee8c61ec52c4fe6cf896d5f
 SHA512:
-  metadata.gz: 9b6da43a7b73f7df6d262188cdff389a7f077fe4c03707bf180d07aed408bb967ac05f21f8b9e9b9a9d341e7b6893659e2f7f2d323f7fc28648c49c44a32fb66
-  data.tar.gz: 56f4ee85472296c2e8fb48bd1fdd8a895462b4a629533faeebf3b274b9a5b600305c412dc38b940ef54cba2301decd8638a852f00dff595f91401f7faa2a1b52
+  metadata.gz: 477bdb94a6eba5ac3a302ae590bf4b926e0dd8f236c0d5cedf57112a5ccea8f058d6e391eff50000a317baba7f5c332eabe0606defab8d4cb333c9ff85268eed
+  data.tar.gz: ca4c41bed4759fc726fe72bae286c897d5fd97f52fd6fe71abde8843864121ea047f82d13609b2527de9d8a359b190c2948f71c88f6f957c5f6fd79ed06e7cef

data/README.md

@@ -1,7 +1,7 @@
 # MailAddress [![Build Status](https://travis-ci.org/kizashi1122/mail_address.svg)](https://travis-ci.org/kizashi1122/mail_address) [![Coverage Status](https://coveralls.io/repos/kizashi1122/mail_address/badge.png)](https://coveralls.io/r/kizashi1122/mail_address)
 
 MailAddress is a simple email address parser.
-This library is implemented based on Perl Module Mail::Address.
+This library is implemented based on Perl Module Mail::Address and added some improvements.
 
 [mail](https://github.com/mikel/mail) is a great gem library. But some email addresses (mostly are violated RFC) are unparsable with mail gem which is strictly RFC compliant. In perl, [Mail::Address](http://search.cpan.org/~markov/MailTools-2.14/lib/Mail/Address.pod) is a very common library to parse email address. Mail::Address conviniently can parse even RFC-violated email addresses such as:
 
@@ -11,7 +11,11 @@ Ello [Do Not Reply] <do-not-reply@ello.co> # [, ] are not permitted according to
 大阪 太郎<osaka@example.com> # no whitespace just before <
 ```
 
-So I straightforwardly converted Perl module Mail::Address to Ruby gem.
+So I straightforwardly converted Perl module Mail::Address to Ruby gem. Then I reviced it because original Mail::Address also has some bad points. For example:
+
+- if no ending parenthesis in name part, cannot parse correctly.
+- Modifications of name part are too much.
+
 
 ## Installation
 
@@ -32,6 +36,7 @@ Or install it yourself as:
 ## Usage
 
 It's almost the same as Mail::Address(Perl).
+But in this module, removed `comment` property from address class in version v1.0.0. Most people don't reallize comment I think.
 
 ```rb
 require 'mail_address'
@@ -39,22 +44,24 @@ require 'mail_address'
 line = "John 'M' Doe <john@example.com> (this is a comment), 大阪 太郎 <osaka@example.jp>"
 addrs = MailAddress.parse(line)
 
-p addrs[0].format     # "John 'M' Doe <john@example.com> (this is a comment)"
+p addrs[0].format     # "\"John 'M' Doe (this is a comment)\" <john@example.com>"
 p addrs[0].address    # "john@example.com"
-p addrs[0].name       # "John 'M' Doe"
-p addrs[0].comment    # "(this is a comment)"
-p addrs[0].phrase     # "John 'M' Doe"
+p addrs[0].name       # "John 'M' Doe (this is a comment)"
+p addrs[0].phrase     # "John 'M' Doe (this is a comment)"
 p addrs[0].host       # "example.com"
 p addrs[0].user       # "john"
+p addrs[0].original   # "John 'M' Doe <john@example.com> (this is a comment)"
 
 p addrs[1].format     # "\"大阪 太郎\" <osaka@example.jp>"
 p addrs[1].address    # "osaka@example.jp"
 p addrs[1].name       # "大阪 太郎"
-p addrs[1].comment    # ""
 p addrs[1].phrase     # "大阪 太郎"
 p addrs[1].host       # "example.jp"
 p addrs[1].user       # "osaka"
+p addrs[1].original   # "大阪 太郎 <osaka@example.jp>"
 ```
+`address.name` and `address.phrase` are almost same. 
+`address.phrase` keeps outermost double quotes or parentheses.
 
 ## Contributing
 

data/lib/mail_address/address.rb

@@ -2,16 +2,19 @@ module MailAddress
 
   class Address
 
-    def initialize(phrase, address)
+    def initialize(phrase, address_, original)
       @phrase = phrase
-      @address = address
+      @address = address_
+      @original = original
     end
-    attr_accessor :phrase, :address
+    attr_accessor :phrase, :original
 
     ATEXT = '[\-\w !#$%&\'*+/=?^`{|}~]'
 
     def format
       addr = []
+      return @original if address.nil?
+
       if !@phrase.nil? && @phrase.length > 0 then
         if @phrase.match(/\A\(/) && @phrase.match(/\)\z/)
           addr.push(@address) if !@address.nil? && @address.length > 0
@@ -52,14 +55,23 @@ module MailAddress
       name.length > 0 ? name : nil
     end
 
+    def address
+      # double check
+      if @address.nil? || @original.include?(@address)
+        @address
+      else
+        nil # wrongly constructed address
+      end
+    end
+
     def host
-      addr = @address || '';
+      addr = address || '';
       i = addr.rindex('@')
       i.nil? ? nil : addr[i + 1 .. -1]
     end
 
     def user
-      addr = @address || '';
+      addr = address || '';
       i = addr.rindex('@')
       i.nil? ? addr : addr[0 ... i]
     end

data/lib/mail_address/mail_address.rb

@@ -7,15 +7,16 @@ module MailAddress
 
     # undisclosed-recipient
     if line.match(/undisclosed[ \-]recipients?: ?;?/i)
-      return [ MailAddress::Address.new(line, nil) ]
+      return [ MailAddress::Address.new(line, nil, line) ]
     end
 
     # obviously invalid email address
     unless line.match(/[a-z0-9\.\-"_\+]@/i)
-      return [ MailAddress::Address.new(line, nil) ]
+      return [ MailAddress::Address.new(line, nil, line) ]
     end
 
     phrase, address, objs = [], [], []
+    original = ''
     depth, idx, end_paren_idx = 0, 0, 0
 
     tokens = _tokenize lines
@@ -24,13 +25,14 @@ module MailAddress
 
     for idx in 0 ... len do
 
+      token = tokens[idx]
+      substr = token[0, 1]
+      original << token
+
       if (end_paren_idx > 0 && end_paren_idx >= idx)
         next
       end
 
-      token = tokens[idx]
-      substr = token[0, 1]
-
       if (substr == '(' && !address.empty?) then
         end_paren_idx = _find_next_paren(idx, tokens, len)
         raise "cannot find end paren" if end_paren_idx == -1 # end paren must exists after address
@@ -42,11 +44,15 @@ module MailAddress
         depth -= 1 if depth > 0
       elsif (substr == ',' || substr == ';') then
         raise "Unmatched '<>' in line" if depth > 0
-        o = _complete(phrase, address)
+
+        original.sub!(/[,;]\s*\z/, '')
+
+        o = _complete(phrase, address, original)
 
         objs.push(o) if o
         depth = 0
         end_paren_idx = 0
+        original = ''
         _next = _find_next idx+1, tokens, len
       elsif (depth > 0) then
         token.strip!
@@ -110,9 +116,9 @@ module MailAddress
     -1
   end
 
-  def self._complete (phrase, address)
+  def self._complete (phrase, address, original)
     phrase.length > 0 || address.length > 0 or return nil
-    new_address = MailAddress::Address.new(phrase.join('').strip, address.join(''))
+    new_address = MailAddress::Address.new(phrase.join('').strip, address.join(''), original)
     phrase.clear; address.clear
     new_address
   end

data/lib/mail_address/version.rb

@@ -1,3 +1,3 @@
 module MailAddress
-  VERSION = "1.0.0"
+  VERSION = "1.1.0"
 end

data/spec/mail_address_spec.rb

@@ -14,6 +14,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
 
     # <address> only
     line = '<johndoe@example.com>'
@@ -24,6 +25,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
 
     # name + <address> (single byte only)
     line = 'John Doe <johndoe@example.com>'
@@ -34,6 +36,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("John Doe")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
 
     # name + <address> (multi byte)
     line = 'ジョン ドゥ <johndoe@example.com>'
@@ -44,6 +47,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("ジョン ドゥ")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
 
     line = 'Amazon.co.jp アソシエイト・プログラム <associates@amazon.co.jp>'
     results = MailAddress.parse(line)
@@ -53,6 +57,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("Amazon.co.jp アソシエイト・プログラム")
     expect(results[0].host).to eq("amazon.co.jp")
     expect(results[0].user).to eq('associates')
+    expect(results[0].original).to eq(line)
 
     # name (includes parens) + <address>
     line = 'Example (Twitterより) <notify@twitter.com>'
@@ -63,6 +68,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("Example (Twitterより)")
     expect(results[0].host).to eq("twitter.com")
     expect(results[0].user).to eq('notify')
+    expect(results[0].original).to eq(line)
 
     # name + <address> (multi byte) name is quoted
     line = '"ジョン ドゥ" <johndoe@example.com>'
@@ -73,6 +79,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq('"ジョン ドゥ"')
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
 
     # address + (note)
     line = 'johndoe@example.com (John Doe)'
@@ -83,6 +90,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("(John Doe)")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
 
     # address + (note) # nested paren
     line = 'johndoe@example.com (John (Mid) Doe)'
@@ -93,6 +101,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("(John (Mid) Doe)")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
 
     # address + (note) # note has special char
     line = 'johndoe@example.com (John@Doe)'
@@ -103,6 +112,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("(John@Doe)")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
 
     line = 'johndoe@example.com (John, Doe)'
     results = MailAddress.parse(line)
@@ -112,6 +122,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("(John, Doe)")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
 
     # name + <address> + (note)
     line = 'John Doe <johndoe@example.com> (Extra)'
@@ -122,6 +133,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("John Doe (Extra)")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
 
     # name + <address> (name has starting paren but doesn't have ending paren)
     line = 'John(Doe <johndoe@example.com>'
@@ -132,6 +144,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("John(Doe")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
 
     # ditto
     line = 'John ( Doe <johndoe@example.com>'
@@ -142,6 +155,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("John ( Doe")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
 
     # "address1" <address2>
     line = '"michael@example.jp" <johndoe@example.com>'
@@ -152,6 +166,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq('"michael@example.jp"')
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('johndoe')
+    expect(results[0].original).to eq(line)
   end
 
   it "normal case (multiple address)" do
@@ -164,6 +179,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("John 'M' Doe")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq("john")
+    expect(results[0].original).to eq("John 'M' Doe <john@example.com>")
 
     # Perl module Mail::Address returns '大阪 太郎 <osaka@example.jp>' (no double quote)
     # because regular expression \w matches even multibyte characters.
@@ -173,6 +189,7 @@ describe MailAddress do
     expect(results[1].phrase).to eq("大阪 太郎")
     expect(results[1].host).to eq("example.jp")
     expect(results[1].user).to eq("osaka")
+    expect(results[1].original).to eq('大阪 太郎 <osaka@example.jp>')
   end
 
   it "normal case (rfc-violated(RFC822) but commonly used in AU/DoCoMo)" do
@@ -186,6 +203,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("John Doe")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('"johndoe."')
+    expect(results[0].original).to eq(line)
 
     # contains '..'
     line = 'John Doe <"john..doe"@example.com>'
@@ -197,6 +215,7 @@ describe MailAddress do
     expect(results[0].phrase).to eq("John Doe")
     expect(results[0].host).to eq("example.com")
     expect(results[0].user).to eq('"john..doe"')
+    expect(results[0].original).to eq(line)
   end
 
   it "unparsable with mail gem (includes non-permitted char'[')" do
@@ -208,6 +227,7 @@ describe MailAddress do
     expect(results[0].phrase).to  eq("Ello [Do Not Reply]")
     expect(results[0].host).to    eq("ello.co")
     expect(results[0].user).to    eq("do-not-reply")
+    expect(results[0].original).to eq(line)
   end
 
   it "unparsable with mail gem (no whitespace before <)" do
@@ -219,6 +239,7 @@ describe MailAddress do
     expect(results[0].phrase).to  eq("大阪 太郎")
     expect(results[0].host).to    eq("example.jp")
     expect(results[0].user).to    eq("osaka")
+    expect(results[0].original).to eq(line)
   end
 
   it "local part only" do
@@ -231,6 +252,7 @@ describe MailAddress do
     expect(results[0].phrase).to  eq("localpartonly")
     expect(results[0].host).to    be_nil
     expect(results[0].user).to    eq("")
+    expect(results[0].original).to eq(line)
   end
 
   it "a lot of types of undisclosed recipients" do
@@ -246,12 +268,14 @@ describe MailAddress do
 
     array.each do |line|
       results = MailAddress.parse(line)
-      expect(results[0].format).to  eq(%Q("#{line}"))
+#      expect(results[0].format).to  eq(%Q("#{line}"))
+      expect(results[0].format).to eq(line)
       expect(results[0].address).to be_nil
       expect(results[0].name).to    eq(line)
       expect(results[0].phrase).to  eq(line)
       expect(results[0].host).to    be_nil
       expect(results[0].user).to    eq("")
+      expect(results[0].original).to eq(line)
     end
 
     array = [
@@ -267,9 +291,10 @@ describe MailAddress do
       expect(results[0].phrase).to  eq(line)
       expect(results[0].host).to    be_nil
       expect(results[0].user).to    eq("")
+      expect(results[0].original).to eq(line)
     end
 
-    # valid address
+    # a seemingly valid address
     line = '"Undisclosed" <"recipients:"@nifty.com>'
     results = MailAddress.parse(line)
     expect(results[0].format).to  eq(line)
@@ -278,6 +303,7 @@ describe MailAddress do
     expect(results[0].phrase).to  eq('"Undisclosed"')
     expect(results[0].host).to    eq("nifty.com")
     expect(results[0].user).to    eq('"recipients:"')
+    expect(results[0].original).to eq(line)
   end
 
   it "specify mime-encoded address" do
@@ -289,6 +315,7 @@ describe MailAddress do
     expect(results[0].phrase).to  eq("=?ISO-2022-JP?B?GyRCQmc6ZRsoQiAbJEJCQE86GyhC?=")
     expect(results[0].host).to    eq("example.jp")
     expect(results[0].user).to    eq("osaka")
+    expect(results[0].original).to eq(line)
   end
 
   it "obviously invalid address (has no '@')" do
@@ -299,16 +326,38 @@ describe MailAddress do
 
     array.each do |line|
       results = MailAddress.parse(line)
-      expect(results[0].format).to  eq(%Q("#{line}"))
+      expect(results[0].format).to  eq(line)
       expect(results[0].address).to be_nil
       expect(results[0].name).to    eq(line)
       expect(results[0].phrase).to  eq(line)
       expect(results[0].host).to    be_nil
       expect(results[0].user).to    eq("")
+      expect(results[0].original).to eq(line)
     end
-
   end
 
+  it "empty string or nil" do
+    # empty string
+    line = ''
+    results = MailAddress.parse(line)
+    expect(results[0].format).to eq('')
+    expect(results[0].address).to be_nil
+    expect(results[0].name).to be_nil
+    expect(results[0].phrase).to eq('')
+    expect(results[0].host).to be_nil
+    expect(results[0].user).to eq('')
+    expect(results[0].original).to eq(line)
+    # nil
+    line = nil
+    results = MailAddress.parse(line)
+    expect(results[0].format).to eq('')
+    expect(results[0].address).to be_nil
+    expect(results[0].name).to be_nil
+    expect(results[0].phrase).to eq('')
+    expect(results[0].host).to be_nil
+    expect(results[0].user).to eq('')
+    expect(results[0].original).to eq('') # Note that it returns empty string NOT nil
+  end
 
   it "corrupted address" do
     line = 'john <john@example.com' # lack of right angle bracket
@@ -322,7 +371,20 @@ describe MailAddress do
     }.to raise_error(StandardError)
   end
 
-  it 'Perl Module Pod test data' do
+  it "unbelievable but existed address" do
+    line = 'Sf 山田 太郎@example.com, valid@example.com'
+    results = MailAddress.parse(line)
+    expect(results[0].format).to eq('Sf 山田 太郎@example.com')
+    expect(results[0].address).not_to eq("Sf@example.com") ## important!
+    expect(results[0].address).to be_nil                   ## important!
+    expect(results[0].name).to eq('山田 太郎')             ## I don't care whatever returns
+    expect(results[0].phrase).to eq('山田 太郎')           ## I don't care whatever returns
+    expect(results[0].host).to be_nil
+    expect(results[0].user).to eq('')
+    expect(results[0].original).to eq('Sf 山田 太郎@example.com')
+  end
+
+  it 'Perl Module TAP test data' do
     data = [
       # [ '"Joe & J. Harvey" <ddd @Org>, JJV @ BBN',
       #   '"Joe & J. Harvey" <ddd@Org>',
@@ -422,9 +484,9 @@ describe MailAddress do
       ['Derek.Roskell%dero@msg.ti.com',
         'Derek.Roskell%dero@msg.ti.com',
         'Derek Roskell'],
-      ['":sysmail"@ Some-Group. Some-Org, Muhammed.(I am the greatest) Ali @(the)Vegas.WBA',
-        '":sysmail"@Some-Group.Some-Org',
-        ''],
+      # ['":sysmail"@ Some-Group. Some-Org, Muhammed.(I am the greatest) Ali @(the)Vegas.WBA',
+      #   '":sysmail"@Some-Group.Some-Org',
+      #   ''],
       # ["david d `zoo' zuhn <zoo@aggregate.com>",
       #   "david d `zoo' zuhn <zoo@aggregate.com>",
       #   "David D `Zoo' Zuhn"],

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: mail_address
 version: !ruby/object:Gem::Version
-  version: 1.0.0
+  version: 1.1.0
 platform: ruby
 authors:
 - Kizashi Nagata
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-04-15 00:00:00.000000000 Z
+date: 2015-04-16 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler