xhtml-compiler/XHTMLCompiler/DOMFilter/Acronymizer.php

   1 <?php
   2
   3 /**
   4  * Based on a list of known acronyms, populates of the title attribute
   5  * of acronym elements in documents.
   6  */
   7 class XHTMLCompiler_DOMFilter_Acronymizer extends XHTMLCompiler_DOMFilter
   8 {
   9
  10     protected $name = 'Acronymizer';
  11
  12     /**
  13      * Associative array of recognized abbreviations/initialisms
  14      * @note To be used when each letter should be spelled out, such
  15      *       as HTML.
  16      */
  17     protected $abbreviations = array(
  18         // markup languages and related technologies
  19         'SGML' => 'Standard Generalized Markup Language',
  20         'HTML' => 'HyperText Markup Language',
  21         'XHTML' => 'Extensible HyperText Markup Language',
  22         'XML' => 'Extensible Markup Language',
  23         'RSS' => 'Really Simple Syndication',
  24         'DTD' => 'Document Type Definition',
  25         'CSS' => 'Cascading Style Sheets',
  26         // programming/apis
  27         'PHP' => 'PHP: HyperText Preprocessor',
  28         'CMS' => 'Content Management System',
  29         'API' => 'Application Programming Interface',
  30         'SVN' => 'Subversion',
  31         // web-app security
  32         'XSS' => 'Cross-Site Scripting',
  33         // organizations/groups
  34         'W3C' => 'World Wide Web Consortium',
  35         'RFC' => 'Request for Comment',
  36         'PECL' => 'PHP Extension Community Library',
  37         // character encodings
  38         'UTF-8' => '8-bit Unicode Transformation Format',
  39         // other
  40         'INI' => 'Initialization',
  41         'CPU' => 'Central Processing Unit',
  42         'LGPL' => 'Lesser GNU Public License',
  43         'URI' => 'Uniform Resource Identifier',
  44     );
  45
  46     /**
  47      * Array of recognized acronyms.
  48      * @note Acronyms can be spoken literally, if in doubt, make it
  49      *       an abbreviation.
  50      * @todo Make a public API for this, allow multiple acronym sets
  51      *       and different precedences for them.
  52      */
  53     protected $acronyms = array(
  54         // programming
  55         'SAX' => 'Simple API for XML',
  56         'DOM' => 'Document Object Module',
  57         'PEAR' => 'PHP Extension and Application Repository',
  58         'ASCII' => 'American Standard Code for Information Interchange',
  59         'SHA-1' => 'Secure Hash Algorithm',
  60         // paradigms
  61         'WYSIWYG' => 'What You See Is What You Get',
  62         'WYSIWYM' => 'What You See Is What You Mean',
  63     );
  64
  65     public function process(DOMDocument $dom, $page, $manager) {
  66         $nodes = $this->query("//html:acronym[not(@title)]");
  67         foreach ($nodes as $node) $this->addAdvisoryTitle($node, $this->acronyms);
  68
  69         $nodes = $this->query("//html:abbr[not(@title)]");
  70         foreach ($nodes as $node) $this->addAdvisoryTitle($node, $this->abbreviations);
  71
  72         // add self as dependency; when acronym lists change, so does the page
  73         $manager->addDependency(__FILE__);
  74     }
  75
  76     protected function addAdvisoryTitle($node, $lookup) {
  77         $key = $node->textContent;
  78         if (!isset($lookup[$key])) {
  79             // not fatal, but good to let the document author know
  80             trigger_error(htmlspecialchars($key) .
  81                 ' is not a recognized acronym/abbreviation (missing title attribute)');
  82             return;
  83         }
  84         $node->setAttribute('title', $lookup[$key]);
  85     }
  86
  87 }
  88
  89 ?>