includes/site/SiteImporter.php

   1 <?php
   2
   3 /**
   4  * Utility for importing site entries from XML.
   5  * For the expected format of the input, see docs/sitelist.txt and docs/sitelist-1.0.xsd.
   6  *
   7  * This program is free software; you can redistribute it and/or modify
   8  * it under the terms of the GNU General Public License as published by
   9  * the Free Software Foundation; either version 2 of the License, or
  10  * (at your option) any later version.
  11  *
  12  * This program is distributed in the hope that it will be useful,
  13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  15  * GNU General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU General Public License along
  18  * with this program; if not, write to the Free Software Foundation, Inc.,
  19  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  20  * http://www.gnu.org/copyleft/gpl.html
  21  *
  22  * @since 1.25
  23  *
  24  * @file
  25  * @ingroup Site
  26  *
  27  * @license GPL-2.0-or-later
  28  * @author Daniel Kinzler
  29  */
  30 class SiteImporter {
  31
  32         /**
  33          * @var SiteStore
  34          */
  35         private $store;
  36
  37         /**
  38          * @var callable|null
  39          */
  40         private $exceptionCallback;
  41
  42         /**
  43          * @param SiteStore $store
  44          */
  45         public function __construct( SiteStore $store ) {
  46                 $this->store = $store;
  47         }
  48
  49         /**
  50          * @return callable
  51          */
  52         public function getExceptionCallback() {
  53                 return $this->exceptionCallback;
  54         }
  55
  56         /**
  57          * @param callable $exceptionCallback
  58          */
  59         public function setExceptionCallback( $exceptionCallback ) {
  60                 $this->exceptionCallback = $exceptionCallback;
  61         }
  62
  63         /**
  64          * @param string $file
  65          */
  66         public function importFromFile( $file ) {
  67                 $xml = file_get_contents( $file );
  68
  69                 if ( $xml === false ) {
  70                         throw new RuntimeException( 'Failed to read ' . $file . '!' );
  71                 }
  72
  73                 $this->importFromXML( $xml );
  74         }
  75
  76         /**
  77          * @param string $xml
  78          *
  79          * @throws InvalidArgumentException
  80          */
  81         public function importFromXML( $xml ) {
  82                 $document = new DOMDocument();
  83
  84                 $oldLibXmlErrors = libxml_use_internal_errors( true );
  85                 $ok = $document->loadXML( $xml, LIBXML_NONET );
  86
  87                 if ( !$ok ) {
  88                         $errors = libxml_get_errors();
  89                         libxml_use_internal_errors( $oldLibXmlErrors );
  90
  91                         foreach ( $errors as $error ) {
  92                                 /** @var LibXMLError $error */
  93                                 throw new InvalidArgumentException(
  94                                         'Malformed XML: ' . $error->message . ' in line ' . $error->line
  95                                 );
  96                         }
  97
  98                         throw new InvalidArgumentException( 'Malformed XML!' );
  99                 }
 100
 101                 libxml_use_internal_errors( $oldLibXmlErrors );
 102                 $this->importFromDOM( $document->documentElement );
 103         }
 104
 105         /**
 106          * @param DOMElement $root
 107          */
 108         private function importFromDOM( DOMElement $root ) {
 109                 $sites = $this->makeSiteList( $root );
 110                 $this->store->saveSites( $sites );
 111         }
 112
 113         /**
 114          * @param DOMElement $root
 115          *
 116          * @return Site[]
 117          */
 118         private function makeSiteList( DOMElement $root ) {
 119                 $sites = [];
 120
 121                 // Old sites, to get the row IDs that correspond to the global site IDs.
 122                 // TODO: Get rid of internal row IDs, they just get in the way. Get rid of ORMRow, too.
 123                 $oldSites = $this->store->getSites();
 124
 125                 $current = $root->firstChild;
 126                 while ( $current ) {
 127                         if ( $current instanceof DOMElement && $current->tagName === 'site' ) {
 128                                 try {
 129                                         $site = $this->makeSite( $current );
 130                                         $key = $site->getGlobalId();
 131
 132                                         if ( $oldSites->hasSite( $key ) ) {
 133                                                 $oldSite = $oldSites->getSite( $key );
 134                                                 $site->setInternalId( $oldSite->getInternalId() );
 135                                         }
 136
 137                                         $sites[$key] = $site;
 138                                 } catch ( Exception $ex ) {
 139                                         $this->handleException( $ex );
 140                                 }
 141                         }
 142
 143                         $current = $current->nextSibling;
 144                 }
 145
 146                 return $sites;
 147         }
 148
 149         /**
 150          * @param DOMElement $siteElement
 151          *
 152          * @return Site
 153          * @throws InvalidArgumentException
 154          */
 155         public function makeSite( DOMElement $siteElement ) {
 156                 if ( $siteElement->tagName !== 'site' ) {
 157                         throw new InvalidArgumentException( 'Expected <site> tag, found ' . $siteElement->tagName );
 158                 }
 159
 160                 $type = $this->getAttributeValue( $siteElement, 'type', Site::TYPE_UNKNOWN );
 161                 $site = Site::newForType( $type );
 162
 163                 $site->setForward( $this->hasChild( $siteElement, 'forward' ) );
 164                 $site->setGlobalId( $this->getChildText( $siteElement, 'globalid' ) );
 165                 $site->setGroup( $this->getChildText( $siteElement, 'group', Site::GROUP_NONE ) );
 166                 $site->setSource( $this->getChildText( $siteElement, 'source', Site::SOURCE_LOCAL ) );
 167
 168                 $pathTags = $siteElement->getElementsByTagName( 'path' );
 169                 for ( $i = 0; $i < $pathTags->length; $i++ ) {
 170                         $pathElement = $pathTags->item( $i );
 171                         '@phan-var DOMElement $pathElement';
 172                         $pathType = $this->getAttributeValue( $pathElement, 'type' );
 173                         $path = $pathElement->textContent;
 174
 175                         $site->setPath( $pathType, $path );
 176                 }
 177
 178                 $idTags = $siteElement->getElementsByTagName( 'localid' );
 179                 for ( $i = 0; $i < $idTags->length; $i++ ) {
 180                         $idElement = $idTags->item( $i );
 181                         '@phan-var DOMElement $idElement';
 182                         $idType = $this->getAttributeValue( $idElement, 'type' );
 183                         $id = $idElement->textContent;
 184
 185                         $site->addLocalId( $idType, $id );
 186                 }
 187
 188                 // @todo: import <data>
 189                 // @todo: import <config>
 190
 191                 return $site;
 192         }
 193
 194         /**
 195          * @param DOMElement $element
 196          * @param string $name
 197          * @param string|null|bool $default
 198          *
 199          * @return null|string
 200          * @throws MWException If the attribute is not found and no default is provided
 201          */
 202         private function getAttributeValue( DOMElement $element, $name, $default = false ) {
 203                 $node = $element->getAttributeNode( $name );
 204
 205                 if ( !$node ) {
 206                         if ( $default !== false ) {
 207                                 return $default;
 208                         } else {
 209                                 throw new MWException(
 210                                         'Required ' . $name . ' attribute not found in <' . $element->tagName . '> tag'
 211                                 );
 212                         }
 213                 }
 214
 215                 return $node->textContent;
 216         }
 217
 218         /**
 219          * @param DOMElement $element
 220          * @param string $name
 221          * @param string|null|bool $default
 222          *
 223          * @return null|string
 224          * @throws MWException If the child element is not found and no default is provided
 225          */
 226         private function getChildText( DOMElement $element, $name, $default = false ) {
 227                 $elements = $element->getElementsByTagName( $name );
 228
 229                 if ( $elements->length < 1 ) {
 230                         if ( $default !== false ) {
 231                                 return $default;
 232                         } else {
 233                                 throw new MWException(
 234                                         'Required <' . $name . '> tag not found inside <' . $element->tagName . '> tag'
 235                                 );
 236                         }
 237                 }
 238
 239                 $node = $elements->item( 0 );
 240                 return $node->textContent;
 241         }
 242
 243         /**
 244          * @param DOMElement $element
 245          * @param string $name
 246          *
 247          * @return bool
 248          * @throws MWException
 249          */
 250         private function hasChild( DOMElement $element, $name ) {
 251                 return $this->getChildText( $element, $name, null ) !== null;
 252         }
 253
 254         /**
 255          * @param Exception $ex
 256          */
 257         private function handleException( Exception $ex ) {
 258                 if ( $this->exceptionCallback ) {
 259                         call_user_func( $this->exceptionCallback, $ex );
 260                 } else {
 261                         wfLogWarning( $ex->getMessage() );
 262                 }
 263         }
 264
 265 }