includes/site/SiteImporter.php

   1 <?php
   2 /**
   3  * This program is free software; you can redistribute it and/or modify
   4  * it under the terms of the GNU General Public License as published by
   5  * the Free Software Foundation; either version 2 of the License, or
   6  * (at your option) any later version.
   7  *
   8  * This program is distributed in the hope that it will be useful,
   9  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  10  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  11  * GNU General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU General Public License along
  14  * with this program; if not, write to the Free Software Foundation, Inc.,
  15  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  16  * http://www.gnu.org/copyleft/gpl.html
  17  *
  18  * @file
  19  */
  20
  21 namespace MediaWiki\Site;
  22
  23 use DOMDocument;
  24 use DOMElement;
  25 use Exception;
  26 use InvalidArgumentException;
  27 use RuntimeException;
  28 use Wikimedia\RequestTimeout\TimeoutException;
  29
  30 /**
  31  * Utility for importing site entries from XML.
  32  *
  33  * For the expected format of the input, see docs/sitelist.md and docs/sitelist-1.0.xsd.
  34  *
  35  * @since 1.25
  36  * @ingroup Site
  37  * @author Daniel Kinzler
  38  */
  39 class SiteImporter {
  40
  41         /**
  42          * @var SiteStore
  43          */
  44         private $store;
  45
  46         /**
  47          * @var callable|null
  48          */
  49         private $exceptionCallback;
  50
  51         public function __construct( SiteStore $store ) {
  52                 $this->store = $store;
  53         }
  54
  55         /**
  56          * @return callable
  57          */
  58         public function getExceptionCallback() {
  59                 return $this->exceptionCallback;
  60         }
  61
  62         /**
  63          * @param callable $exceptionCallback
  64          */
  65         public function setExceptionCallback( $exceptionCallback ) {
  66                 $this->exceptionCallback = $exceptionCallback;
  67         }
  68
  69         /**
  70          * @param string $file
  71          */
  72         public function importFromFile( $file ) {
  73                 $xml = file_get_contents( $file );
  74
  75                 if ( $xml === false ) {
  76                         throw new RuntimeException( 'Failed to read ' . $file . '!' );
  77                 }
  78
  79                 $this->importFromXML( $xml );
  80         }
  81
  82         /**
  83          * @param string $xml
  84          *
  85          */
  86         public function importFromXML( $xml ) {
  87                 $document = new DOMDocument();
  88
  89                 $oldLibXmlErrors = libxml_use_internal_errors( true );
  90                 // phpcs:ignore Generic.PHP.NoSilencedErrors -- suppress deprecation per T268847
  91                 $oldDisable = @libxml_disable_entity_loader( true );
  92                 $ok = $document->loadXML( $xml, LIBXML_NONET );
  93
  94                 if ( !$ok ) {
  95                         $errors = libxml_get_errors();
  96                         libxml_use_internal_errors( $oldLibXmlErrors );
  97                         // phpcs:ignore Generic.PHP.NoSilencedErrors
  98                         @libxml_disable_entity_loader( $oldDisable );
  99
 100                         foreach ( $errors as $error ) {
 101                                 /** @var LibXMLError $error */
 102                                 throw new InvalidArgumentException(
 103                                         'Malformed XML: ' . $error->message . ' in line ' . $error->line
 104                                 );
 105                         }
 106
 107                         throw new InvalidArgumentException( 'Malformed XML!' );
 108                 }
 109
 110                 libxml_use_internal_errors( $oldLibXmlErrors );
 111                 // phpcs:ignore Generic.PHP.NoSilencedErrors
 112                 @libxml_disable_entity_loader( $oldDisable );
 113                 $sites = $this->makeSiteList( $document->documentElement );
 114                 $this->store->saveSites( $sites );
 115         }
 116
 117         /**
 118          * @param DOMElement $root
 119          *
 120          * @return Site[]
 121          */
 122         private function makeSiteList( DOMElement $root ) {
 123                 $sites = [];
 124
 125                 // Old sites, to get the row IDs that correspond to the global site IDs.
 126                 // TODO: Get rid of internal row IDs, they just get in the way. Get rid of ORMRow, too.
 127                 $oldSites = $this->store->getSites();
 128
 129                 $current = $root->firstChild;
 130                 while ( $current ) {
 131                         if ( $current instanceof DOMElement && $current->tagName === 'site' ) {
 132                                 try {
 133                                         $site = $this->makeSite( $current );
 134                                         $key = $site->getGlobalId();
 135
 136                                         if ( $oldSites->hasSite( $key ) ) {
 137                                                 $oldSite = $oldSites->getSite( $key );
 138                                                 $site->setInternalId( $oldSite->getInternalId() );
 139                                         }
 140
 141                                         $sites[$key] = $site;
 142                                 } catch ( TimeoutException $e ) {
 143                                         throw $e;
 144                                 } catch ( Exception $ex ) {
 145                                         $this->handleException( $ex );
 146                                 }
 147                         }
 148
 149                         $current = $current->nextSibling;
 150                 }
 151
 152                 return $sites;
 153         }
 154
 155         /**
 156          * @param DOMElement $siteElement
 157          *
 158          * @return Site
 159          */
 160         public function makeSite( DOMElement $siteElement ) {
 161                 if ( $siteElement->tagName !== 'site' ) {
 162                         throw new InvalidArgumentException( 'Expected <site> tag, found ' . $siteElement->tagName );
 163                 }
 164
 165                 $type = $this->getAttributeValue( $siteElement, 'type', Site::TYPE_UNKNOWN );
 166                 $site = Site::newForType( $type );
 167
 168                 $site->setForward( $this->hasChild( $siteElement, 'forward' ) );
 169                 $site->setGlobalId( $this->getChildText( $siteElement, 'globalid' ) );
 170                 $site->setGroup( $this->getChildText( $siteElement, 'group', Site::GROUP_NONE ) );
 171                 $site->setSource( $this->getChildText( $siteElement, 'source', Site::SOURCE_LOCAL ) );
 172
 173                 $pathTags = $siteElement->getElementsByTagName( 'path' );
 174                 for ( $i = 0; $i < $pathTags->length; $i++ ) {
 175                         $pathElement = $pathTags->item( $i );
 176                         '@phan-var DOMElement $pathElement';
 177                         $pathType = $this->getAttributeValue( $pathElement, 'type' );
 178                         $path = $pathElement->textContent;
 179
 180                         $site->setPath( $pathType, $path );
 181                 }
 182
 183                 $idTags = $siteElement->getElementsByTagName( 'localid' );
 184                 for ( $i = 0; $i < $idTags->length; $i++ ) {
 185                         $idElement = $idTags->item( $i );
 186                         '@phan-var DOMElement $idElement';
 187                         $idType = $this->getAttributeValue( $idElement, 'type' );
 188                         $id = $idElement->textContent;
 189
 190                         $site->addLocalId( $idType, $id );
 191                 }
 192
 193                 // @todo: import <data>
 194                 // @todo: import <config>
 195
 196                 return $site;
 197         }
 198
 199         /**
 200          * @param DOMElement $element
 201          * @param string $name
 202          * @param string|null|false $default
 203          *
 204          * @return null|string
 205          */
 206         private function getAttributeValue( DOMElement $element, $name, $default = false ) {
 207                 $node = $element->getAttributeNode( $name );
 208
 209                 if ( !$node ) {
 210                         if ( $default !== false ) {
 211                                 return $default;
 212                         } else {
 213                                 throw new RuntimeException(
 214                                         'Required ' . $name . ' attribute not found in <' . $element->tagName . '> tag'
 215                                 );
 216                         }
 217                 }
 218
 219                 return $node->textContent;
 220         }
 221
 222         /**
 223          * @param DOMElement $element
 224          * @param string $name
 225          * @param string|null|false $default
 226          *
 227          * @return null|string
 228          */
 229         private function getChildText( DOMElement $element, $name, $default = false ) {
 230                 $elements = $element->getElementsByTagName( $name );
 231
 232                 if ( $elements->length < 1 ) {
 233                         if ( $default !== false ) {
 234                                 return $default;
 235                         } else {
 236                                 throw new RuntimeException(
 237                                         'Required <' . $name . '> tag not found inside <' . $element->tagName . '> tag'
 238                                 );
 239                         }
 240                 }
 241
 242                 $node = $elements->item( 0 );
 243                 return $node->textContent;
 244         }
 245
 246         /**
 247          * @param DOMElement $element
 248          * @param string $name
 249          *
 250          * @return bool
 251          */
 252         private function hasChild( DOMElement $element, $name ) {
 253                 return $this->getChildText( $element, $name, null ) !== null;
 254         }
 255
 256         private function handleException( Exception $ex ) {
 257                 if ( $this->exceptionCallback ) {
 258                         call_user_func( $this->exceptionCallback, $ex );
 259                 } else {
 260                         wfLogWarning( $ex->getMessage() );
 261                 }
 262         }
 263
 264 }
 265
 266 /** @deprecated class alias since 1.42 */
 267 class_alias( SiteImporter::class, 'SiteImporter' );