From 430c066f7797328afe292dabde608f1b784a1ab6 Mon Sep 17 00:00:00 2001
From: Francois Zaninotto <fzaninotto@gmail.com>
Date: Fri, 7 Nov 2014 09:05:12 +0100
Subject: [PATCH 1/2] Add regexify formatter

---
 src/Faker/Provider/Base.php      | 73 ++++++++++++++++++++++++++++++++
 test/Faker/Provider/BaseTest.php | 50 +++++++++++++++++++++-
 2 files changed, 121 insertions(+), 2 deletions(-)

diff --git a/src/Faker/Provider/Base.php b/src/Faker/Provider/Base.php
index de61c8b49a..e5f2c0fa48 100644
--- a/src/Faker/Provider/Base.php
+++ b/src/Faker/Provider/Base.php
@@ -370,6 +370,79 @@ public static function asciify($string = '****')
         return preg_replace_callback('/\*/u', 'static::randomAscii', $string);
     }
 
+    /**
+     * Transforms a basic regular expression into a random string satisfying the expression.
+     *
+     * @example $faker->regexify('[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,4}'); // sm0@y8k96a.ej
+     *
+     * Regex delimiters '/.../' and begin/end markers '^...$' are ignored.
+     *
+     * Only supports a small subset of the regex syntax. For instance,
+     * unicode, negated classes, unbouned ranges, subpatterns, back references,
+     * assertions, recursive patterns, and comments are not supported. Escaping
+     * support is extremely fragile.
+     *
+     * This method is also VERY slow. Use it only when no other formatter
+     * can generate the fake data you want. For instance, prefer calling
+     * `$faker->email` rather than `regexify` with the previous regular
+     * expression.
+     *
+     * Also note than `bothify` can probably do most of what this method does,
+     * but much faster. For instance, for a dummy email generation, try
+     * `$faker->bothify('?????????@???.???')`.
+     *
+     * @see https://github.com/icomefromthenet/ReverseRegex for a more robust implementation
+     *
+     * @param string $regex A regular expression (delimiters are optional)
+     * @return string
+     */
+    public static function regexify($regex = '')
+    {
+        // ditch the anchors
+        $regex = preg_replace('/^\/?\^?/', '', $regex);
+        $regex = preg_replace('/\$?\/?$/', '', $regex);
+        // All {2} become {2,2}
+        $regex = preg_replace('/\{(\d+)\}/', '{\1,\1}', $regex);
+        // Single-letter quantifiers (?, *, +) become bracket quantifiers ({0,1}, {0,rand}, {1, rand})
+        $regex = preg_replace('/(?<!\\\)\?/', '{0,1}', $regex);
+        $regex = preg_replace('/(?<!\\\)\*/', '{0,' . self::randomDigitNotNull() . '}', $regex);
+        $regex = preg_replace('/(?<!\\\)\+/', '{1,' . self::randomDigitNotNull() . '}', $regex);
+        // [12]{1,2} becomes [12] or [12][12]
+        $regex = preg_replace_callback('/(\[[^\]]+\])\{(\d+),(\d+)\}/', function ($matches) {
+            return str_repeat($matches[1], static::randomElement(range($matches[2], $matches[3])));
+        }, $regex);
+        // (12|34){1,2} becomes (12|34) or (12|34)(12|34)
+        $regex = preg_replace_callback('/(\([^\)]+\))\{(\d+),(\d+)\}/', function ($matches) {
+            return str_repeat($matches[1], static::randomElement(range($matches[2], $matches[3])));
+        }, $regex);
+        // A{1,2} becomes A or AA or \d{3} becomes \d\d\d
+        $regex = preg_replace_callback('/(\\\?.)\{(\d+),(\d+)\}/', function ($matches) {
+            return str_repeat($matches[1], static::randomElement(range($matches[2], $matches[3])));
+        }, $regex);
+        // (this|that) becomes 'this' or 'that'
+        $regex = preg_replace_callback('/\((.*?)\)/', function ($matches) {
+            return static::randomElement(explode('|', str_replace(array('(', ')'), '', $matches[1])));
+        }, $regex);
+        // All A-F inside of [] become ABCDEF
+        $regex = preg_replace_callback('/\[([^\]]+)\]/', function ($matches) {
+            return '[' . preg_replace_callback('/(\w|\d)\-(\w|\d)/', function ($range) {
+                return join(range($range[1], $range[2]), '');
+            }, $matches[1]) . ']';
+        }, $regex);
+        // All [ABC] become B (or A or C)
+        $regex = preg_replace_callback('/\[([^\]]+)\]/', function ($matches) {
+            return static::randomElement(str_split($matches[1]));
+        }, $regex);
+        // replace \d with number and \w with letter and . with ascii
+        $regex = preg_replace_callback('/\\\w/', 'static::randomLetter', $regex);
+        $regex = preg_replace_callback('/\\\d/', 'static::randomDigit', $regex);
+        $regex = preg_replace_callback('/(?<!\\\)\./', 'static::randomAscii', $regex);
+        // remove remaining backslashes
+        $regex = str_replace('\\', '', $regex);
+        // phew
+        return $regex;
+    }
+
     /**
      * Converts string to lowercase.
      * Uses mb_string extension if available.
diff --git a/test/Faker/Provider/BaseTest.php b/test/Faker/Provider/BaseTest.php
index 875d19db6a..95e694ab6e 100644
--- a/test/Faker/Provider/BaseTest.php
+++ b/test/Faker/Provider/BaseTest.php
@@ -271,9 +271,55 @@ public function testAsciifyReturnsSameStringWhenItContainsNoStarSign()
         $this->assertEquals('fooBar?', BaseProvider::asciify('fooBar?'));
     }
 
-    public function testNumerifyReturnsStringWithStarSignsReplacedByAsciiChars()
+    public function testAsciifyReturnsStringWithStarSignsReplacedByAsciiChars()
     {
-        $this->assertRegExp('/foo.Ba.r/', BaseProvider::numerify('foo*Ba*r'));
+        $this->assertRegExp('/foo.Ba.r/', BaseProvider::asciify('foo*Ba*r'));
+    }
+
+    public function regexifyBasicDataProvider()
+    {
+        return array(
+            array('azeQSDF1234', 'azeQSDF1234', 'does not change non regex chars'),
+            array('foo(bar){1}', 'foobar', 'replaces regex characters'),
+            array('', '', 'supports empty string'),
+            array('/^foo(bar){1}$/', 'foobar', 'ignores regex delimiters')
+        );
+    }
+
+    /**
+     * @dataProvider regexifyBasicDataProvider
+     */
+    public function testRegexifyBasicFeatures($input, $output, $message)
+    {
+        $this->assertEquals($output, BaseProvider::regexify($input), $message);
+    }
+
+    public function regexifyDataProvider()
+    {
+        return array(
+            array('\d', 'numbers'),
+            array('\w', 'letters'),
+            array('(a|b)', 'alternation'),
+            array('[aeiou]', 'basic character class'),
+            array('[a-z]', 'character class range'),
+            array('[a-z1-9]', 'multiple character class range'),
+            array('a*b+c?', 'single character quantifiers'),
+            array('a{2}', 'brackets quantifiers'),
+            array('a{2,3}', 'min-max brackets quantifiers'),
+            array('[aeiou]{2,3}', 'brackets quantifiers on basic character class'),
+            array('[a-z]{2,3}', 'brackets quantifiers on character class range'),
+            array('(a|b){2,3}', 'brackets quantifiers on alternation'),
+            array('\.\*\?\+', 'escaped characters'),
+            array('[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,4}', 'complex regex')
+        );
+    }
+
+    /**
+     * @dataProvider regexifyDataProvider
+     */
+    public function testRegexifySupportedRegexSyntax($pattern, $message)
+    {
+        $this->assertRegExp('/' . $pattern . '/', BaseProvider::regexify($pattern), 'Regexify supports ' . $message);
     }
 
     public function testOptionalReturnsProviderValueWhenCalledWithWeight1()

From b7a0aeeedee473ce9487ff45a78e4373be4148f1 Mon Sep 17 00:00:00 2001
From: Francois Zaninotto <fzaninotto@gmail.com>
Date: Fri, 7 Nov 2014 09:13:22 +0100
Subject: [PATCH 2/2] Fix regexify on PHP 5.3

---
 src/Faker/Provider/Base.php | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/src/Faker/Provider/Base.php b/src/Faker/Provider/Base.php
index e5f2c0fa48..ee5b3a0056 100644
--- a/src/Faker/Provider/Base.php
+++ b/src/Faker/Provider/Base.php
@@ -405,23 +405,23 @@ public static function regexify($regex = '')
         $regex = preg_replace('/\{(\d+)\}/', '{\1,\1}', $regex);
         // Single-letter quantifiers (?, *, +) become bracket quantifiers ({0,1}, {0,rand}, {1, rand})
         $regex = preg_replace('/(?<!\\\)\?/', '{0,1}', $regex);
-        $regex = preg_replace('/(?<!\\\)\*/', '{0,' . self::randomDigitNotNull() . '}', $regex);
-        $regex = preg_replace('/(?<!\\\)\+/', '{1,' . self::randomDigitNotNull() . '}', $regex);
+        $regex = preg_replace('/(?<!\\\)\*/', '{0,' . static::randomDigitNotNull() . '}', $regex);
+        $regex = preg_replace('/(?<!\\\)\+/', '{1,' . static::randomDigitNotNull() . '}', $regex);
         // [12]{1,2} becomes [12] or [12][12]
         $regex = preg_replace_callback('/(\[[^\]]+\])\{(\d+),(\d+)\}/', function ($matches) {
-            return str_repeat($matches[1], static::randomElement(range($matches[2], $matches[3])));
+            return str_repeat($matches[1], Base::randomElement(range($matches[2], $matches[3])));
         }, $regex);
         // (12|34){1,2} becomes (12|34) or (12|34)(12|34)
         $regex = preg_replace_callback('/(\([^\)]+\))\{(\d+),(\d+)\}/', function ($matches) {
-            return str_repeat($matches[1], static::randomElement(range($matches[2], $matches[3])));
+            return str_repeat($matches[1], Base::randomElement(range($matches[2], $matches[3])));
         }, $regex);
         // A{1,2} becomes A or AA or \d{3} becomes \d\d\d
         $regex = preg_replace_callback('/(\\\?.)\{(\d+),(\d+)\}/', function ($matches) {
-            return str_repeat($matches[1], static::randomElement(range($matches[2], $matches[3])));
+            return str_repeat($matches[1], Base::randomElement(range($matches[2], $matches[3])));
         }, $regex);
         // (this|that) becomes 'this' or 'that'
         $regex = preg_replace_callback('/\((.*?)\)/', function ($matches) {
-            return static::randomElement(explode('|', str_replace(array('(', ')'), '', $matches[1])));
+            return Base::randomElement(explode('|', str_replace(array('(', ')'), '', $matches[1])));
         }, $regex);
         // All A-F inside of [] become ABCDEF
         $regex = preg_replace_callback('/\[([^\]]+)\]/', function ($matches) {
@@ -431,7 +431,7 @@ public static function regexify($regex = '')
         }, $regex);
         // All [ABC] become B (or A or C)
         $regex = preg_replace_callback('/\[([^\]]+)\]/', function ($matches) {
-            return static::randomElement(str_split($matches[1]));
+            return Base::randomElement(str_split($matches[1]));
         }, $regex);
         // replace \d with number and \w with letter and . with ascii
         $regex = preg_replace_callback('/\\\w/', 'static::randomLetter', $regex);