Ticket #29256: miqro-29256.2.patch

src/wp-includes/formatting.php

@@ -30 +30 @@
 function wptexturize($text, $reset = false) {
         global $wp_cockneyreplace;
         static $static_characters, $static_replacements, $dynamic_characters, $dynamic_replacements,
-                $default_no_texturize_tags, $default_no_texturize_shortcodes, $run_texturize = true;
+                $default_no_texturize_tags, $default_no_texturize_shortcodes, $run_texturize = true,
+                $apos_flag, $apos, $prime, $double_prime, $opening_quote, $closing_quote, $opening_single_quote,
+                $closing_single_quote;
 
         // If there's nothing to do, just stop.
         if ( empty( $text ) || false === $run_texturize ) {
@@ -105 +107 @@
                 $dynamic_replacements = array( 'apos' => array(), 'quote' => array(), 'dash' => array() );
                 $dynamic = array();
                 $spaces = wp_spaces_regexp();
+                $apos_flag = '<!--wp-apos-->';  // We need a semantic representation that is not identical to a quote.
 
                 // '99' and '99" are ambiguous among other patterns; assume it's an abbreviated year at the end of a quotation.
                 if ( "'" !== $apos || "'" !== $closing_single_quote ) {
-                        $dynamic[ '/\'(\d\d)\'(?=\Z|[.,)}\-\]]|&gt;|' . $spaces . ')/' ] = $apos . '$1' . $closing_single_quote;
+                        $dynamic[ '/\'(\d\d)\'(?=\Z|[.,)}\-\]]|&gt;|' . $spaces . ')/' ] = $apos_flag . '$1' . $closing_single_quote;
                 }
                 if ( "'" !== $apos || '"' !== $closing_quote ) {
-                        $dynamic[ '/\'(\d\d)"(?=\Z|[.,)}\-\]]|&gt;|' . $spaces . ')/' ] = $apos . '$1' . $closing_quote;
+                        $dynamic[ '/\'(\d\d)"(?=\Z|[.,)}\-\]]|&gt;|' . $spaces . ')/' ] = $apos_flag . '$1' . $closing_quote;
                 }
 
                 // '99 '99s '99's (apostrophe)  But never '9 or '99% or '999 or '99.0.
                 if ( "'" !== $apos ) {
-                        $dynamic[ '/\'(?=\d\d(?:\Z|(?![%\d]|[.,]\d)))/' ] = $apos;
+                        $dynamic[ '/\'(?=\d\d(?:\Z|(?![%\d]|[.,]\d)))/' ] = $apos_flag;
                 }
 
                 // Quoted Numbers like '0.42'
@@ -131 +134 @@
 
                 // Apostrophe in a word.  No spaces, double apostrophes, or other punctuation.
                 if ( "'" !== $apos ) {
-                        $dynamic[ '/(?<!' . $spaces . ')\'(?!\Z|[.,:;"\'(){}[\]\-]|&[lg]t;|' . $spaces . ')/' ] = $apos;
+                        $dynamic[ '/(?<!' . $spaces . ')\'(?!\Z|[.,:;"\'(){}[\]\-]|&[lg]t;|' . $spaces . ')/' ] = $apos_flag;
                 }
 
-                // 9' (prime)
-                if ( "'" !== $prime ) {
-                        $dynamic[ '/(?<=\d)\'/' ] = $prime;
-                }
-
-                // Single quotes followed by spaces or ending punctuation.
-                if ( "'" !== $closing_single_quote ) {
-                        $dynamic[ '/\'(?=\Z|[.,)}\-\]]|&gt;|' . $spaces . ')/' ] = $closing_single_quote;
-                }
-
                 $dynamic_characters['apos'] = array_keys( $dynamic );
                 $dynamic_replacements['apos'] = array_values( $dynamic );
                 $dynamic = array();
@@ -153 +146 @@
                         $dynamic[ '/(?<=\A|' . $spaces . ')"(\d[.,\d]*)"/' ] = $opening_quote . '$1' . $closing_quote;
                 }
 
-                // 9" (double prime)
-                if ( '"' !== $double_prime ) {
-                        $dynamic[ '/(?<=\d)"/' ] = $double_prime;
-                }
-
                 // Double quote at start, or preceded by (, {, <, [, -, or spaces, and not followed by spaces.
                 if ( '"' !== $opening_quote ) {
                         $dynamic[ '/(?<=\A|[([{\-]|&lt;|' . $spaces . ')"(?!' . $spaces . ')/' ] = $opening_quote;
                 }
 
-                // Any remaining double quotes.
-                if ( '"' !== $closing_quote ) {
-                        $dynamic[ '/"/' ] = $closing_quote;
-                }
-                
                 $dynamic_characters['quote'] = array_keys( $dynamic );
                 $dynamic_replacements['quote'] = array_values( $dynamic );
                 $dynamic = array();
@@ -261 +244 @@
 
                         if ( false !== strpos( $curl, "'" ) ) {
                                 $curl = preg_replace( $dynamic_characters['apos'], $dynamic_replacements['apos'], $curl );
+                                $curl = wptexturize_primes( $curl, "'", $prime, $opening_single_quote, $closing_single_quote );
+                                $curl = str_replace( $apos_flag, $apos, $curl );
                         }
                         if ( false !== strpos( $curl, '"' ) ) {
                                 $curl = preg_replace( $dynamic_characters['quote'], $dynamic_replacements['quote'], $curl );
+                                $curl = wptexturize_primes( $curl, '"', $double_prime, $opening_quote, $closing_quote );
                         }
                         if ( false !== strpos( $curl, '-' ) ) {
                                 $curl = preg_replace( $dynamic_characters['dash'], $dynamic_replacements['dash'], $curl );
@@ -285 +271 @@
 }
 
 /**
+ * Implements a logic tree to determine whether or not "7'." represents seven feet,
+ * then converts the special char into either a prime char or a closing quote char.
+ *
+ * @since 4.1.0
+ *
+ * @param string $haystack The plain text to be searched.
+ * @param string $needle The character to search for such as ' or ".
+ * @param string $prime The prime char to use for replacement.
+ * @param string $open_quote The opening quote char. Opening quote replacement must be accomplished already.
+ * @param string $close_quote The closing quote char to use for replacement.
+ * @return string The $haystack value after primes and quotes replacements.
+ */
+function wptexturize_primes( $haystack, $needle, $prime, $open_quote, $close_quote ) {
+
+        $spaces = wp_spaces_regexp();
+        $flag = '<!--wp-prime-or-quote-->';
+        $quote_pattern = "/$needle(?=\\Z|[.,)}\\-\\]]|&gt;|" . $spaces . ")/";
+        $prime_pattern    = "/(?<=\\d)$needle/";
+        $flag_after_digit = "/(?<=\\d)$flag/";
+        $flag_no_digit    = "/(?<!\\d)$flag/";
+
+        if ( $open_quote == $close_quote ) {
+                // Algorithm assumes some quotes were already translated, replaced, and are not identical.
+                return $haystack;
+        }
+        
+        $sentences = explode( $open_quote, $haystack );
+
+        foreach( $sentences as $key => &$sentence ) {
+                if ( false === strpos( $sentence, $needle ) ) {
+                        continue;
+                } elseif ( 0 !== $key && substr_count( $sentence, $close_quote ) === 0 ) {
+                        $sentence = preg_replace( $quote_pattern, $flag, $sentence, -1, $count );
+                        if ( $count > 1) {
+                                // This sentence appears to have multiple closing quotes.  Attempt Vulcan logic.
+                                $sentence = preg_replace( $flag_no_digit, $close_quote, $sentence, -1, $count2 );
+                                if ( 0 === $count2 ) {
+                                        // Closing quote still ambiguous.  Look for a quote followed by a period.
+                                        $count2 = substr_count( $sentence, "$flag." );
+                                        if ( $count2 > 0 ) {
+                                                // Assume the rightmost quote-period match is the end of quotation.
+                                                $pos = strrpos( $sentence, "$flag." );
+                                        } else {
+                                                // When all else fails, make the rightmost candidate a closing quote.
+                                                // This is most likely to be problematic in the context of bug #18549.
+                                                $pos = strrpos( $sentence, $flag );
+                                        }
+                                        $sentence = substr_replace( $sentence, $close_quote, $pos, strlen( $flag ) );
+                                }
+                                // Use conventional replacement on any remaining primes and quotes.
+                                $sentence = preg_replace( $prime_pattern, $prime, $sentence );
+                                $sentence = preg_replace( $flag_after_digit, $prime, $sentence );
+                                $sentence = str_replace( $flag, $close_quote, $sentence );
+                        } elseif ( 1 == $count ) {
+                                // Found only one closing quote candidate, so give it priority over primes.
+                                $sentence = str_replace( $flag, $close_quote, $sentence );
+                                $sentence = preg_replace( $prime_pattern, $prime, $sentence );
+                        } else {
+                                // No closing quotes found.  Just run primes pattern.
+                                $sentence = preg_replace( $prime_pattern, $prime, $sentence );
+                        }
+                } else {
+                        $sentence = preg_replace( $prime_pattern, $prime, $sentence );
+                        $sentence = preg_replace( $quote_pattern, $close_quote, $sentence );
+                }
+                if ( '"' == $needle ) {
+                        $sentence = str_replace( '"', $close_quote, $sentence );
+                }
+        }
+
+        return implode( $open_quote, $sentences );
+}
+
+/**
  * Search for disabled element tags. Push element to stack on tag open and pop
  * on tag close.
  *

tests/phpunit/tests/formatting/WPTexturize.php

@@ -91 +91 @@
                 //$this->assertEquals('Here is &#8220;<a href="http://example.com">a test with a link</a>&#8221;&#8230; and ellipses.', wptexturize('Here is "<a href="http://example.com">a test with a link</a>"... and ellipses.'));
                 //$this->assertEquals('Here is &#8220;a test <a href="http://example.com">with a link</a>&#8221;.', wptexturize('Here is "a test <a href="http://example.com">with a link</a>".'));
                 //$this->assertEquals('Here is &#8220;<a href="http://example.com">a test with a link</a>&#8221;and a work stuck to the end.', wptexturize('Here is "<a href="http://example.com">a test with a link</a>"and a work stuck to the end.'));
-                //$this->assertEquals('A test with a finishing number, &#8220;like 23&#8221;.', wptexturize('A test with a finishing number, "like 23".'));
-                //$this->assertEquals('A test with a number, &#8220;like 62&#8221;, is nice to have.', wptexturize('A test with a number, "like 62", is nice to have.'));
+                $this->assertEquals('A test with a finishing number, &#8220;like 23&#8221;.', wptexturize('A test with a finishing number, "like 23".'));
+                $this->assertEquals('A test with a number, &#8220;like 62&#8221;, is nice to have.', wptexturize('A test with a number, "like 62", is nice to have.'));
         }
 
         /**
@@ -115 +115 @@
                 $this->assertEquals('&#8216;Class of &#8217;99&#8217;', wptexturize("'Class of '99'"));
                 $this->assertEquals('&#8216;Class of &#8217;99&#8217;s&#8217;', wptexturize("'Class of '99's'"));
                 $this->assertEquals('&#8216;Class of &#8217;99&#8217;s&#8217;', wptexturize("'Class of '99&#8217;s'"));
-                //$this->assertEquals('&#8220;Class of 99&#8221;', wptexturize("\"Class of 99\""));
+                $this->assertEquals('&#8220;Class of 99&#8221;', wptexturize("\"Class of 99\""));
                 $this->assertEquals('&#8220;Class of &#8217;99&#8221;', wptexturize("\"Class of '99\""));
                 $this->assertEquals('{&#8220;Class of &#8217;99&#8221;}', wptexturize("{\"Class of '99\"}"));
                 $this->assertEquals(' &#8220;Class of &#8217;99&#8221; ', wptexturize(" \"Class of '99\" "));
@@ -1727 +1727 @@
                         ),
                 );
         }
+
+        /**
+         * Ensure primes logic is not too greedy at the end of a quotation.
+         *
+         * @ticket 29256
+         * @dataProvider primes_vs_quotes
+         */
+        function test_primes_vs_quotes( $input, $output ) {
+                return $this->assertEquals( $output, wptexturize( $input ) );
+        }
+
+        function data_primes_vs_quotes() {
+                return array(
+                        array(
+                                "George's porch is 99' long.",
+                                "George&#8217;s porch is 99&#8242; long.",
+                        ),
+                        array(
+                                'The best year "was that time in 2012" when everyone partied, he said.',
+                                'The best year &#8220;was that time in 2012&#8221; when everyone partied, he said.',
+                        ),
+                        array(
+                                "I need 4 x 20' = 80' of trim.", // Works only with a space before the = char.
+                                "I need 4 x 20&#8242; = 80&#8242; of trim.",
+                        ),
+                        array(
+                                '"Lorem ipsum dolor sit amet 1234"',
+                                '&#8220;Lorem ipsum dolor sit amet 1234&#8221;',
+                        ),
+                        array(
+                                "'Etiam eu egestas dui 1234'",
+                                "&#8216;Etiam eu egestas dui 1234&#8217;",
+                        ),
+                        array(
+                                'according to our source, "33% of all students scored less than 50" on the test.',
+                                'according to our source, &#8220;33% of all students scored less than 50&#8221; on the test.',
+                        ),
+                        array(
+                                "The doctor said, 'An average height is between 5' and 6' in study group 7'.  He then produced a 6' chart of averages.  A man of 7', incredibly, is very possible.",
+                                "The doctor said, &#8216;An average height is between 5&#8242; and 6&#8242; in study group 7&#8217;.  He then produced a 6&#8242; chart of averages.  A man of 7&#8242;, incredibly, is very possible.",
+                        ),
+                        array(
+                                'Pirates have voted on "The Expendables 3" with their clicks -- and it turns out the Sylvester Stallone-starrer hasn\'t been astoundingly popular among digital thieves, relatively speaking.
+
+As of Sunday, 5.12 million people worldwide had pirated "Expendables 3" since a high-quality copy hit torrent-sharing sites July 23, according to piracy-tracking firm Excipio.
+
+That likely contributed to the action movie\'s dismal box-office debut this weekend. But over the same July 23-Aug. 18 time period, the movie was No. 4 in downloads, after "Captain America: The Winter Soldier" (7.31 million), "Divergent" (6.29 million) and "The Amazing Spider-Man 2" (5.88 million). Moreover, that\'s despite "Expendables 3" becoming available more than three weeks prior to the film\'s U.S. theatrical debut.
+
+String with a number followed by a single quote \'Expendables 3\' vestibulum in arcu mi.',
+
+                                'Pirates have voted on &#8220;The Expendables 3&#8221; with their clicks &#8212; and it turns out the Sylvester Stallone-starrer hasn&#8217;t been astoundingly popular among digital thieves, relatively speaking.
+
+As of Sunday, 5.12 million people worldwide had pirated &#8220;Expendables 3&#8221; since a high-quality copy hit torrent-sharing sites July 23, according to piracy-tracking firm Excipio.
+
+That likely contributed to the action movie&#8217;s dismal box-office debut this weekend. But over the same July 23-Aug. 18 time period, the movie was No. 4 in downloads, after &#8220;Captain America: The Winter Soldier&#8221; (7.31 million), &#8220;Divergent&#8221; (6.29 million) and &#8220;The Amazing Spider-Man 2&#8221; (5.88 million). Moreover, that&#8217;s despite &#8220;Expendables 3&#8221; becoming available more than three weeks prior to the film&#8217;s U.S. theatrical debut.
+
+String with a number followed by a single quote &#8216;Expendables 3&#8217; vestibulum in arcu mi.',
+                        ),
+                );
+        }
 }
+ No newline at end of file