Added test proof (#110)

Fixes #91 and #74 for v1.2.x

.travis.yml

@@ -1,3 +1,5 @@
+sudo: false
+
 language: php
 
 php:
@@ -5,16 +7,29 @@ php:
   - 5.4
   - 5.5
   - 5.6
+  - 7.0
   - hhvm
 
-before_script:
- - wget http://getcomposer.org/composer.phar
- - php composer.phar install --dev --no-interaction
+env:
+  global:
+    - deps=high
+
+matrix:
+  fast_finish: true
+  include:
+    - php: 5.3
+      env: deps=low
+    - php: 5.4
+      env: deps=no
+    - php: 5.5
+      env: deps=no
+
+install:
+  - if [ "$deps" = "no" ]; then composer install; fi
+  - if [ "$deps" = "low" ]; then composer update --prefer-lowest; fi
+  - if [ "$deps" = "high" ]; then composer update; fi
 
 script:
  - mkdir -p build/logs
  - phpunit --coverage-clover build/logs/clover.xml
 
-after_script:
- - php vendor/bin/coveralls
-

README.md

@@ -1,51 +1,59 @@
-EmailValidator [![Build Status](https://travis-ci.org/egulias/EmailValidator.png?branch=master)](https://travis-ci.org/egulias/EmailValidator) [![Coverage Status](https://coveralls.io/repos/egulias/EmailValidator/badge.png?branch=master)](https://coveralls.io/r/egulias/EmailValidator?branch=master) [![SensioLabsInsight](https://insight.sensiolabs.com/projects/b18d473e-bd5a-4d88-a7b2-7aeaee0ebd7b/small.png)](https://insight.sensiolabs.com/projects/b18d473e-bd5a-4d88-a7b2-7aeaee0ebd7b)[![Scrutinizer Code Quality](https://scrutinizer-ci.com/g/egulias/EmailValidator/badges/quality-score.png?b=master)](https://scrutinizer-ci.com/g/egulias/EmailValidator/?branch=master)
-=============================
-
-##Installation##
-Install via composer. Add to your current compooser.json ```require``` key: ```"egulias/email-validator":"1.0.x-dev" ```
-
-##Usage##
-
-Simple example:
-
-```php
-<?php
-
-use Egulias\EmailValidator\EmailValidator;
-
-$validator = new EmailValidator;
-if ($validator->isValid($email)) {
-	echo $email . ' is a valid email address';
-}
-```
-
-More advanced example (returns detailed diagnostic error codes):
-
-```php
-<?php
-
-use egulias\EmailValidator\EmailValidator;
-
-$validator = new EmailValidator;
-$email = 'dominic@sayers.cc';
-$result = $validator->isValid($email);
-
-if ($result) {
-	echo $email . ' is a valid email address';
-} else if ($validator->hasWarnings()) {
-	echo 'Warning! ' . $email . ' has unusual/deprecated features (result code ' . var_export($validator->getWarnings(), true) . ')';
-} else {
-	echo $email . ' is not a valid email address (result code ' . $validator->getError() . ')';
-}
-```
-
-##Contributors##
-As this is a port from another library and work, here are other people related to the previous:
-
-* Ricard Clau [@ricardclau](http://github.com/ricardclau):      	Performance against PHP built-in filter_var
-* Josepf Bielawski [@stloyd](http://github.com/stloyd):      		For its first re-work of Dominic's lib
-* Dominic Sayers [@dominicsayers](http://github.com/dominicsayers):  	The original isemail function
-
-##Licence##
-Released under the MIT Licence attached with this code.
-
+#EmailValidator
+[![Build Status](https://travis-ci.org/egulias/EmailValidator.png?branch=master)](https://travis-ci.org/egulias/EmailValidator) [![Coverage Status](https://coveralls.io/repos/egulias/EmailValidator/badge.png?branch=master)](https://coveralls.io/r/egulias/EmailValidator?branch=master) [![Scrutinizer Code Quality](https://scrutinizer-ci.com/g/egulias/EmailValidator/badges/quality-score.png?b=master)](https://scrutinizer-ci.com/g/egulias/EmailValidator/?branch=master) [![SensioLabsInsight](https://insight.sensiolabs.com/projects/22ba6692-9c02-42e5-a65d-1c5696bfffc6/small.png)](https://insight.sensiolabs.com/projects/22ba6692-9c02-42e5-a65d-1c5696bfffc6)
+=============================
+With the help of
+
+![Powered by PhpStorm](https://www.jetbrains.com/phpstorm/documentation/docs/logo_phpstorm.png)
+##Installation##
+
+Run the command below to install via Composer
+
+```shell
+composer require egulias/email-validator "~1.2"
+```
+
+##Usage##
+
+Simple example:
+
+```php
+<?php
+
+use Egulias\EmailValidator\EmailValidator;
+
+$validator = new EmailValidator;
+if ($validator->isValid($email)) {
+	echo $email . ' is a valid email address';
+}
+```
+
+More advanced example (returns detailed diagnostic error codes):
+
+```php
+<?php
+
+use Egulias\EmailValidator\EmailValidator;
+
+$validator = new EmailValidator;
+$email = 'dominic@sayers.cc';
+$result = $validator->isValid($email);
+
+if ($result) {
+	echo $email . ' is a valid email address';
+} else if ($validator->hasWarnings()) {
+	echo 'Warning! ' . $email . ' has unusual/deprecated features (result code ' . var_export($validator->getWarnings(), true) . ')';
+} else {
+	echo $email . ' is not a valid email address (result code ' . $validator->getError() . ')';
+}
+```
+
+##Contributors##
+As this is a port from another library and work, here are other people related to the previous:
+
+* Ricard Clau [@ricardclau](http://github.com/ricardclau):      	Performance against PHP built-in filter_var
+* Josepf Bielawski [@stloyd](http://github.com/stloyd):      		For its first re-work of Dominic's lib
+* Dominic Sayers [@dominicsayers](http://github.com/dominicsayers):  	The original isemail function
+
+##License##
+Released under the MIT License attached with this code.
+

composer.json

@@ -3,22 +3,22 @@
   "description":  "A library for validating emails",
   "homepage":     "https://github.com/egulias/EmailValidator",
   "type":         "Library",
-  "keywords":     ["email", "validation", "validator"],
+  "keywords":     ["email", "validation", "validator", "emailvalidation", "emailvalidator"],
   "license":      "MIT",
   "authors": [
     {"name": "Eduardo Gulias Davis"}
   ],
   "extra": {
     "branch-alias": {
-      "dev-master": "1.0.x-dev"
+      "dev-master": "2.0.x-dev"
     }
   },
   "require":      {
     "php": ">= 5.3.3",
-    "doctrine/lexer": "~1.0"
+    "doctrine/lexer": "^1.0.1"
   },
   "require-dev" :   {
-    "satooshi/php-coveralls": "dev-master"
+    "phpunit/phpunit": "^4.8.24"
   },
   "autoload": {
     "psr-0": {

documentation/RFC5321BNF.html

@@ -1,89 +1,89 @@
-<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
-<html xmlns="http://www.w3.org/1999/xhtml">
-
-<head>
-<meta content="text/html; charset=utf-8" http-equiv="Content-Type" />
-<title>The BNF from RFC 5321 defining parts of a valid SMTP address</title>
-</head>
-
-<body>
-<pre style="font-size:11px;">
-   Mailbox        = Local-part "@" ( Domain / address-literal )
-
-   Local-part     = Dot-string / Quoted-string
-                  ; MAY be case-sensitive
-
-
-   Dot-string     = Atom *("."  Atom)
-
-   Atom           = 1*atext
-
-   Quoted-string  = DQUOTE *QcontentSMTP DQUOTE
-
-   QcontentSMTP   = qtextSMTP / quoted-pairSMTP
-
-   quoted-pairSMTP  = %d92 %d32-126
-                    ; i.e., backslash followed by any ASCII
-                    ; graphic (including itself) or SPace
-
-   qtextSMTP      = %d32-33 / %d35-91 / %d93-126
-                  ; i.e., within a quoted string, any
-                  ; ASCII graphic or space is permitted
-                  ; without blackslash-quoting except
-                  ; double-quote and the backslash itself.
-
-   Domain         = sub-domain *("." sub-domain)
-
-   sub-domain     = Let-dig [Ldh-str]
-
-   Let-dig        = ALPHA / DIGIT
-
-   Ldh-str        = *( ALPHA / DIGIT / "-" ) Let-dig
-
-   address-literal  = "[" ( IPv4-address-literal /
-                    IPv6-address-literal /
-                    General-address-literal ) "]"
-                    ; See Section 4.1.3
-
-   IPv4-address-literal  = Snum 3("."  Snum)
-
-   IPv6-address-literal  = "IPv6:" IPv6-addr
-
-   General-address-literal  = Standardized-tag ":" 1*dcontent
-
-   Standardized-tag  = Ldh-str
-                     ; Standardized-tag MUST be specified in a
-                     ; Standards-Track RFC and registered with IANA
-
-   dcontent       = %d33-90 / ; Printable US-ASCII
-                  %d94-126 ; excl. "[", "\", "]"
-
-   Snum           = 1*3DIGIT
-                  ; representing a decimal integer
-                  ; value in the range 0 through 255
-
-   IPv6-addr      = IPv6-full / IPv6-comp / IPv6v4-full / IPv6v4-comp
-
-   IPv6-hex       = 1*4HEXDIG
-
-   IPv6-full      = IPv6-hex 7(":" IPv6-hex)
-
-   IPv6-comp      = [IPv6-hex *5(":" IPv6-hex)] "::"
-                  [IPv6-hex *5(":" IPv6-hex)]
-                  ; The "::" represents at least 2 16-bit groups of
-                  ; zeros.  No more than 6 groups in addition to the
-                  ; "::" may be present.
-
-   IPv6v4-full    = IPv6-hex 5(":" IPv6-hex) ":" IPv4-address-literal
-
-   IPv6v4-comp    = [IPv6-hex *3(":" IPv6-hex)] "::"
-                  [IPv6-hex *3(":" IPv6-hex) ":"]
-                  IPv4-address-literal
-                  ; The "::" represents at least 2 16-bit groups of
-                  ; zeros.  No more than 4 groups in addition to the
-                  ; "::" and IPv4-address-literal may be present.
-
-</pre>
-</body>
-
-</html>
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml">
+
+<head>
+<meta content="text/html; charset=utf-8" http-equiv="Content-Type" />
+<title>The BNF from RFC 5321 defining parts of a valid SMTP address</title>
+</head>
+
+<body>
+<pre style="font-size:11px;">
+   Mailbox        = Local-part "@" ( Domain / address-literal )
+
+   Local-part     = Dot-string / Quoted-string
+                  ; MAY be case-sensitive
+
+
+   Dot-string     = Atom *("."  Atom)
+
+   Atom           = 1*atext
+
+   Quoted-string  = DQUOTE *QcontentSMTP DQUOTE
+
+   QcontentSMTP   = qtextSMTP / quoted-pairSMTP
+
+   quoted-pairSMTP  = %d92 %d32-126
+                    ; i.e., backslash followed by any ASCII
+                    ; graphic (including itself) or SPace
+
+   qtextSMTP      = %d32-33 / %d35-91 / %d93-126
+                  ; i.e., within a quoted string, any
+                  ; ASCII graphic or space is permitted
+                  ; without blackslash-quoting except
+                  ; double-quote and the backslash itself.
+
+   Domain         = sub-domain *("." sub-domain)
+
+   sub-domain     = Let-dig [Ldh-str]
+
+   Let-dig        = ALPHA / DIGIT
+
+   Ldh-str        = *( ALPHA / DIGIT / "-" ) Let-dig
+
+   address-literal  = "[" ( IPv4-address-literal /
+                    IPv6-address-literal /
+                    General-address-literal ) "]"
+                    ; See Section 4.1.3
+
+   IPv4-address-literal  = Snum 3("."  Snum)
+
+   IPv6-address-literal  = "IPv6:" IPv6-addr
+
+   General-address-literal  = Standardized-tag ":" 1*dcontent
+
+   Standardized-tag  = Ldh-str
+                     ; Standardized-tag MUST be specified in a
+                     ; Standards-Track RFC and registered with IANA
+
+   dcontent       = %d33-90 / ; Printable US-ASCII
+                  %d94-126 ; excl. "[", "\", "]"
+
+   Snum           = 1*3DIGIT
+                  ; representing a decimal integer
+                  ; value in the range 0 through 255
+
+   IPv6-addr      = IPv6-full / IPv6-comp / IPv6v4-full / IPv6v4-comp
+
+   IPv6-hex       = 1*4HEXDIG
+
+   IPv6-full      = IPv6-hex 7(":" IPv6-hex)
+
+   IPv6-comp      = [IPv6-hex *5(":" IPv6-hex)] "::"
+                  [IPv6-hex *5(":" IPv6-hex)]
+                  ; The "::" represents at least 2 16-bit groups of
+                  ; zeros.  No more than 6 groups in addition to the
+                  ; "::" may be present.
+
+   IPv6v4-full    = IPv6-hex 5(":" IPv6-hex) ":" IPv4-address-literal
+
+   IPv6v4-comp    = [IPv6-hex *3(":" IPv6-hex)] "::"
+                  [IPv6-hex *3(":" IPv6-hex) ":"]
+                  IPv4-address-literal
+                  ; The "::" represents at least 2 16-bit groups of
+                  ; zeros.  No more than 4 groups in addition to the
+                  ; "::" and IPv4-address-literal may be present.
+
+</pre>
+</body>
+
+</html>

documentation/RFC5322BNF.html

@@ -1,141 +1,141 @@
-<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
-<html xmlns="http://www.w3.org/1999/xhtml">
-
-<head>
-<meta content="text/html; charset=utf-8" http-equiv="Content-Type" />
-<title>The BNF from RFC 5322 defining parts of a valid internet message address</title>
-</head>
-
-<body>
-<pre style="font-size:11px;">
-   addr-spec       =   local-part "@" domain
-
-   local-part      =   dot-atom / quoted-string / obs-local-part
-
-   dot-atom        =   [CFWS] dot-atom-text [CFWS]
-
-   CFWS            =   (1*([FWS] comment) [FWS]) / FWS
-
-   FWS             =   ([*WSP CRLF] 1*WSP) /  obs-FWS
-                                          ; Folding white space
-
-   WSP             =   SP / HTAB          ; white space
-
-   obs-FWS         =   1*([CRLF] WSP)     ; As amended in erratum #1908
-
-   ctext           =   %d33-39 /          ; Printable US-ASCII
-                       %d42-91 /          ;  characters not including
-                       %d93-126 /         ;  "(", ")", or "\"
-                       obs-ctext
-
-   obs-ctext       =   obs-NO-WS-CTL
-   ccontent        =   ctext / quoted-pair / comment
-
-   comment         =   "(" *([FWS] ccontent) [FWS] ")"
-
-   dot-atom-text   =   1*atext *("." 1*atext)
-
-   atext           =   ALPHA / DIGIT /    ; Printable US-ASCII
-                       "!" / "#" /        ;  characters not including
-                       "$" / "%" /        ;  specials.  Used for atoms.
-                       "&amp;" / "'" /
-                       "*" / "+" /
-                       "-" / "/" /
-                       "=" / "?" /
-                       "^" / "_" /
-                       "`" / "{" /
-                       "|" / "}" /
-                       "~"
-
-   specials        =   "(" / ")" /        ; Special characters that do
-                       "&lt;" / "&gt;" /        ;  not appear in atext
-                       "[" / "]" /
-                       ":" / ";" /
-                       "@" / "\" /
-                       "," / "." /
-                       DQUOTE
-
-   quoted-string   =   [CFWS]
-                       DQUOTE *([FWS] qcontent) [FWS] DQUOTE
-                       [CFWS]
-
-   qcontent        =   qtext / quoted-pair
-
-   qtext           =   %d33 /             ; Printable US-ASCII
-                       %d35-91 /          ;  characters not including
-                       %d93-126 /         ;  "\" or the quote character
-                       obs-qtext
-
-   obs-qtext       =   obs-NO-WS-CTL
-
-   obs-NO-WS-CTL   =   %d1-8 /            ; US-ASCII control
-                       %d11 /             ;  characters that do not
-                       %d12 /             ;  include the carriage
-                       %d14-31 /          ;  return, line feed, and
-                       %d127              ;  white space characters
-
-   quoted-pair     =   ("\" (VCHAR / WSP)) / obs-qp
-
-   VCHAR           =   %x21-7E            ; visible (printing) characters
-
-   obs-qp          =   "\" (%d0 / obs-NO-WS-CTL / LF / CR)
-
-   obs-local-part  =   word *("." word)
-
-   word            =   atom / quoted-string
-
-   atom            =   [CFWS] 1*atext [CFWS]
-
-   domain          =   dot-atom / domain-literal / obs-domain
-
-   domain-literal  =   [CFWS] "[" *([FWS] dtext) [FWS] "]" [CFWS]
-
-   dtext           =   %d33-90 /          ; Printable US-ASCII
-                       %d94-126 /         ;  characters not including
-                       obs-dtext          ;  "[", "]", or "\"
-
-   obs-dtext       =   obs-NO-WS-CTL / quoted-pair
-
-   obs-domain      =   atom *("." atom)
-
-NB For SMTP mail, the domain-literal is restricted by RFC5321 as follows:
-
-   Mailbox        = Local-part "@" ( Domain / address-literal )
-
-   address-literal  = "[" ( IPv4-address-literal /
-                    IPv6-address-literal /
-                    General-address-literal ) "]"
-
-   IPv4-address-literal  = Snum 3("."  Snum)
-
-   IPv6-address-literal  = "IPv6:" IPv6-addr
-
-   Snum           = 1*3DIGIT
-                  ; representing a decimal integer
-                  ; value in the range 0 through 255
-
-   IPv6-addr      = IPv6-full / IPv6-comp / IPv6v4-full / IPv6v4-comp
-
-   IPv6-hex       = 1*4HEXDIG
-
-   IPv6-full      = IPv6-hex 7(":" IPv6-hex)
-
-   IPv6-comp      = [IPv6-hex *5(":" IPv6-hex)] "::"
-                  [IPv6-hex *5(":" IPv6-hex)]
-                  ; The "::" represents at least 2 16-bit groups of
-                  ; zeros.  No more than 6 groups in addition to the
-                  ; "::" may be present.
-
-   IPv6v4-full    = IPv6-hex 5(":" IPv6-hex) ":" IPv4-address-literal
-
-   IPv6v4-comp    = [IPv6-hex *3(":" IPv6-hex)] "::"
-                  [IPv6-hex *3(":" IPv6-hex) ":"]
-                  IPv4-address-literal
-                  ; The "::" represents at least 2 16-bit groups of
-                  ; zeros.  No more than 4 groups in addition to the
-                  ; "::" and IPv4-address-literal may be present.
-
-</pre>
-</body>
-
-</html>
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml">
+
+<head>
+<meta content="text/html; charset=utf-8" http-equiv="Content-Type" />
+<title>The BNF from RFC 5322 defining parts of a valid internet message address</title>
+</head>
+
+<body>
+<pre style="font-size:11px;">
+   addr-spec       =   local-part "@" domain
+
+   local-part      =   dot-atom / quoted-string / obs-local-part
+
+   dot-atom        =   [CFWS] dot-atom-text [CFWS]
+
+   CFWS            =   (1*([FWS] comment) [FWS]) / FWS
+
+   FWS             =   ([*WSP CRLF] 1*WSP) /  obs-FWS
+                                          ; Folding white space
+
+   WSP             =   SP / HTAB          ; white space
+
+   obs-FWS         =   1*([CRLF] WSP)     ; As amended in erratum #1908
+
+   ctext           =   %d33-39 /          ; Printable US-ASCII
+                       %d42-91 /          ;  characters not including
+                       %d93-126 /         ;  "(", ")", or "\"
+                       obs-ctext
+
+   obs-ctext       =   obs-NO-WS-CTL
+   ccontent        =   ctext / quoted-pair / comment
+
+   comment         =   "(" *([FWS] ccontent) [FWS] ")"
+
+   dot-atom-text   =   1*atext *("." 1*atext)
+
+   atext           =   ALPHA / DIGIT /    ; Printable US-ASCII
+                       "!" / "#" /        ;  characters not including
+                       "$" / "%" /        ;  specials.  Used for atoms.
+                       "&amp;" / "'" /
+                       "*" / "+" /
+                       "-" / "/" /
+                       "=" / "?" /
+                       "^" / "_" /
+                       "`" / "{" /
+                       "|" / "}" /
+                       "~"
+
+   specials        =   "(" / ")" /        ; Special characters that do
+                       "&lt;" / "&gt;" /        ;  not appear in atext
+                       "[" / "]" /
+                       ":" / ";" /
+                       "@" / "\" /
+                       "," / "." /
+                       DQUOTE
+
+   quoted-string   =   [CFWS]
+                       DQUOTE *([FWS] qcontent) [FWS] DQUOTE
+                       [CFWS]
+
+   qcontent        =   qtext / quoted-pair
+
+   qtext           =   %d33 /             ; Printable US-ASCII
+                       %d35-91 /          ;  characters not including
+                       %d93-126 /         ;  "\" or the quote character
+                       obs-qtext
+
+   obs-qtext       =   obs-NO-WS-CTL
+
+   obs-NO-WS-CTL   =   %d1-8 /            ; US-ASCII control
+                       %d11 /             ;  characters that do not
+                       %d12 /             ;  include the carriage
+                       %d14-31 /          ;  return, line feed, and
+                       %d127              ;  white space characters
+
+   quoted-pair     =   ("\" (VCHAR / WSP)) / obs-qp
+
+   VCHAR           =   %x21-7E            ; visible (printing) characters
+
+   obs-qp          =   "\" (%d0 / obs-NO-WS-CTL / LF / CR)
+
+   obs-local-part  =   word *("." word)
+
+   word            =   atom / quoted-string
+
+   atom            =   [CFWS] 1*atext [CFWS]
+
+   domain          =   dot-atom / domain-literal / obs-domain
+
+   domain-literal  =   [CFWS] "[" *([FWS] dtext) [FWS] "]" [CFWS]
+
+   dtext           =   %d33-90 /          ; Printable US-ASCII
+                       %d94-126 /         ;  characters not including
+                       obs-dtext          ;  "[", "]", or "\"
+
+   obs-dtext       =   obs-NO-WS-CTL / quoted-pair
+
+   obs-domain      =   atom *("." atom)
+
+NB For SMTP mail, the domain-literal is restricted by RFC5321 as follows:
+
+   Mailbox        = Local-part "@" ( Domain / address-literal )
+
+   address-literal  = "[" ( IPv4-address-literal /
+                    IPv6-address-literal /
+                    General-address-literal ) "]"
+
+   IPv4-address-literal  = Snum 3("."  Snum)
+
+   IPv6-address-literal  = "IPv6:" IPv6-addr
+
+   Snum           = 1*3DIGIT
+                  ; representing a decimal integer
+                  ; value in the range 0 through 255
+
+   IPv6-addr      = IPv6-full / IPv6-comp / IPv6v4-full / IPv6v4-comp
+
+   IPv6-hex       = 1*4HEXDIG
+
+   IPv6-full      = IPv6-hex 7(":" IPv6-hex)
+
+   IPv6-comp      = [IPv6-hex *5(":" IPv6-hex)] "::"
+                  [IPv6-hex *5(":" IPv6-hex)]
+                  ; The "::" represents at least 2 16-bit groups of
+                  ; zeros.  No more than 6 groups in addition to the
+                  ; "::" may be present.
+
+   IPv6v4-full    = IPv6-hex 5(":" IPv6-hex) ":" IPv4-address-literal
+
+   IPv6v4-comp    = [IPv6-hex *3(":" IPv6-hex)] "::"
+                  [IPv6-hex *3(":" IPv6-hex) ":"]
+                  IPv4-address-literal
+                  ; The "::" represents at least 2 16-bit groups of
+                  ; zeros.  No more than 4 groups in addition to the
+                  ; "::" and IPv4-address-literal may be present.
+
+</pre>
+</body>
+
+</html>

src/Egulias/EmailValidator/EmailLexer.php

@@ -31,10 +31,13 @@ class EmailLexer extends AbstractLexer
     const S_SEMICOLON        = 275;
     const S_OPENQBRACKET     = 276;
     const S_CLOSEQBRACKET    = 277;
+    const S_SLASH            = 278;
     const S_EMPTY            = null;
     const GENERIC            = 300;
     const CRLF               = 301;
     const INVALID            = 302;
+    const ASCII_INVALID_FROM = 127;
+    const ASCII_INVALID_TO   = 199;
 
     /**
      * US-ASCII visible characters not valid for atext (@link http://tools.ietf.org/html/rfc5322#section-3.2.3)
@@ -52,6 +55,7 @@ class EmailLexer extends AbstractLexer
         ';'    => self::S_SEMICOLON,
         '@'    => self::S_AT,
         '\\'   => self::S_BACKSLASH,
+        '/'    => self::S_SLASH,
         ','    => self::S_COMMA,
         '.'    => self::S_DOT,
         '"'    => self::S_DQUOTE,
@@ -63,18 +67,31 @@ class EmailLexer extends AbstractLexer
         "\n"   => self::S_LF,
         "\r\n" => self::CRLF,
         'IPv6' => self::S_IPV6TAG,
-        '<'    => self::S_LOWERTHAN,
-        '>'    => self::S_GREATERTHAN,
         '{'    => self::S_OPENQBRACKET,
         '}'    => self::S_CLOSEQBRACKET,
-        ''     => self::S_EMPTY
+        ''     => self::S_EMPTY,
+        '\0'   => self::C_NUL,
     );
 
+    protected $hasInvalidTokens = false;
+
     protected $previous;
 
+    public function reset()
+    {
+        $this->hasInvalidTokens = false;
+        parent::reset();
+    }
+
+    public function hasInvalidTokens()
+    {
+        return $this->hasInvalidTokens;
+    }
+
     /**
      * @param $type
      * @throws \UnexpectedValueException
+     * @return boolean
      */
     public function find($type)
     {
@@ -84,6 +101,7 @@ class EmailLexer extends AbstractLexer
         if (!$search->lookahead) {
             throw new \UnexpectedValueException($type . ' not found');
         }
+        return true;
     }
 
     /**
@@ -99,7 +117,7 @@ class EmailLexer extends AbstractLexer
     /**
      * moveNext
      *
-     * @return mixed
+     * @return boolean
      */
     public function moveNext()
     {
@@ -111,29 +129,29 @@ class EmailLexer extends AbstractLexer
     /**
      * Lexical catchable patterns.
      *
-     * @return array
+     * @return string[]
      */
     protected function getCatchablePatterns()
     {
         return array(
-            '[a-zA-Z_]+[4,6]?',
+            '[a-zA-Z_]+[46]?', //ASCII and domain literal
+            '[^\x00-\x7F]',  //UTF-8
             '[0-9]+',
             '\r\n',
             '::',
-            '\s+',
-            '[\x1-\x1F]+',
-            '.'
+            '\s+?',
+            '.',
             );
     }
 
     /**
      * Lexical non-catchable patterns.
      *
-     * @return array
+     * @return string[]
      */
     protected function getNonCatchablePatterns()
     {
-        return array('[\x7f-\xff]+');
+        return array('[\xA0-\xff]+');
     }
 
     /**
@@ -145,14 +163,59 @@ class EmailLexer extends AbstractLexer
      */
     protected function getType(&$value)
     {
-        if (isset($this->charValue[$value])) {
+        if ($this->isNullType($value)) {
+            return self::C_NUL;
+        }
+
+        if ($this->isValid($value)) {
             return $this->charValue[$value];
         }
 
-        if (preg_match('/[\x1-\x1F]+/', $value)) {
+        if ($this->isUTF8Invalid($value)) {
+            $this->hasInvalidTokens = true;
             return self::INVALID;
         }
 
         return  self::GENERIC;
     }
+
+    protected function isValid($value)
+    {
+        if (isset($this->charValue[$value])) {
+            return true;
+        }
+
+        return false;
+    }
+
+    /**
+     * @param $value
+     * @return bool
+     */
+    protected function isNullType($value)
+    {
+        if ($value === "\0") {
+            return true;
+        }
+
+        return false;
+    }
+
+    /**
+     * @param $value
+     * @return bool
+     */
+    protected function isUTF8Invalid($value)
+    {
+        if (preg_match('/\p{Cc}+/u', $value)) {
+            return true;
+        }
+
+        return false;
+    }
+
+    protected function getModifiers()
+    {
+        return 'iu';
+    }
 }

src/Egulias/EmailValidator/EmailParser.php

@@ -28,6 +28,10 @@ class EmailParser
         $this->domainPartParser = new DomainPart($this->lexer);
     }
 
+    /**
+     * @param $str
+     * @return array
+     */
     public function parse($str)
     {
         $this->lexer->setInput($str);
@@ -36,11 +40,16 @@ class EmailParser
             throw new \InvalidArgumentException('ERR_NOLOCALPART');
         }
 
+
         $this->localPartParser->parse($str);
         $this->domainPartParser->parse($str);
 
         $this->setParts($str);
 
+        if ($this->lexer->hasInvalidTokens()) {
+            throw new \InvalidArgumentException('ERR_INVALID_ATEXT');
+        }
+
         return array('local' => $this->localPart, 'domain' => $this->domainPart);
     }
 
@@ -78,6 +87,10 @@ class EmailParser
         return true;
     }
 
+    /**
+     * @param string $localPart
+     * @param string $parsedDomainPart
+     */
     protected function addLongEmailWarning($localPart, $parsedDomainPart)
     {
         if (strlen($localPart . '@' . $parsedDomainPart) > self::EMAIL_MAX_LENGTH) {

src/Egulias/EmailValidator/EmailValidator.php

@@ -2,8 +2,6 @@
 
 namespace Egulias\EmailValidator;
 
-use Egulias\EmailValidator\Parser\LocalPart;
-
 /**
  * EmailValidator
  *
@@ -11,47 +9,50 @@ use Egulias\EmailValidator\Parser\LocalPart;
  */
 class EmailValidator
 {
+    /**
+     * Critical validation errors used to indicate that
+     * an email address is invalid:
+     */
     const ERR_CONSECUTIVEATS     = 128;
     const ERR_EXPECTING_DTEXT    = 129;
     const ERR_NOLOCALPART        = 130;
     const ERR_NODOMAIN           = 131;
     const ERR_CONSECUTIVEDOTS    = 132;
     const ERR_ATEXT_AFTER_CFWS   = 133;
-    const ERR_ATEXT_AFTER_QS     = 134;
-    const ERR_ATEXT_AFTER_DOMLIT = 135;
     const ERR_EXPECTING_QPAIR    = 136;
     const ERR_EXPECTING_ATEXT    = 137;
-    const ERR_EXPECTING_QTEXT    = 138;
     const ERR_EXPECTING_CTEXT    = 139;
-    const ERR_BACKSLASHEND       = 140;
     const ERR_DOT_START          = 141;
     const ERR_DOT_END            = 142;
-    const ERR_DOMAINHYPHENSTART  = 143;
     const ERR_DOMAINHYPHENEND    = 144;
     const ERR_UNCLOSEDQUOTEDSTR  = 145;
     const ERR_UNCLOSEDCOMMENT    = 146;
-    const ERR_UNCLOSEDDOMLIT     = 147;
     const ERR_FWS_CRLF_X2        = 148;
     const ERR_FWS_CRLF_END       = 149;
     const ERR_CR_NO_LF           = 150;
     const ERR_DEPREC_REACHED     = 151;
+    const ERR_UNOPENEDCOMMENT    = 152;
+    const ERR_ATEXT_AFTER_QS     = 134; // not in use
+    const ERR_ATEXT_AFTER_DOMLIT = 135; // not in use
+    const ERR_EXPECTING_QTEXT    = 138; // not in use
+    const ERR_BACKSLASHEND       = 140; // not in use
+    const ERR_DOMAINHYPHENSTART  = 143; // not in use
+    const ERR_UNCLOSEDDOMLIT     = 147; // not in use
+
+    /**
+     * Informational validation warnings regarding unusual or
+     * deprecated features found in an email address:
+     */
+    // Address is valid for SMTP (RFC-5321), but has unusual elements.
     const RFC5321_TLD             = 9;
-    const RFC5321_TLDNUMERIC      = 10;
     const RFC5321_QUOTEDSTRING    = 11;
     const RFC5321_ADDRESSLITERAL  = 12;
     const RFC5321_IPV6DEPRECATED  = 13;
-    const CFWS_COMMENT            = 17;
-    const CFWS_FWS                = 18;
-    const DEPREC_LOCALPART        = 33;
-    const DEPREC_FWS              = 34;
-    const DEPREC_QTEXT            = 35;
-    const DEPREC_QP               = 36;
-    const DEPREC_COMMENT          = 37;
-    const DEPREC_CTEXT            = 38;
-    const DEPREC_CFWS_NEAR_AT     = 49;
+    const RFC5321_TLDNUMERIC      = 10; // not in use
+    // Address is only valid according to the broad
+    // definition of RFC-5322. It is otherwise invalid.
     const RFC5322_LOCAL_TOOLONG   = 64;
     const RFC5322_LABEL_TOOLONG   = 63;
-    const RFC5322_DOMAIN          = 65;
     const RFC5322_TOOLONG         = 66;
     const RFC5322_DOMAIN_TOOLONG  = 255;
     const RFC5322_DOMAINLITERAL   = 70;
@@ -62,12 +63,48 @@ class EmailValidator
     const RFC5322_IPV6_MAXGRPS    = 75;
     const RFC5322_IPV6_COLONSTRT  = 76;
     const RFC5322_IPV6_COLONEND   = 77;
+    const RFC5322_DOMAIN          = 65; // not in use
+    // Address contains deprecated elements, but may
+    // still be valid in restricted contexts.
+    const DEPREC_QP               = 36;
+    const DEPREC_COMMENT          = 37;
+    const DEPREC_CFWS_NEAR_AT     = 49;
+    const DEPREC_LOCALPART        = 33; // not in use
+    const DEPREC_FWS              = 34; // not in use
+    const DEPREC_QTEXT            = 35; // not in use
+    const DEPREC_CTEXT            = 38; // not in use
+    // Address is valid within the message,
+    // but cannot be used unmodified in the envelope.
+    const CFWS_COMMENT            = 17;
+    const CFWS_FWS                = 18;
+    // Hostname DNS checks were unsuccessful.
     const DNSWARN_NO_MX_RECORD    = 5;
     const DNSWARN_NO_RECORD       = 6;
 
+    /**
+     * @var EmailParser
+     */
     protected $parser;
+
+    /**
+     * Contains any informational warnings regarding unusual/deprecated
+     * features that were encountered during validation.
+     *
+     * @var array
+     */
     protected $warnings = array();
+
+    /**
+     * If a critical validation problem is encountered, this will be
+     * set to the value of one of this class's ERR_* constants.
+     *
+     * @var int
+     */
     protected $error;
+
+    /**
+     * @var int
+     */
     protected $threshold = 255;
 
     public function __construct()
@@ -75,6 +112,28 @@ class EmailValidator
         $this->parser = new EmailParser(new EmailLexer());
     }
 
+    /**
+     * Validates an email address against the following standards:
+     *
+     * RFC-5321: Simple Mail Transfer Protocol
+     * RFC-5322: Internet Message Format
+     * RFC-6530: Overview and Framework for Internationalized Email
+     * RFC-6531: SMTP Extension for Internationalized Email
+     * RFC-6532: Internationalized Email Headers
+     * RFC-1123 section 2.1: Requirements for Internet Hosts -- Application and Support
+     * RFC-4291 section 2.2: IP Version 6 Addressing Architecture
+     *
+     * @param string $email    The email address to validate.
+     * @param bool   $checkDNS Whether or not the email address's hostname should
+     *                         be confirmed with a DNS lookup. This only comes
+     *                         into play if strict mode is also enabled.
+     * @param bool   $strict   If this is true, and any informational warnings
+     *                         were raised during validation, the email address
+     *                         will be considered invalid. Additionally, if
+     *                         $checkDNS is true and the DNS lookup failed,
+     *                         the email address will be considered invalid.
+     * @return bool
+     */
     public function isValid($email, $checkDNS = false, $strict = false)
     {
         try {
@@ -86,18 +145,14 @@ class EmailValidator
             return false;
         }
 
-        $dns = false;
-        if ($checkDNS) {
-            $dns = $this->checkDNS();
-        }
+        $dnsProblemExists = ($checkDNS ? !$this->checkDNS() : false);
 
         if ($this->hasWarnings() && ((int) max($this->warnings) > $this->threshold)) {
             $this->error = self::ERR_DEPREC_REACHED;
-
             return false;
         }
 
-        return ($strict) ? (!$this->hasWarnings() && $dns) : true;
+        return ($strict ? (!$this->hasWarnings() && !$dnsProblemExists) : true);
     }
 
     /**
@@ -144,26 +199,20 @@ class EmailValidator
         return $this->threshold;
     }
 
+    /**
+     * @return bool Whether or not an MX record exists for the
+     *              email address's host name.
+     */
     protected function checkDNS()
     {
-        $checked = false;
-        if (!function_exists('dns_get_record') && (
-            in_array(self::DNSWARN_NO_RECORD, $this->warnings) &&
-            in_array(self::DNSWARN_NO_MX_RECORD, $this->warnings)
-        )) {
-            return $checked;
-        }
+        $mxRecordExists = checkdnsrr(trim($this->parser->getParsedDomainPart()), 'MX');
 
-        $result = checkdnsrr(trim($this->parser->getParsedDomainPart()), 'MX');
-        $checked = true;
-
-        if (!$result) {
+        if (!$mxRecordExists) {
             $this->warnings[] = self::DNSWARN_NO_RECORD;
-            $checked = false;
             $this->addTLDWarnings();
         }
 
-        return $checked;
+        return $mxRecordExists;
     }
 
     protected function addTLDWarnings()

src/Egulias/EmailValidator/Parser/DomainPart.php

@@ -4,7 +4,6 @@
 namespace Egulias\EmailValidator\Parser;
 
 use Egulias\EmailValidator\EmailLexer;
-use Egulias\EmailValidator\Parser\Parser;
 use Egulias\EmailValidator\EmailValidator;
 
 class DomainPart extends Parser
@@ -23,6 +22,9 @@ class DomainPart extends Parser
         if ($this->lexer->token['type'] === EmailLexer::S_EMPTY) {
             throw new \InvalidArgumentException('ERR_NODOMAIN');
         }
+        if ($this->lexer->token['type'] === EmailLexer::S_HYPHEN) {
+            throw new \InvalidArgumentException('ERR_DOMAINHYPHENEND');
+        }
 
         if ($this->lexer->token['type'] === EmailLexer::S_OPENPARENTHESIS) {
             $this->warnings[] = EmailValidator::DEPREC_COMMENT;
@@ -100,18 +102,34 @@ class DomainPart extends Parser
     protected function doParseDomainPart()
     {
         $domain = '';
+        $openedParenthesis = 0;
+        $openBrackets = false;
         do {
             $prev = $this->lexer->getPrevious();
 
+            $this->checkNotAllowedChars($this->lexer->token);
+
             if ($this->lexer->token['type'] === EmailLexer::S_OPENPARENTHESIS) {
                 $this->parseComments();
+                $openedParenthesis += $this->getOpenedParenthesis();
                 $this->lexer->moveNext();
+                $tmpPrev = $this->lexer->getPrevious();
+                if ($tmpPrev['type'] === EmailLexer::S_CLOSEPARENTHESIS) {
+                    $openedParenthesis--;
+                }
+            }
+            if ($this->lexer->token['type'] === EmailLexer::S_CLOSEPARENTHESIS) {
+                if ($openedParenthesis === 0) {
+                    throw new \InvalidArgumentException('ERR_UNOPENEDCOMMENT');
+                } else {
+                    $openedParenthesis--;
+                }
             }
 
             $this->checkConsecutiveDots();
             $this->checkDomainPartExceptions($prev);
 
-            if ($this->hasBrackets()) {
+            if ($openBrackets = $this->hasBrackets($openBrackets)) {
                 $this->parseDomainLiteral();
             }
 
@@ -128,6 +146,14 @@ class DomainPart extends Parser
         return $domain;
     }
 
+    private function checkNotAllowedChars($token)
+    {
+        $notAllowed = array(EmailLexer::S_BACKSLASH => true, EmailLexer::S_SLASH=> true);
+        if (isset($notAllowed[$token['type']])) {
+            throw new \InvalidArgumentException('ERR_DOMAIN_CHAR_NOT_ALLOWED');
+        }
+    }
+
     protected function parseDomainLiteral()
     {
         if ($this->lexer->isNextToken(EmailLexer::S_COLON)) {
@@ -172,7 +198,7 @@ class DomainPart extends Parser
             }
 
             if ($this->lexer->isNextToken(EmailLexer::S_CR)) {
-                throw new \InvalidArgumentException("ERR_CR_NO_LF");
+                throw new \InvalidArgumentException('ERR_CR_NO_LF');
             }
             if ($this->lexer->token['type'] === EmailLexer::S_BACKSLASH) {
                 $this->warnings[] = EmailValidator::RFC5322_DOMLIT_OBSDTEXT;
@@ -235,12 +261,29 @@ class DomainPart extends Parser
 
     protected function checkDomainPartExceptions($prev)
     {
+        $invalidDomainTokens = array(
+            EmailLexer::S_DQUOTE => true,
+            EmailLexer::S_SEMICOLON => true,
+            EmailLexer::S_GREATERTHAN => true,
+            EmailLexer::S_LOWERTHAN => true,
+        );
+
+        if (isset($invalidDomainTokens[$this->lexer->token['type']])) {
+            throw new \InvalidArgumentException('ERR_EXPECTING_ATEXT');
+        }
+
+        if ($this->lexer->token['type'] === EmailLexer::S_COMMA) {
+            throw new \InvalidArgumentException('ERR_COMMA_IN_DOMAIN');
+        }
+
         if ($this->lexer->token['type'] === EmailLexer::S_AT) {
             throw new \InvalidArgumentException('ERR_CONSECUTIVEATS');
         }
+
         if ($this->lexer->token['type'] === EmailLexer::S_OPENQBRACKET && $prev['type'] !== EmailLexer::S_AT) {
             throw new \InvalidArgumentException('ERR_EXPECTING_ATEXT');
         }
+
         if ($this->lexer->token['type'] === EmailLexer::S_HYPHEN && $this->lexer->isNextToken(EmailLexer::S_DOT)) {
             throw new \InvalidArgumentException('ERR_DOMAINHYPHENEND');
         }
@@ -251,8 +294,12 @@ class DomainPart extends Parser
         }
     }
 
-    protected function hasBrackets()
+    protected function hasBrackets($openBrackets)
     {
+        if ($this->lexer->token['type'] === EmailLexer::S_CLOSEBRACKET && !$openBrackets) {
+            throw new \InvalidArgumentException('ERR_EXPECTING_OPENBRACKET');
+        }
+
         if ($this->lexer->token['type'] !== EmailLexer::S_OPENBRACKET) {
             return false;
         }
@@ -289,5 +336,4 @@ class DomainPart extends Parser
             throw new \InvalidArgumentException('ERR_EXPECTING_ATEXT');
         }
     }
-
-}
+}

src/Egulias/EmailValidator/Parser/LocalPart.php

@@ -3,43 +3,48 @@
 namespace Egulias\EmailValidator\Parser;
 
 use Egulias\EmailValidator\EmailLexer;
-use Egulias\EmailValidator\Parser\Parser;
 use Egulias\EmailValidator\EmailValidator;
 
-
 class LocalPart extends Parser
 {
     public function parse($localPart)
     {
+        $parseDQuote = true;
         $closingQuote = false;
-        while ($this->lexer->token['type'] !== EmailLexer::S_AT && $this->lexer->token) {
+        $openedParenthesis = 0;
 
+        while ($this->lexer->token['type'] !== EmailLexer::S_AT && $this->lexer->token) {
             if ($this->lexer->token['type'] === EmailLexer::S_DOT && !$this->lexer->getPrevious()) {
                 throw new \InvalidArgumentException('ERR_DOT_START');
             }
 
             $closingQuote = $this->checkDQUOTE($closingQuote);
+            if ($closingQuote && $parseDQuote) {
+                $parseDQuote = $this->parseDoubleQuote();
+            }
 
             if ($this->lexer->token['type'] === EmailLexer::S_OPENPARENTHESIS) {
                 $this->parseComments();
+                $openedParenthesis += $this->getOpenedParenthesis();
+            }
+            if ($this->lexer->token['type'] === EmailLexer::S_CLOSEPARENTHESIS) {
+                if ($openedParenthesis === 0) {
+                    throw new \InvalidArgumentException('ERR_UNOPENEDCOMMENT');
+                } else {
+                    $openedParenthesis--;
+                }
             }
 
             $this->checkConsecutiveDots();
 
-            if ($this->lexer->token['type'] === EmailLexer::S_DOT && $this->lexer->isNextToken(EmailLexer::S_AT)) {
+            if ($this->lexer->token['type'] === EmailLexer::S_DOT &&
+                $this->lexer->isNextToken(EmailLexer::S_AT)
+            ) {
                 throw new \InvalidArgumentException('ERR_DOT_END');
             }
 
             $this->warnEscaping();
-
-            if ($this->lexer->isNextTokenAny(
-                array(
-                    EmailLexer::INVALID, EmailLexer::S_LOWERTHAN, EmailLexer::S_GREATERTHAN
-                )
-            )
-            ) {
-                throw new \InvalidArgumentException('ERR_EXPECTING_ATEXT');
-            }
+            $this->isInvalidToken($this->lexer->token, $closingQuote);
 
             if ($this->isFWS()) {
                 $this->parseFWS();
@@ -53,4 +58,70 @@ class LocalPart extends Parser
             $this->warnings[] = EmailValidator::RFC5322_LOCAL_TOOLONG;
         }
     }
+
+    protected function parseDoubleQuote()
+    {
+        $parseAgain = true;
+        $special = array(
+            EmailLexer::S_CR => true,
+            EmailLexer::S_HTAB => true,
+            EmailLexer::S_LF => true
+        );
+
+        $invalid = array(
+            EmailLexer::C_NUL => true,
+            EmailLexer::S_HTAB => true,
+            EmailLexer::S_CR => true,
+            EmailLexer::S_LF => true
+        );
+        $setSpecialsWarning = true;
+
+        $this->lexer->moveNext();
+
+        while ($this->lexer->token['type'] !== EmailLexer::S_DQUOTE && $this->lexer->token) {
+            $parseAgain = false;
+            if (isset($special[$this->lexer->token['type']]) && $setSpecialsWarning) {
+                $this->warnings[] = EmailValidator::CFWS_FWS;
+                $setSpecialsWarning = false;
+            }
+
+            $this->lexer->moveNext();
+
+            if (!$this->escaped() && isset($invalid[$this->lexer->token['type']])) {
+                throw new \InvalidArgumentException('ERR_EXPECTED_ATEXT');
+            }
+        }
+
+        $prev = $this->lexer->getPrevious();
+
+        if ($prev['type'] === EmailLexer::S_BACKSLASH) {
+            if (!$this->checkDQUOTE(false)) {
+                throw new \InvalidArgumentException('ERR_UNCLOSED_DQUOTE');
+            }
+        }
+
+        if (!$this->lexer->isNextToken(EmailLexer::S_AT) && $prev['type'] !== EmailLexer::S_BACKSLASH) {
+            throw new \InvalidArgumentException('ERR_EXPECED_AT');
+        }
+
+        return $parseAgain;
+    }
+
+    protected function isInvalidToken($token, $closingQuote)
+    {
+        $forbidden = array(
+            EmailLexer::S_COMMA,
+            EmailLexer::S_CLOSEBRACKET,
+            EmailLexer::S_OPENBRACKET,
+            EmailLexer::S_GREATERTHAN,
+            EmailLexer::S_LOWERTHAN,
+            EmailLexer::S_COLON,
+            EmailLexer::S_SEMICOLON,
+            EmailLexer::INVALID
+        );
+
+        if (in_array($token['type'], $forbidden) && !$closingQuote) {
+            throw new \InvalidArgumentException('ERR_EXPECTING_ATEXT');
+        }
+    }
 }

src/Egulias/EmailValidator/Parser/Parser.php

@@ -9,6 +9,7 @@ abstract class Parser
 {
     protected $warnings = array();
     protected $lexer;
+    protected $openedParenthesis = 0;
 
     public function __construct(EmailLexer $lexer)
     {
@@ -20,7 +21,13 @@ abstract class Parser
         return $this->warnings;
     }
 
-    abstract function parse($str);
+    abstract public function parse($str);
+
+    /** @return int */
+    public function getOpenedParenthesis()
+    {
+        return $this->openedParenthesis;
+    }
 
     /**
      * validateQuotedPair
@@ -35,15 +42,15 @@ abstract class Parser
         $this->warnings[] = EmailValidator::DEPREC_QP;
     }
 
-    /**
-     * @return string the the comment
-     * @throws \InvalidArgumentException
-     */
     protected function parseComments()
     {
+        $this->openedParenthesis = 1;
         $this->isUnclosedComment();
         $this->warnings[] = EmailValidator::CFWS_COMMENT;
         while (!$this->lexer->isNextToken(EmailLexer::S_CLOSEPARENTHESIS)) {
+            if ($this->lexer->isNextToken(EmailLexer::S_OPENPARENTHESIS)) {
+                $this->openedParenthesis++;
+            }
             $this->warnEscaping();
             $this->lexer->moveNext();
         }
@@ -75,11 +82,11 @@ abstract class Parser
         $this->checkCRLFInFWS();
 
         if ($this->lexer->token['type'] === EmailLexer::S_CR) {
-            throw new \InvalidArgumentException("ERR_CR_NO_LF");
+            throw new \InvalidArgumentException('ERR_CR_NO_LF');
         }
 
         if ($this->lexer->isNextToken(EmailLexer::GENERIC) && $previous['type']  !== EmailLexer::S_AT) {
-            throw new \InvalidArgumentException("ERR_ATEXT_AFTER_CFWS");
+            throw new \InvalidArgumentException('ERR_ATEXT_AFTER_CFWS');
         }
 
         if ($this->lexer->token['type'] === EmailLexer::S_LF || $this->lexer->token['type'] === EmailLexer::C_NUL) {
@@ -102,6 +109,10 @@ abstract class Parser
 
     protected function isFWS()
     {
+        if ($this->escaped()) {
+            return false;
+        }
+
         if ($this->lexer->token['type'] === EmailLexer::S_SP ||
             $this->lexer->token['type'] === EmailLexer::S_HTAB ||
             $this->lexer->token['type'] === EmailLexer::S_CR ||
@@ -114,6 +125,20 @@ abstract class Parser
         return false;
     }
 
+    protected function escaped()
+    {
+        $previous = $this->lexer->getPrevious();
+
+        if ($previous['type'] === EmailLexer::S_BACKSLASH
+            &&
+            $this->lexer->token['type'] !== EmailLexer::GENERIC
+        ) {
+            return true;
+        }
+
+        return false;
+    }
+
     protected function warnEscaping()
     {
         if ($this->lexer->token['type'] !== EmailLexer::S_BACKSLASH) {
@@ -142,9 +167,10 @@ abstract class Parser
             return $hasClosingQuote;
         }
         $previous = $this->lexer->getPrevious();
-        if ($this->lexer->isNextToken(EmailLexer::GENERIC) && $previous['type'] === EmailLexer::GENERIC) {
+        if ($previous['type'] === EmailLexer::GENERIC && $this->lexer->isNextToken(EmailLexer::GENERIC)) {
             throw new \InvalidArgumentException('ERR_EXPECTING_ATEXT');
         }
+
         $this->warnings[] = EmailValidator::RFC5321_QUOTEDSTRING;
         try {
             $this->lexer->find(EmailLexer::S_DQUOTE);
@@ -162,10 +188,10 @@ abstract class Parser
             return;
         }
         if ($this->lexer->isNextToken(EmailLexer::CRLF)) {
-            throw new \InvalidArgumentException("ERR_FWS_CRLF_X2");
+            throw new \InvalidArgumentException('ERR_FWS_CRLF_X2');
         }
         if (!$this->lexer->isNextTokenAny(array(EmailLexer::S_SP, EmailLexer::S_HTAB))) {
-            throw new \InvalidArgumentException("ERR_FWS_CRLF_END");
+            throw new \InvalidArgumentException('ERR_FWS_CRLF_END');
         }
     }
-}
+}

tests/egulias/Tests/EmailValidator/EmailLexerTest.php

@@ -26,6 +26,73 @@ class EmailLexerTests extends \PHPUnit_Framework_TestCase
         $this->assertEquals($token, $lexer->token['type']);
     }
 
+    public function testLexerParsesMultipleSpaces()
+    {
+        $lexer = new EmailLexer();
+        $lexer->setInput('  ');
+        $lexer->moveNext();
+        $lexer->moveNext();
+        $this->assertEquals(EmailLexer::S_SP, $lexer->token['type']);
+        $lexer->moveNext();
+        $this->assertEquals(EmailLexer::S_SP, $lexer->token['type']);
+    }
+
+    /**
+     * @dataProvider invalidUTF8CharsProvider
+     */
+    public function testLexerParsesInvalidUTF8($char)
+    {
+        $lexer = new EmailLexer();
+        $lexer->setInput($char);
+        $lexer->moveNext();
+        $lexer->moveNext();
+
+        $this->assertEquals(EmailLexer::INVALID, $lexer->token['type']);
+    }
+
+    public function invalidUTF8CharsProvider()
+    {
+        $chars = array();
+        for ($i = 0; $i < 0x100; ++$i) {
+            $c = $this->utf8Chr($i);
+            if (preg_match('/(?=\p{Cc})(?=[^\t\n\n\r])/u', $c) && !preg_match('/\x{0000}/u', $c)) {
+                $chars[] = array($c);
+            }
+        }
+
+        return $chars;
+    }
+
+    protected function utf8Chr($code_point)
+    {
+
+        if ($code_point < 0 || 0x10FFFF < $code_point || (0xD800 <= $code_point && $code_point <= 0xDFFF)) {
+            return '';
+        }
+
+        if ($code_point < 0x80) {
+            $hex[0] = $code_point;
+            $ret = chr($hex[0]);
+        } elseif ($code_point < 0x800) {
+            $hex[0] = 0x1C0 | $code_point >> 6;
+            $hex[1] = 0x80  | $code_point & 0x3F;
+            $ret = chr($hex[0]).chr($hex[1]);
+        } elseif ($code_point < 0x10000) {
+            $hex[0] = 0xE0 | $code_point >> 12;
+            $hex[1] = 0x80 | $code_point >> 6 & 0x3F;
+            $hex[2] = 0x80 | $code_point & 0x3F;
+            $ret = chr($hex[0]).chr($hex[1]).chr($hex[2]);
+        } else {
+            $hex[0] = 0xF0 | $code_point >> 18;
+            $hex[1] = 0x80 | $code_point >> 12 & 0x3F;
+            $hex[2] = 0x80 | $code_point >> 6 & 0x3F;
+            $hex[3] = 0x80 | $code_point  & 0x3F;
+            $ret = chr($hex[0]).chr($hex[1]).chr($hex[2]).chr($hex[3]);
+        }
+
+        return $ret;
+    }
+
     public function testLexerForTab()
     {
         $lexer = new EmailLexer();
@@ -36,6 +103,25 @@ class EmailLexerTests extends \PHPUnit_Framework_TestCase
         $this->assertEquals(EmailLexer::S_HTAB, $lexer->token['type']);
     }
 
+    public function testLexerForUTF8()
+    {
+        $lexer = new EmailLexer();
+        $lexer->setInput("áÇ@bar.com");
+        $lexer->moveNext();
+        $lexer->moveNext();
+        $this->assertEquals(EmailLexer::GENERIC, $lexer->token['type']);
+        $lexer->moveNext();
+        $this->assertEquals(EmailLexer::GENERIC, $lexer->token['type']);
+    }
+
+    public function testLexerSearchToken()
+    {
+        $lexer = new EmailLexer();
+        $lexer->setInput("foo\tbar");
+        $lexer->moveNext();
+        $this->assertTrue($lexer->find(EmailLexer::S_HTAB));
+    }
+
     public function getTokens()
     {
         return array(
@@ -53,6 +139,7 @@ class EmailLexerTests extends \PHPUnit_Framework_TestCase
             array("\"", EmailLexer::S_DQUOTE),
             array("-", EmailLexer::S_HYPHEN),
             array("\\", EmailLexer::S_BACKSLASH),
+            array("/", EmailLexer::S_SLASH),
             array("(", EmailLexer::S_OPENPARENTHESIS),
             array(")", EmailLexer::S_CLOSEPARENTHESIS),
             array('<', EmailLexer::S_LOWERTHAN),
@@ -66,7 +153,9 @@ class EmailLexerTests extends \PHPUnit_Framework_TestCase
             array('{', EmailLexer::S_OPENQBRACKET),
             array('}', EmailLexer::S_CLOSEQBRACKET),
             array('',  EmailLexer::S_EMPTY),
-            array(chr(31),  EmailLexer::INVALID)
+            array(chr(31),  EmailLexer::INVALID),
+            array(chr(226),  EmailLexer::GENERIC),
+            array(chr(0),  EmailLexer::C_NUL)
         );
     }
 }

tests/egulias/Tests/EmailValidator/EmailValidatorTest.php

@@ -26,26 +26,52 @@ class EmailValidatorTest extends \PHPUnit_Framework_TestCase
         $this->assertTrue($this->validator->isValid($email));
     }
 
+    public function testInvalidUTF8Email()
+    {
+        $validator = new EmailValidator;
+        $email     = "\x80\x81\x82@\x83\x84\x85.\x86\x87\x88";
+
+        $this->assertFalse($validator->isValid($email));
+    }
+
     public function getValidEmails()
     {
         return array(
+            array('â@iana.org'),
             array('fabien@symfony.com'),
             array('example@example.co.uk'),
             array('fabien_potencier@example.fr'),
             array('example@localhost'),
             array('fab\'ien@symfony.com'),
+            array('fab\ ien@symfony.com'),
             array('example((example))@fakedfake.co.uk'),
             array('example@faked(fake).co.uk'),
             array('fabien+@symfony.com'),
             array('инфо@письмо.рф'),
-
+            array('"username"@example.com'),
+            array('"user,name"@example.com'),
+            array('"user name"@example.com'),
+            array('"user@name"@example.com'),
+            array('"\a"@iana.org'),
+            array('"test\ test"@iana.org'),
+            array('""@iana.org'),
+            array('"\""@iana.org'),
+            array('müller@möller.de'),
+            array('test@email*'),
+            array('test@email!'),
+            array('test@email&'),
+            array('test@email^'),
+            array('test@email%'),
+            array('test@email$'),
+            array('test@email.com.au'),
+            array('1500111@профи-инвест.рф'),
         );
     }
 
     /**
      * @dataProvider getInvalidEmails
      */
-    public function testAInvalidEmails($email)
+    public function testInvalidEmails($email)
     {
         $this->assertFalse($this->validator->isValid($email));
     }
@@ -53,18 +79,55 @@ class EmailValidatorTest extends \PHPUnit_Framework_TestCase
     public function getInvalidEmails()
     {
         return array(
+            array('test@example.com test'),
+            array('user  name@example.com'),
+            array('user   name@example.com'),
             array('example.@example.co.uk'),
             array('example@example@example.co.uk'),
-            array('(fabien_potencier@example.fr)'),
+            array('(test_exampel@example.fr)'),
             array('example(example)example@example.co.uk'),
             array('.example@localhost'),
             array('ex\ample@localhost'),
             array('example@local\host'),
+            array('example@localhost\\'),
             array('example@localhost.'),
             array('user name@example.com'),
             array('username@ example . com'),
             array('example@(fake).com'),
             array('example@(fake.com'),
+            array('username@example,com'),
+            array('usern,ame@example.com'),
+            array('user[na]me@example.com'),
+            array('"""@iana.org'),
+            array('"\"@iana.org'),
+            array('"test"test@iana.org'),
+            array('"test""test"@iana.org'),
+            array('"test"."test"@iana.org'),
+            array('"test".test@iana.org'),
+            array('"test"' . chr(0) . '@iana.org'),
+            array('"test\"@iana.org'),
+            array(chr(226) . '@iana.org'),
+            array('test@' . chr(226) . '.org'),
+            array('\r\ntest@iana.org'),
+            array('\r\n test@iana.org'),
+            array('\r\n \r\ntest@iana.org'),
+            array('\r\n \r\ntest@iana.org'),
+            array('\r\n \r\n test@iana.org'),
+            array('test@iana.org \r\n'),
+            array('test@iana.org \r\n '),
+            array('test@iana.org \r\n \r\n'),
+            array('test@iana.org \r\n\r\n'),
+            array('test@iana.org  \r\n\r\n '),
+            array('test@iana/icann.org'),
+            array('test@foo;bar.com'),
+            array('test;123@foobar.com'),
+            array('test@example..com'),
+            array('email.email@email."'),
+            array('test@email>'),
+            array('test@email<'),
+            array('test@email{'),
+            array('test@email.com]'),
+            array('test@ema[il.com'),
         );
     }
 
@@ -94,6 +157,11 @@ class EmailValidatorTest extends \PHPUnit_Framework_TestCase
             array(EmailValidator::ERR_DOT_END, 'example@localhost.'),
             array(EmailValidator::ERR_DOT_END, 'example.@example.co.uk'),
             array(EmailValidator::ERR_UNCLOSEDCOMMENT, '(example@localhost'),
+            array(EmailValidator::ERR_UNOPENEDCOMMENT, 'comment)example@localhost'),
+            array(EmailValidator::ERR_UNOPENEDCOMMENT, 'example(comment))@localhost'),
+            array(EmailValidator::ERR_UNOPENEDCOMMENT, 'example@comment)localhost'),
+            array(EmailValidator::ERR_UNOPENEDCOMMENT, 'example@localhost(comment))'),
+            array(EmailValidator::ERR_UNOPENEDCOMMENT, 'example@(comment))example.com'),
             array(EmailValidator::ERR_UNCLOSEDQUOTEDSTR, '"example@localhost'),
             array(EmailValidator::ERR_EXPECTING_ATEXT, 'exa"mple@localhost'),
             //This was the original. But atext is not allowed after \n
@@ -104,39 +172,76 @@ class EmailValidatorTest extends \PHPUnit_Framework_TestCase
             array(EmailValidator::ERR_CR_NO_LF, "example@exa\rmple.co.uk"),
             array(EmailValidator::ERR_CR_NO_LF, "example@[\r]"),
             array(EmailValidator::ERR_CR_NO_LF, "exam\rple@example.co.uk"),
-            array(EmailValidator::ERR_CR_NO_LF, "\"\r\"@localhost"),
         );
     }
 
     /**
      * @dataProvider getInvalidEmailsWithWarnings
      */
-    public function testInvalidEmailsWithWarningsCheck($warnings, $email)
+    public function testValidEmailsWithWarningsCheck($warnings, $email)
     {
         $this->assertTrue($this->validator->isValid($email, true));
 
         $this->assertEquals($warnings, $this->validator->getWarnings());
     }
 
+    /**
+     * @dataProvider getInvalidEmailsWithWarnings
+     */
+    public function testInvalidEmailsWithDnsCheckAndStrictMode($warnings, $email)
+    {
+        $this->assertFalse($this->validator->isValid($email, true, true));
+
+        $this->assertEquals($warnings, $this->validator->getWarnings());
+    }
+
     public function getInvalidEmailsWithWarnings()
     {
         return array(
-            array(array( EmailValidator::DEPREC_CFWS_NEAR_AT,), 'example @example.co.uk'),
-            array(array( EmailValidator::DEPREC_CFWS_NEAR_AT,), 'example@ example.co.uk'),
-            array(array( EmailValidator::CFWS_COMMENT,), 'example@example(examplecomment).co.uk'),
+            array(
+                array(
+                    EmailValidator::DEPREC_CFWS_NEAR_AT,
+                    EmailValidator::DNSWARN_NO_RECORD
+                ),
+                'example @invalid.example.com'
+            ),
+            array(
+                array(
+                    EmailValidator::DEPREC_CFWS_NEAR_AT,
+                    EmailValidator::DNSWARN_NO_RECORD
+                ),
+                'example@ invalid.example.com'
+            ),
+            array(
+                array(
+                    EmailValidator::CFWS_COMMENT,
+                    EmailValidator::DNSWARN_NO_RECORD
+                ),
+                'example@invalid.example(examplecomment).com'
+            ),
             array(
                 array(
                     EmailValidator::CFWS_COMMENT,
                     EmailValidator::DEPREC_CFWS_NEAR_AT,
+                    EmailValidator::DNSWARN_NO_RECORD,
                 ),
-                'example(examplecomment)@example.co.uk'
+                'example(examplecomment)@invalid.example.com'
             ),
             array(
                 array(
                     EmailValidator::RFC5321_QUOTEDSTRING,
                     EmailValidator::CFWS_FWS,
+                    EmailValidator::DNSWARN_NO_RECORD,
                 ),
-                "\"\t\"@example.co.uk"
+                "\"\t\"@invalid.example.com"
+            ),
+            array(
+                array(
+                    EmailValidator::RFC5321_QUOTEDSTRING,
+                    EmailValidator::CFWS_FWS,
+                    EmailValidator::DNSWARN_NO_RECORD
+                ),
+                "\"\r\"@invalid.example.com"
             ),
             array(
                 array(
@@ -226,14 +331,16 @@ class EmailValidatorTest extends \PHPUnit_Framework_TestCase
             array(
                 array(
                     EmailValidator::RFC5321_QUOTEDSTRING,
+                    EmailValidator::DNSWARN_NO_RECORD
                 ),
-                '"example"@example.co.uk'
+                '"example"@invalid.example.com'
             ),
             array(
                 array(
                     EmailValidator::RFC5322_LOCAL_TOOLONG,
+                    EmailValidator::DNSWARN_NO_RECORD
                 ),
-                'too_long_localpart_too_long_localpart_too_long_localpart_too_long_localpart@example.co.uk'
+                'too_long_localpart_too_long_localpart_too_long_localpart_too_long_localpart@invalid.example.com'
             ),
             array(
                 array(
@@ -262,6 +369,17 @@ class EmailValidatorTest extends \PHPUnit_Framework_TestCase
                 'parttoolonglocalparttoolonglocalparttoolonglocalparttoolonglocalparttoolonglocalparttoolonglocalpart'.
                 'toolonglocalparttoolonglocalparttoolonglocalparttoolonglocalpar'
             ),
+            array(
+                array(
+                    EmailValidator::DNSWARN_NO_RECORD,
+                ),
+                'test@test'
+            ),
         );
     }
+
+    public function testInvalidEmailsWithStrict()
+    {
+        $this->assertFalse($this->validator->isValid('"test"@test', false, true));
+    }
 }