Changeset 33292

Timestamp:

07/16/2015 09:44:43 AM (9 years ago)

Author:

iseulde

Message:

Editor: word count: exclude more characters

Also only exclude these characters for the words type. They should be counted for other types.
Add the ASCIIOnly option to the uglify config to preserve escaped unicode characters.

See #30966. Fixes #27391.

Location:

trunk

Files:

• Gruntfile.js (modified) (1 diff)
• src/wp-admin/js/word-count.js (modified) (2 diffs)
• tests/qunit/wp-admin/js/word-count.js (modified) (1 diff)

trunk/Gruntfile.js

@@ -400 +400 @@
         },
         uglify: {
+            options: {
+                ASCIIOnly: true
+            },
             core: {
                 expand: true,

trunk/src/wp-admin/js/word-count.js

@@ -15 +15 @@
         HTMLRegExp: /<\/?[a-z][^>]*?>/gi,
         spaceRegExp: /&nbsp;|&#160;/gi,
-        removeRegExp: /[0-9.(),;:!?%#$¿'"_+=\\\/-]+/g,
+        connectorRegExp: /--|\u2014/gi,
+        removeRegExp: new RegExp( [
+            '[',
+                // Basic Latin (extract)
+                '\u0021-\u0040\u005B-\u0060\u007B-\u007E',
+                // Latin-1 Supplement (extract)
+                '\u0080-\u00BF\u00D7\u00F7',
+                // General Punctuation
+                // Superscripts and Subscripts
+                // Currency Symbols
+                // Combining Diacritical Marks for Symbols
+                // Letterlike Symbols
+                // Number Forms
+                // Arrows
+                // Mathematical Operators
+                // Miscellaneous Technical
+                // Control Pictures
+                // Optical Character Recognition
+                // Enclosed Alphanumerics
+                // Box Drawing
+                // Block Elements
+                // Geometric Shapes
+                // Miscellaneous Symbols
+                // Dingbats
+                // Miscellaneous Mathematical Symbols-A
+                // Supplemental Arrows-A
+                // Braille Patterns
+                // Supplemental Arrows-B
+                // Miscellaneous Mathematical Symbols-B
+                // Supplemental Mathematical Operators
+                // Miscellaneous Symbols and Arrows
+                '\u2000-\u2BFF',
+                // Supplemental Punctuation
+                '\u2E00-\u2E7F',
+            ']'
+        ].join( '' ), 'g' ),
         wordsRegExp: /\S\s+/g,
         charactersRegExp: /\S/g,
@@ -32 +67 @@
             text = text.replace( this.settings.HTMLRegExp, '\n' );
             text = text.replace( this.settings.spaceRegExp, ' ' );
-            text = text.replace( this.settings.removeRegExp, '' );
+
+            if ( type === 'words' ) {
+                text = text.replace( this.settings.connectorRegExp, ' ' );
+                text = text.replace( this.settings.removeRegExp, '' );
+            }
 
             text = text.match( this.settings[ type + 'RegExp' ] );

trunk/tests/qunit/wp-admin/js/word-count.js

@@ -34 +34 @@
             {
                 message: 'Punctuation.',
-                string: 'It\'s two three... 4?',
+                string: 'It\'s two three \u2026 4?',
                 words: 3,
-                characters: 11,
+                characters: 15,
+                all: 19
+            },
+            {
+                message: 'Em dash.',
+                string: 'one\u2014two--three',
+                words: 3,
+                characters: 14,
                 all: 14
             }