XHTMLCompiler/DOMFilter/Acronymizer.php

   1 <?php
   2
   3 /**
   4  * Based on a list of known acronyms, populates of the title attribute
   5  * of acronym and abbr elements in documents.
   6  * @note Title is a bit of legacy baggage.
   7  */
   8 class XHTMLCompiler_DOMFilter_Acronymizer extends XHTMLCompiler_DOMFilter
   9 {
  10
  11     protected $name = 'Acronymizer';
  12
  13     /**
  14      * Associative array of recognized abbreviations/initialisms
  15      * @note To be used when each letter should be spelled out, such
  16      *       as HTML.
  17      */
  18     protected $abbreviations = array(
  19         // markup languages and related technologies
  20         'SGML' => 'Standard Generalized Markup Language',
  21         'HTML' => 'HyperText Markup Language',
  22         'XHTML' => 'Extensible HyperText Markup Language',
  23         'XML' => 'Extensible Markup Language',
  24         'RSS' => 'Really Simple Syndication',
  25         'DTD' => 'Document Type Definition',
  26         'CSS' => 'Cascading Style Sheets',
  27         'HTTP' => 'HyperText Transfer Protocol',
  28         // programming/apis
  29         'PHP' => 'PHP: HyperText Preprocessor',
  30         'CMS' => 'Content Management System',
  31         'API' => 'Application Programming Interface',
  32         'SVN' => 'Subversion',
  33         'XSLT' => 'Extensible Stylesheet Language Transformations',
  34         'SQL' => 'Structured Query Language',
  35         // web-app security
  36         'XSS' => 'Cross-Site Scripting',
  37         // organizations/groups
  38         'W3C' => 'World Wide Web Consortium',
  39         'RFC' => 'Request for Comment',
  40         'PECL' => 'PHP Extension Community Library',
  41         // character encodings
  42         'UTF-8' => '8-bit Unicode Transformation Format',
  43         // other
  44         'INI' => 'Initialization',
  45         'CPU' => 'Central Processing Unit',
  46         'LGPL' => 'Lesser GNU Public License',
  47         'FTP' => 'File Transfer Protocol',
  48         'URI' => 'Uniform Resource Identifier',
  49     );
  50
  51     /**
  52      * Array of recognized acronyms.
  53      * @note Acronyms can be spoken literally, if in doubt, make it
  54      *       an abbreviation.
  55      * @todo Make a public API for this, allow multiple acronym sets
  56      *       and different precedences for them.
  57      */
  58     protected $acronyms = array(
  59         // programming
  60         'SAX' => 'Simple API for XML',
  61         'DOM' => 'Document Object Module',
  62         'PEAR' => 'PHP Extension and Application Repository',
  63         'ASCII' => 'American Standard Code for Information Interchange',
  64         'SHA-1' => 'Secure Hash Algorithm',
  65         // paradigms
  66         'WYSIWYG' => 'What You See Is What You Get',
  67         'WYSIWYM' => 'What You See Is What You Mean',
  68     );
  69
  70     public function process(DOMDocument $dom, $page, $manager) {
  71         $nodes = $this->query("//html:acronym[not(@title)]");
  72         foreach ($nodes as $node) $this->addAdvisoryTitle($node, $this->acronyms);
  73
  74         $nodes = $this->query("//html:abbr[not(@title)]");
  75         foreach ($nodes as $node) $this->addAdvisoryTitle($node, $this->abbreviations);
  76
  77         // add self as dependency; when acronym lists change, so does the page
  78         $manager->addDependency(__FILE__);
  79     }
  80
  81     protected function addAdvisoryTitle($node, $lookup) {
  82         $key = $node->textContent;
  83         if (!isset($lookup[$key])) {
  84             // not fatal, but good to let the document author know
  85             trigger_error(htmlspecialchars($key) .
  86                 ' is not a recognized acronym/abbreviation (missing title attribute)');
  87             return;
  88         }
  89         $node->setAttribute('title', $lookup[$key]);
  90     }
  91
  92 }