Supply the image alt text when converting images

In order to provide an accessible version of HTML content, where an alt tag
has been provided, this should be used.

Where text refers to an image inline (for example, an emoticon), complete
removal of that image can confuse the meaning of the text. As per WCAG 2.0
guidelines (http://www.w3.org/TR/WCAG20-TECHS/H37.html), images which
convey words should be provided with alt text which inclues those words.

This also relates to images which convey meaning in some other fashion.

We have been using this modification to Html2Text since August 2010.

src/Html2Text.php

@@ -67,6 +67,7 @@ class Html2Text
         '/(<tr[^>]*>|<\/tr>)/i',                          // <tr> and </tr>
         '/<td[^>]*>(.*?)<\/td>/i',                        // <td> and </td>
         '/<span class="_html2text_ignore">.+?<\/span>/i', // <span class="_html2text_ignore">...</span>
+        '/<(img)[^>]*alt=\"([^>"]+)\"[^>]*>/i',           // <img> with alt tag
     );
 
     /**
@@ -97,7 +98,8 @@ class Html2Text
         "\n\n",                          // <table> and </table>
         "\n",                            // <tr> and </tr>
         "\t\t\\1\n",                     // <td> and </td>
-        ""                               // <span class="_html2text_ignore">...</span>
+        "",                              // <span class="_html2text_ignore">...</span>
+        '[\\2]',                         // <img> with alt tag
     );
 
     /**

test/ImageTest.php

@@ -0,0 +1,42 @@
+<?php
+
+namespace Html2Text;
+
+class ImageTest extends \PHPUnit_Framework_TestCase
+{
+    public function testImageDataProvider() {
+        return array(
+            'Without alt tag' => array(
+                'html' => '<img src="http://example.com/example.jpg">',
+                'expected'  => '',
+            ),
+            'Without alt tag, wrapped in text' => array(
+                'html' => 'xx<img src="http://example.com/example.jpg">xx',
+                'expected'  => 'xxxx',
+            ),
+            'With alt tag' => array(
+                'html' => '<img src="http://example.com/example.jpg" alt="An example image">',
+                'expected'  => '[An example image]',
+            ),
+            'With alt, and title tags' => array(
+                'html' => '<img src="http://example.com/example.jpg" alt="An example image" title="Should be ignored">',
+                'expected'  => '[An example image]',
+            ),
+            'With alt tag, wrapped in text' => array(
+                'html' => 'xx<img src="http://example.com/example.jpg" alt="An example image">xx',
+                'expected'  => 'xx[An example image]xx',
+            ),
+        );
+    }
+
+    /**
+     * @dataProvider testImageDataProvider
+     */
+    public function testImages($html, $expected)
+    {
+        $html2text = new Html2Text($html);
+        $output = $html2text->getText();
+
+        $this->assertEquals($expected, $output);
+    }
+}