includes/Metadata.php

   1 <?php
   2 /**
   3  * Metadata.php -- provides DublinCore and CreativeCommons metadata
   4  * Copyright 2004, Evan Prodromou <evan@wikitravel.org>.
   5  *
   6  *  This program is free software; you can redistribute it and/or modify
   7  *  it under the terms of the GNU General Public License as published by
   8  *  the Free Software Foundation; either version 2 of the License, or
   9  *  (at your option) any later version.
  10  *
  11  * This program is distributed in the hope that it will be useful,
  12  *  but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14  *  GNU General Public License for more details.
  15  *
  16  *  You should have received a copy of the GNU General Public License
  17  *  along with this program; if not, write to the Free Software
  18  *  Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
  19  *
  20  * @author Evan Prodromou <evan@wikitravel.org>
  21  */
  22
  23 /**
  24  * TODO: Perhaps make this file into a Metadata class, with static methods (declared
  25  * as private where indicated), to move these functions out of the global namespace?
  26  */
  27 define('RDF_TYPE_PREFS', "application/rdf+xml,text/xml;q=0.7,application/xml;q=0.5,text/rdf;q=0.1");
  28
  29 function wfDublinCoreRdf($article) {
  30
  31         $url = dcReallyFullUrl($article->mTitle);
  32
  33         if (rdfSetup()) {
  34                 dcPrologue($url);
  35                 dcBasics($article);
  36                 dcEpilogue();
  37         }
  38 }
  39
  40 function wfCreativeCommonsRdf($article) {
  41
  42         if (rdfSetup()) {
  43                 global $wgRightsUrl;
  44
  45                 $url = dcReallyFullUrl($article->mTitle);
  46
  47                 ccPrologue();
  48                 ccSubPrologue('Work', $url);
  49                 dcBasics($article);
  50                 if (isset($wgRightsUrl)) {
  51                         $url = htmlspecialchars( $wgRightsUrl );
  52                         print "    <cc:license rdf:resource=\"$url\" />\n";
  53                 }
  54
  55                 ccSubEpilogue('Work');
  56
  57                 if (isset($wgRightsUrl)) {
  58                         $terms = ccGetTerms($wgRightsUrl);
  59                         if ($terms) {
  60                                 ccSubPrologue('License', $wgRightsUrl);
  61                                 ccLicense($terms);
  62                                 ccSubEpilogue('License');
  63                         }
  64                 }
  65         }
  66
  67         ccEpilogue();
  68 }
  69
  70 /**
  71  * @private
  72  */
  73 function rdfSetup() {
  74         global $wgOut, $_SERVER;
  75
  76         $rdftype = wfNegotiateType(wfAcceptToPrefs($_SERVER['HTTP_ACCEPT']), wfAcceptToPrefs(RDF_TYPE_PREFS));
  77
  78         if (!$rdftype) {
  79                 wfHttpError(406, "Not Acceptable", wfMsg("notacceptable"));
  80                 return false;
  81         } else {
  82                 $wgOut->disable();
  83                 header( "Content-type: {$rdftype}" );
  84                 $wgOut->sendCacheControl();
  85                 return true;
  86         }
  87 }
  88
  89 /**
  90  * @private
  91  */
  92 function dcPrologue($url) {
  93         global $wgOutputEncoding;
  94
  95         $url = htmlspecialchars( $url );
  96         print "<" . "?xml version=\"1.0\" encoding=\"{$wgOutputEncoding}\" ?" . ">
  97
  98                                                                                                                                                           <!DOCTYPE rdf:RDF PUBLIC \"-//DUBLIN CORE//DCMES DTD 2002/07/31//EN\" \"http://dublincore.org/documents/2002/07/31/dcmes-xml/dcmes-xml-dtd.dtd\">
  99
 100                                                                                                                                                           <rdf:RDF xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\"
 101                                                                                                                                                           xmlns:dc=\"http://purl.org/dc/elements/1.1/\">
 102                                                                                                                                                           <rdf:Description rdf:about=\"$url\">
 103                                                                                                                                                           ";
 104 }
 105
 106 /**
 107  * @private
 108  */
 109 function dcEpilogue() {
 110         print "
 111                         </rdf:Description>
 112                         </rdf:RDF>
 113                         ";
 114 }
 115
 116 /**
 117  * @private
 118  */
 119 function dcBasics($article) {
 120         global $wgContLanguageCode, $wgSitename;
 121
 122         dcElement('title', $article->mTitle->getText());
 123         dcPageOrString('publisher', wfMsg('aboutpage'), $wgSitename);
 124         dcElement('language', $wgContLanguageCode);
 125         dcElement('type', 'Text');
 126         dcElement('format', 'text/html');
 127         dcElement('identifier', dcReallyFullUrl($article->mTitle));
 128         dcElement('date', dcDate($article->getTimestamp()));
 129
 130         $last_editor = $article->getUser();
 131
 132         if ($last_editor == 0) {
 133                 dcPerson('creator', 0);
 134         } else {
 135                 dcPerson('creator', $last_editor, $article->getUserText(),
 136                                  User::whoIsReal($last_editor));
 137         }
 138
 139         $contributors = $article->getContributors();
 140
 141         foreach ($contributors as $user_parts) {
 142                 dcPerson('contributor', $user_parts[0], $user_parts[1], $user_parts[2]);
 143         }
 144
 145         dcRights();
 146 }
 147
 148 /**
 149  * @private
 150  */
 151 function ccPrologue() {
 152         global $wgOutputEncoding;
 153
 154         echo "<" . "?xml version='1.0'  encoding='{$wgOutputEncoding}' ?" . ">
 155
 156                                                                                                                                                   <rdf:RDF xmlns:cc=\"http://web.resource.org/cc/\"
 157                                                                                                                                                   xmlns:dc=\"http://purl.org/dc/elements/1.1/\"
 158                                                                                                                                                   xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">
 159                                                                                                                                                   ";
 160 }
 161
 162 /**
 163  * @private
 164  */
 165 function ccSubPrologue($type, $url) {
 166         $url = htmlspecialchars( $url );
 167         echo "  <cc:{$type} rdf:about=\"{$url}\">\n";
 168 }
 169
 170 /**
 171  * @private
 172  */
 173 function ccSubEpilogue($type) {
 174         echo "  </cc:{$type}>\n";
 175 }
 176
 177 /**
 178  * @private
 179  */
 180 function ccLicense($terms) {
 181
 182         foreach ($terms as $term) {
 183                 switch ($term) {
 184                  case 're':
 185                         ccTerm('permits', 'Reproduction'); break;
 186                  case 'di':
 187                         ccTerm('permits', 'Distribution'); break;
 188                  case 'de':
 189                         ccTerm('permits', 'DerivativeWorks'); break;
 190                  case 'nc':
 191                         ccTerm('prohibits', 'CommercialUse'); break;
 192                  case 'no':
 193                         ccTerm('requires', 'Notice'); break;
 194                  case 'by':
 195                         ccTerm('requires', 'Attribution'); break;
 196                  case 'sa':
 197                         ccTerm('requires', 'ShareAlike'); break;
 198                  case 'sc':
 199                         ccTerm('requires', 'SourceCode'); break;
 200                 }
 201         }
 202 }
 203
 204 /**
 205  * @private
 206  */
 207 function ccTerm($term, $name) {
 208         print "    <cc:{$term} rdf:resource=\"http://web.resource.org/cc/{$name}\" />\n";
 209 }
 210
 211 /**
 212  * @private
 213  */
 214 function ccEpilogue() {
 215         echo "</rdf:RDF>\n";
 216 }
 217
 218 /**
 219  * @private
 220  */
 221 function dcElement($name, $value) {
 222         $value = htmlspecialchars( $value );
 223         print "    <dc:{$name}>{$value}</dc:{$name}>\n";
 224 }
 225
 226 /**
 227  * @private
 228  */
 229 function dcDate($timestamp) {
 230         return substr($timestamp, 0, 4) . '-'
 231           . substr($timestamp, 4, 2) . '-'
 232           . substr($timestamp, 6, 2);
 233 }
 234
 235 /**
 236  * @private
 237  */
 238 function dcReallyFullUrl($title) {
 239         return $title->getFullURL();
 240 }
 241
 242 /**
 243  * @private
 244  */
 245 function dcPageOrString($name, $page, $str) {
 246         $nt = Title::newFromText($page);
 247
 248         if (!$nt || $nt->getArticleID() == 0) {
 249                 dcElement($name, $str);
 250         } else {
 251                 dcPage($name, $nt);
 252         }
 253 }
 254
 255 /**
 256  * @private
 257  */
 258 function dcPage($name, $title) {
 259         dcUrl($name, dcReallyFullUrl($title));
 260 }
 261
 262 /**
 263  * @private
 264  */
 265 function dcUrl($name, $url) {
 266         $url = htmlspecialchars( $url );
 267         print "    <dc:{$name} rdf:resource=\"{$url}\" />\n";
 268 }
 269
 270 /**
 271  * @private
 272  */
 273 function dcPerson($name, $id, $user_name='', $user_real_name='') {
 274         global $wgContLang;
 275
 276         if ($id == 0) {
 277                 dcElement($name, wfMsg('anonymous'));
 278         } else if ( !empty($user_real_name) ) {
 279                 dcElement($name, $user_real_name);
 280         } else {
 281                 # XXX: This shouldn't happen.
 282                 if( empty( $user_name ) ) {
 283                         $user_name = User::whoIs($id);
 284                 }
 285                 dcPageOrString($name, $wgContLang->getNsText(NS_USER) . ':' . $user_name, wfMsg('siteuser', $user_name));
 286         }
 287 }
 288
 289 /**
 290  * Takes an arg, for future enhancement with different rights for
 291  * different pages.
 292  * @private
 293  */
 294 function dcRights() {
 295
 296         global $wgRightsPage, $wgRightsUrl, $wgRightsText;
 297
 298         if (isset($wgRightsPage) &&
 299                 ($nt = Title::newFromText($wgRightsPage))
 300                 && ($nt->getArticleID() != 0)) {
 301                 dcPage('rights', $nt);
 302         } else if (isset($wgRightsUrl)) {
 303                 dcUrl('rights', $wgRightsUrl);
 304         } else if (isset($wgRightsText)) {
 305                 dcElement('rights', $wgRightsText);
 306         }
 307 }
 308
 309 /**
 310  * @private
 311  */
 312 function ccGetTerms($url) {
 313         global $wgLicenseTerms;
 314
 315         if (isset($wgLicenseTerms)) {
 316                 return $wgLicenseTerms;
 317         } else {
 318                 $known = getKnownLicenses();
 319                 if( isset( $known[$url] ) ) {
 320                         return $known[$url];
 321                 } else {
 322                         return array();
 323                 }
 324         }
 325 }
 326
 327 /**
 328  * @private
 329  */
 330 function getKnownLicenses() {
 331
 332         $ccLicenses = array('by', 'by-nd', 'by-nd-nc', 'by-nc',
 333                                                 'by-nc-sa', 'by-sa');
 334         $ccVersions = array('1.0', '2.0');
 335         $knownLicenses = array();
 336
 337         foreach ($ccVersions as $version) {
 338                 foreach ($ccLicenses as $license) {
 339                         if( $version == '2.0' && substr( $license, 0, 2) != 'by' ) {
 340                                 # 2.0 dropped the non-attribs licenses
 341                                 continue;
 342                         }
 343                         $lurl = "http://creativecommons.org/licenses/{$license}/{$version}/";
 344                         $knownLicenses[$lurl] = explode('-', $license);
 345                         $knownLicenses[$lurl][] = 're';
 346                         $knownLicenses[$lurl][] = 'di';
 347                         $knownLicenses[$lurl][] = 'no';
 348                         if (!in_array('nd', $knownLicenses[$lurl])) {
 349                                 $knownLicenses[$lurl][] = 'de';
 350                         }
 351                 }
 352         }
 353
 354         /* Handle the GPL and LGPL, too. */
 355
 356         $knownLicenses['http://creativecommons.org/licenses/GPL/2.0/'] =
 357           array('de', 're', 'di', 'no', 'sa', 'sc');
 358         $knownLicenses['http://creativecommons.org/licenses/LGPL/2.1/'] =
 359           array('de', 're', 'di', 'no', 'sa', 'sc');
 360         $knownLicenses['http://www.gnu.org/copyleft/fdl.html'] =
 361           array('de', 're', 'di', 'no', 'sa', 'sc');
 362
 363         return $knownLicenses;
 364 }
 365
 366 ?>