Merge pull request #16 from nojimage/develop

Refactor Regex class
nojimage · Dec 28, 2017 · a2a1fb1 · a2a1fb1
2 parents 4c3d764 + c763596
commit a2a1fb1
Show file tree

Hide file tree

Showing 10 changed files with 2,694 additions and 284 deletions.
diff --git a/.travis.yml b/.travis.yml
@@ -6,6 +6,7 @@ php:
   - 5.6
   - 7.0
   - 7.1
+  - 7.2
 
 dist: trusty
 sudo: false

diff --git a/lib/Twitter/Text/Autolink.php b/lib/Twitter/Text/Autolink.php
@@ -30,7 +30,7 @@
  * @license    http://www.apache.org/licenses/LICENSE-2.0  Apache License v2.0
  * @package    Twitter.Text
  */
-class Autolink extends Regex
+class Autolink
 {
 
     /**
@@ -139,6 +139,13 @@ class Autolink extends Regex
      */
     protected $extractor = null;
 
+    /**
+     * The tweet to be used in parsing.
+     *
+     * @var  string
+     */
+    protected $tweet = '';
+
     /**
      * Provides fluent method chaining.
      *
@@ -170,13 +177,14 @@ public function __construct($tweet = null, $escape = true, $full_encode = false)
     {
         if ($escape && !empty($tweet)) {
             if ($full_encode) {
-                parent::__construct(htmlentities($tweet, ENT_QUOTES, 'UTF-8', false));
+                $this->tweet = htmlentities($tweet, ENT_QUOTES, 'UTF-8', false);
             } else {
-                parent::__construct(htmlspecialchars($tweet, ENT_QUOTES, 'UTF-8', false));
+                $this->tweet = htmlspecialchars($tweet, ENT_QUOTES, 'UTF-8', false);
             }
         } else {
-            parent::__construct($tweet);
+            $this->tweet = $tweet;
         }
+
         $this->extractor = Extractor::create();
     }
 
@@ -650,7 +658,7 @@ public function linkToHashtag($entity, $tweet = null)
         if (!empty($this->class_hash)) {
             $class[] = $this->class_hash;
         }
-        if (preg_match(self::$patterns['rtl_chars'], $linkText)) {
+        if (preg_match(Regex::getRtlCharsMatcher(), $linkText)) {
             $class[] = 'rtl';
         }
         if (!empty($class)) {

diff --git a/lib/Twitter/Text/Extractor.php b/lib/Twitter/Text/Extractor.php
@@ -29,14 +29,21 @@
  * @license    http://www.apache.org/licenses/LICENSE-2.0  Apache License v2.0
  * @package    Twitter.Text
  */
-class Extractor extends Regex
+class Extractor
 {
 
     /**
      * @var boolean
      */
     protected $extractURLWithoutProtocol = true;
 
+    /**
+     * The tweet to be used in parsing.
+     *
+     * @var  string
+     */
+    protected $tweet = '';
+
     /**
      * Provides fluent method chaining.
      *
@@ -60,7 +67,7 @@ public static function create($tweet = null)
      */
     public function __construct($tweet = null)
     {
-        parent::__construct($tweet);
+        $this->tweet = $tweet;
     }
 
     /**
@@ -205,9 +212,9 @@ public function extractReplyScreenname($tweet = null)
         if (is_null($tweet)) {
             $tweet = $this->tweet;
         }
-        $matched = preg_match(self::$patterns['valid_reply'], $tweet, $matches);
+        $matched = preg_match(Regex::getValidReplyMatcher(), $tweet, $matches);
         # Check username ending in
-        if ($matched && preg_match(self::$patterns['end_mention_match'], $matches[2])) {
+        if ($matched && preg_match(Regex::getEndMentionMatcher(), $matches[2])) {
             $matched = false;
         }
         return $matched ? $matches[1] : null;
@@ -243,15 +250,15 @@ public function extractHashtagsWithIndices($tweet = null, $checkUrlOverlap = tru
             return array();
         }
 
-        preg_match_all(self::$patterns['valid_hashtag'], $tweet, $matches, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
+        preg_match_all(Regex::getValidHashtagMatcher(), $tweet, $matches, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
         $tags = array();
 
         foreach ($matches as $match) {
             list($all, $before, $hash, $hashtag, $outer) = array_pad($match, 3, array('', 0));
             $start_position = $hash[1] > 0 ? StringUtils::strlen(substr($tweet, 0, $hash[1])) : $hash[1];
             $end_position = $start_position + StringUtils::strlen($hash[0] . $hashtag[0]);
 
-            if (preg_match(self::$patterns['end_hashtag_match'], $outer[0])) {
+            if (preg_match(Regex::getEndHashtagMatcher(), $outer[0])) {
                 continue;
             }
 
@@ -296,15 +303,15 @@ public function extractCashtagsWithIndices($tweet = null)
             return array();
         }
 
-        preg_match_all(self::$patterns['valid_cashtag'], $tweet, $matches, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
+        preg_match_all(Regex::getValidCashtagMatcher(), $tweet, $matches, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
         $tags = array();
 
         foreach ($matches as $match) {
             list($all, $before, $dollar, $cash_text, $outer) = array_pad($match, 3, array('', 0));
             $start_position = $dollar[1] > 0 ? StringUtils::strlen(substr($tweet, 0, $dollar[1])) : $dollar[1];
             $end_position = $start_position + StringUtils::strlen($dollar[0] . $cash_text[0]);
 
-            if (preg_match(self::$patterns['end_hashtag_match'], $outer[0])) {
+            if (preg_match(Regex::getEndHashtagMatcher(), $outer[0])) {
                 continue;
             }
 
@@ -335,7 +342,7 @@ public function extractURLsWithIndices($tweet = null)
         }
 
         $urls = array();
-        preg_match_all(self::$patterns['valid_url'], $tweet, $matches, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
+        preg_match_all(Regex::getValidUrlMatcher(), $tweet, $matches, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
 
         foreach ($matches as $match) {
             list($all, $before, $url, $protocol, $domain, $port, $path, $query) = array_pad($match, 8, array(''));
@@ -354,14 +361,14 @@ public function extractURLsWithIndices($tweet = null)
             // If protocol is missing and domain contains non-ASCII characters,
             // extract ASCII-only domains.
             if (empty($protocol)) {
-                if (!$this->extractURLWithoutProtocol || preg_match(self::$patterns['invalid_url_without_protocol_preceding_chars'], $before)) {
+                if (!$this->extractURLWithoutProtocol || preg_match(Regex::getInvalidUrlWithoutProtocolPrecedingCharsMatcher(), $before)) {
                     continue;
                 }
 
                 $last_url = null;
                 $ascii_end_position = 0;
 
-                if (preg_match(self::$patterns['valid_ascii_domain'], $domain, $asciiDomain)) {
+                if (preg_match(Regex::getValidAsciiDomainMatcher(), $domain, $asciiDomain)) {
                     $asciiDomain[0] = preg_replace('/' . preg_quote($domain, '/') . '/u', $asciiDomain[0], $url);
                     $ascii_start_position = StringUtils::strpos($domain, $asciiDomain[0], $ascii_end_position);
                     $ascii_end_position = $ascii_start_position + StringUtils::strlen($asciiDomain[0]);
@@ -370,8 +377,8 @@ public function extractURLsWithIndices($tweet = null)
                         'indices' => array($start_position + $ascii_start_position, $start_position + $ascii_end_position),
                     );
                     if (!empty($path)
-                        || preg_match(self::$patterns['valid_special_short_domain'], $asciiDomain[0])
-                        || !preg_match(self::$patterns['invalid_short_domain'], $asciiDomain[0])) {
+                        || preg_match(Regex::getValidSpecialShortDomainMatcher(), $asciiDomain[0])
+                        || !preg_match(Regex::getInvalidCharactersMatcher(), $asciiDomain[0])) {
                         $urls[] = $last_url;
                     }
                 }
@@ -389,7 +396,7 @@ public function extractURLsWithIndices($tweet = null)
                 }
             } else {
                 // In the case of t.co URLs, don't allow additional path characters
-                if (preg_match(self::$patterns['valid_tco_url'], $url, $tcoUrlMatches)) {
+                if (preg_match(Regex::getValidTcoUrlMatcher(), $url, $tcoUrlMatches)) {
                     $url = $tcoUrlMatches[0];
                     $end_position = $start_position + StringUtils::strlen($url);
                 }
@@ -453,7 +460,7 @@ public function extractMentionsOrListsWithIndices($tweet = null)
             return array();
         }
 
-        preg_match_all(self::$patterns['valid_mentions_or_lists'], $tweet, $matches, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
+        preg_match_all(Regex::getValidMentionsOrListsMatcher(), $tweet, $matches, PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
         $results = array();
 
         foreach ($matches as $match) {
@@ -466,7 +473,7 @@ public function extractMentionsOrListsWithIndices($tweet = null)
                 'indices' => array($start_position, $end_position),
             );
 
-            if (preg_match(self::$patterns['end_mention_match'], $outer[0])) {
+            if (preg_match(Regex::getEndMentionMatcher(), $outer[0])) {
                 continue;
             }
 

diff --git a/lib/Twitter/Text/HitHighlighter.php b/lib/Twitter/Text/HitHighlighter.php
@@ -27,7 +27,7 @@
  * @license    http://www.apache.org/licenses/LICENSE-2.0  Apache License v2.0
  * @package    Twitter.Text
  */
-class HitHighlighter extends Regex
+class HitHighlighter
 {
 
     /**
@@ -37,6 +37,13 @@ class HitHighlighter extends Regex
      */
     protected $tag = 'em';
 
+    /**
+     * The tweet to be used in parsing.
+     *
+     * @var  string
+     */
+    protected $tweet = '';
+
     /**
      * Provides fluent method chaining.
      *
@@ -67,12 +74,12 @@ public function __construct($tweet = null, $escape = true, $full_encode = false)
     {
         if (!empty($tweet) && $escape) {
             if ($full_encode) {
-                parent::__construct(htmlentities($tweet, ENT_QUOTES, 'UTF-8', false));
+                $this->tweet = htmlentities($tweet, ENT_QUOTES, 'UTF-8', false);
             } else {
-                parent::__construct(htmlspecialchars($tweet, ENT_QUOTES, 'UTF-8', false));
+                $this->tweet = htmlspecialchars($tweet, ENT_QUOTES, 'UTF-8', false);
             }
         } else {
-            parent::__construct($tweet);
+            $this->tweet = $tweet;
         }
     }
 

diff --git a/lib/Twitter/Text/LooseAutolink.php b/lib/Twitter/Text/LooseAutolink.php
@@ -134,7 +134,7 @@ public function addLinks()
     public function addLinksToHashtags()
     {
         return preg_replace_callback(
-            self::$patterns['valid_hashtag'],
+            Regex::getValidHashtagMatcher(),
             array($this, '_addLinksToHashtags'),
             $this->tweet
         );
@@ -148,7 +148,7 @@ public function addLinksToHashtags()
     public function addLinksToCashtags()
     {
         return preg_replace_callback(
-            self::$patterns['valid_cashtag'],
+            Regex::getValidCashtagMatcher(),
             array($this, '_addLinksToCashtags'),
             $this->tweet
         );
@@ -161,7 +161,7 @@ public function addLinksToCashtags()
      */
     public function addLinksToURLs()
     {
-        return preg_replace_callback(self::$patterns['valid_url'], array($this, '_addLinksToURLs'), $this->tweet);
+        return preg_replace_callback(Regex::getValidUrlMatcher(), array($this, '_addLinksToURLs'), $this->tweet);
     }
 
     /**
@@ -172,7 +172,7 @@ public function addLinksToURLs()
     public function addLinksToUsernamesAndLists()
     {
         return preg_replace_callback(
-            self::$patterns['valid_mentions_or_lists'],
+            Regex::getValidMentionsOrListsMatcher(),
             array($this, '_addLinksToUsernamesAndLists'),
             $this->tweet
         );
@@ -261,15 +261,15 @@ protected function wrapHash($url, $class, $element)
     protected function _addLinksToHashtags($matches)
     {
         list($all, $before, $hash, $tag, $after) = array_pad($matches, 5, '');
-        if (preg_match(self::$patterns['end_hashtag_match'], $after)
+        if (preg_match(Regex::getEndHashtagMatcher(), $after)
             || (!preg_match('!\A["\']!', $before) && preg_match('!\A["\']!', $after)) || preg_match('!\A</!', $after)) {
             return $all;
         }
         $replacement = $before;
         $element = $hash . $tag;
         $url = $this->url_base_hash . $tag;
         $class_hash = $this->class_hash;
-        if (preg_match(self::$patterns['rtl_chars'], $element)) {
+        if (preg_match(Regex::getRtlCharsMatcher(), $element)) {
             $class_hash .= ' rtl';
         }
         $replacement .= $this->wrapHash($url, $class_hash, $element);
@@ -286,7 +286,7 @@ protected function _addLinksToHashtags($matches)
     protected function _addLinksToCashtags($matches)
     {
         list($all, $before, $cash, $tag, $after) = array_pad($matches, 5, '');
-        if (preg_match(self::$patterns['end_cashtag_match'], $after)
+        if (preg_match(Regex::getEndCashtagMatcher(), $after)
             || (!preg_match('!\A["\']!', $before) && preg_match('!\A["\']!', $after)) || preg_match('!\A</!', $after)) {
             return $all;
         }
@@ -331,7 +331,7 @@ protected function _addLinksToUsernamesAndLists($matches)
             $class = $this->class_list;
             $url = $this->url_base_list . $element;
         } else {
-            if (preg_match(self::$patterns['end_mention_match'], $after)) {
+            if (preg_match(Regex::getEndMentionMatcher(), $after)) {
                 return $all;
             }
             # Replace the username