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         // web-app security
  31         'XSS' => 'Cross-Site Scripting',
  32         // organizations/groups
  33         'W3C' => 'World Wide Web Consortium',
  34         'RFC' => 'Request for Comment',
  35         'PECL' => 'PHP Extension Community Library',
  36         // character encodings
  37         'UTF-8' => '8-bit Unicode Transformation Format',
  38         // other
  39         'INI' => 'Initialization',
  40         'CPU' => 'Central Processing Unit',
  41         'LGPL' => 'Lesser GNU Public License',
  42         'URI' => 'Uniform Resource Identifier',
  43     );
  44
  45     /**
  46      * Array of recognized acronyms.
  47      * @note Acronyms can be spoken literally, if in doubt, make it
  48      *       an abbreviation.
  49      * @todo Make a public API for this, allow multiple acronym sets
  50      *       and different precedences for them.
  51      */
  52     protected $acronyms = array(
  53         // programming
  54         'SAX' => 'Simple API for XML',
  55         'DOM' => 'Document Object Module',
  56         'PEAR' => 'PHP Extension and Application Repository',
  57         'ASCII' => 'American Standard Code for Information Interchange',
  58         'SHA-1' => 'Secure Hash Algorithm',
  59         // paradigms
  60         'WYSIWYG' => 'What You See Is What You Get',
  61         'WYSIWYM' => 'What You See Is What You Mean',
  62     );
  63
  64     public function process(DOMDocument $dom, $page, $manager) {
  65         $nodes = $this->query("//html:acronym[not(@title)]");
  66         foreach ($nodes as $node) $this->addAdvisoryTitle($node, $this->acronyms);
  67
  68         $nodes = $this->query("//html:abbr[not(@title)]");
  69         foreach ($nodes as $node) $this->addAdvisoryTitle($node, $this->abbreviations);
  70
  71         // add self as dependency; when acronym lists change, so does the page
  72         $manager->addDependency(__FILE__);
  73     }
  74
  75     protected function addAdvisoryTitle($node, $lookup) {
  76         $key = $node->textContent;
  77         if (!isset($lookup[$key])) {
  78             // not fatal, but good to let the document author know
  79             trigger_error(htmlspecialchars($key) .
  80                 ' is not a recognized acronym/abbreviation (missing title attribute)');
  81             return;
  82         }
  83         $node->setAttribute('title', $lookup[$key]);
  84     }
  85
  86 }
  87
  88 ?>