Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
85.71% |
72 / 84 |
|
36.36% |
4 / 11 |
CRAP | |
0.00% |
0 / 1 |
| SiteImporter | |
86.75% |
72 / 83 |
|
36.36% |
4 / 11 |
29.82 | |
0.00% |
0 / 1 |
| __construct | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| getExceptionCallback | |
0.00% |
0 / 1 |
|
0.00% |
0 / 1 |
2 | |||
| setExceptionCallback | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| importFromFile | |
75.00% |
3 / 4 |
|
0.00% |
0 / 1 |
2.06 | |||
| importFromXML | |
94.12% |
16 / 17 |
|
0.00% |
0 / 1 |
3.00 | |||
| makeSiteList | |
82.35% |
14 / 17 |
|
0.00% |
0 / 1 |
7.27 | |||
| makeSite | |
95.24% |
20 / 21 |
|
0.00% |
0 / 1 |
4 | |||
| getAttributeValue | |
62.50% |
5 / 8 |
|
0.00% |
0 / 1 |
3.47 | |||
| getChildText | |
100.00% |
9 / 9 |
|
100.00% |
1 / 1 |
3 | |||
| hasChild | |
100.00% |
1 / 1 |
|
100.00% |
1 / 1 |
1 | |||
| handleException | |
66.67% |
2 / 3 |
|
0.00% |
0 / 1 |
2.15 | |||
| 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 | /** |
| 52 | * @param SiteStore $store |
| 53 | */ |
| 54 | public function __construct( SiteStore $store ) { |
| 55 | $this->store = $store; |
| 56 | } |
| 57 | |
| 58 | /** |
| 59 | * @return callable |
| 60 | */ |
| 61 | public function getExceptionCallback() { |
| 62 | return $this->exceptionCallback; |
| 63 | } |
| 64 | |
| 65 | /** |
| 66 | * @param callable $exceptionCallback |
| 67 | */ |
| 68 | public function setExceptionCallback( $exceptionCallback ) { |
| 69 | $this->exceptionCallback = $exceptionCallback; |
| 70 | } |
| 71 | |
| 72 | /** |
| 73 | * @param string $file |
| 74 | */ |
| 75 | public function importFromFile( $file ) { |
| 76 | $xml = file_get_contents( $file ); |
| 77 | |
| 78 | if ( $xml === false ) { |
| 79 | throw new RuntimeException( 'Failed to read ' . $file . '!' ); |
| 80 | } |
| 81 | |
| 82 | $this->importFromXML( $xml ); |
| 83 | } |
| 84 | |
| 85 | /** |
| 86 | * @param string $xml |
| 87 | * |
| 88 | * @throws InvalidArgumentException |
| 89 | */ |
| 90 | public function importFromXML( $xml ) { |
| 91 | $document = new DOMDocument(); |
| 92 | |
| 93 | $oldLibXmlErrors = libxml_use_internal_errors( true ); |
| 94 | // phpcs:ignore Generic.PHP.NoSilencedErrors -- suppress deprecation per T268847 |
| 95 | $oldDisable = @libxml_disable_entity_loader( true ); |
| 96 | $ok = $document->loadXML( $xml, LIBXML_NONET ); |
| 97 | |
| 98 | if ( !$ok ) { |
| 99 | $errors = libxml_get_errors(); |
| 100 | libxml_use_internal_errors( $oldLibXmlErrors ); |
| 101 | // phpcs:ignore Generic.PHP.NoSilencedErrors |
| 102 | @libxml_disable_entity_loader( $oldDisable ); |
| 103 | |
| 104 | foreach ( $errors as $error ) { |
| 105 | /** @var LibXMLError $error */ |
| 106 | throw new InvalidArgumentException( |
| 107 | 'Malformed XML: ' . $error->message . ' in line ' . $error->line |
| 108 | ); |
| 109 | } |
| 110 | |
| 111 | throw new InvalidArgumentException( 'Malformed XML!' ); |
| 112 | } |
| 113 | |
| 114 | libxml_use_internal_errors( $oldLibXmlErrors ); |
| 115 | // phpcs:ignore Generic.PHP.NoSilencedErrors |
| 116 | @libxml_disable_entity_loader( $oldDisable ); |
| 117 | $sites = $this->makeSiteList( $document->documentElement ); |
| 118 | $this->store->saveSites( $sites ); |
| 119 | } |
| 120 | |
| 121 | /** |
| 122 | * @param DOMElement $root |
| 123 | * |
| 124 | * @return Site[] |
| 125 | */ |
| 126 | private function makeSiteList( DOMElement $root ) { |
| 127 | $sites = []; |
| 128 | |
| 129 | // Old sites, to get the row IDs that correspond to the global site IDs. |
| 130 | // TODO: Get rid of internal row IDs, they just get in the way. Get rid of ORMRow, too. |
| 131 | $oldSites = $this->store->getSites(); |
| 132 | |
| 133 | $current = $root->firstChild; |
| 134 | while ( $current ) { |
| 135 | if ( $current instanceof DOMElement && $current->tagName === 'site' ) { |
| 136 | try { |
| 137 | $site = $this->makeSite( $current ); |
| 138 | $key = $site->getGlobalId(); |
| 139 | |
| 140 | if ( $oldSites->hasSite( $key ) ) { |
| 141 | $oldSite = $oldSites->getSite( $key ); |
| 142 | $site->setInternalId( $oldSite->getInternalId() ); |
| 143 | } |
| 144 | |
| 145 | $sites[$key] = $site; |
| 146 | } catch ( TimeoutException $e ) { |
| 147 | throw $e; |
| 148 | } catch ( Exception $ex ) { |
| 149 | $this->handleException( $ex ); |
| 150 | } |
| 151 | } |
| 152 | |
| 153 | $current = $current->nextSibling; |
| 154 | } |
| 155 | |
| 156 | return $sites; |
| 157 | } |
| 158 | |
| 159 | /** |
| 160 | * @param DOMElement $siteElement |
| 161 | * |
| 162 | * @return Site |
| 163 | */ |
| 164 | public function makeSite( DOMElement $siteElement ) { |
| 165 | if ( $siteElement->tagName !== 'site' ) { |
| 166 | throw new InvalidArgumentException( 'Expected <site> tag, found ' . $siteElement->tagName ); |
| 167 | } |
| 168 | |
| 169 | $type = $this->getAttributeValue( $siteElement, 'type', Site::TYPE_UNKNOWN ); |
| 170 | $site = Site::newForType( $type ); |
| 171 | |
| 172 | $site->setForward( $this->hasChild( $siteElement, 'forward' ) ); |
| 173 | $site->setGlobalId( $this->getChildText( $siteElement, 'globalid' ) ); |
| 174 | $site->setGroup( $this->getChildText( $siteElement, 'group', Site::GROUP_NONE ) ); |
| 175 | $site->setSource( $this->getChildText( $siteElement, 'source', Site::SOURCE_LOCAL ) ); |
| 176 | |
| 177 | $pathTags = $siteElement->getElementsByTagName( 'path' ); |
| 178 | for ( $i = 0; $i < $pathTags->length; $i++ ) { |
| 179 | $pathElement = $pathTags->item( $i ); |
| 180 | '@phan-var DOMElement $pathElement'; |
| 181 | $pathType = $this->getAttributeValue( $pathElement, 'type' ); |
| 182 | $path = $pathElement->textContent; |
| 183 | |
| 184 | $site->setPath( $pathType, $path ); |
| 185 | } |
| 186 | |
| 187 | $idTags = $siteElement->getElementsByTagName( 'localid' ); |
| 188 | for ( $i = 0; $i < $idTags->length; $i++ ) { |
| 189 | $idElement = $idTags->item( $i ); |
| 190 | '@phan-var DOMElement $idElement'; |
| 191 | $idType = $this->getAttributeValue( $idElement, 'type' ); |
| 192 | $id = $idElement->textContent; |
| 193 | |
| 194 | $site->addLocalId( $idType, $id ); |
| 195 | } |
| 196 | |
| 197 | // @todo: import <data> |
| 198 | // @todo: import <config> |
| 199 | |
| 200 | return $site; |
| 201 | } |
| 202 | |
| 203 | /** |
| 204 | * @param DOMElement $element |
| 205 | * @param string $name |
| 206 | * @param string|null|false $default |
| 207 | * |
| 208 | * @return null|string |
| 209 | */ |
| 210 | private function getAttributeValue( DOMElement $element, $name, $default = false ) { |
| 211 | $node = $element->getAttributeNode( $name ); |
| 212 | |
| 213 | if ( !$node ) { |
| 214 | if ( $default !== false ) { |
| 215 | return $default; |
| 216 | } else { |
| 217 | throw new RuntimeException( |
| 218 | 'Required ' . $name . ' attribute not found in <' . $element->tagName . '> tag' |
| 219 | ); |
| 220 | } |
| 221 | } |
| 222 | |
| 223 | return $node->textContent; |
| 224 | } |
| 225 | |
| 226 | /** |
| 227 | * @param DOMElement $element |
| 228 | * @param string $name |
| 229 | * @param string|null|false $default |
| 230 | * |
| 231 | * @return null|string |
| 232 | */ |
| 233 | private function getChildText( DOMElement $element, $name, $default = false ) { |
| 234 | $elements = $element->getElementsByTagName( $name ); |
| 235 | |
| 236 | if ( $elements->length < 1 ) { |
| 237 | if ( $default !== false ) { |
| 238 | return $default; |
| 239 | } else { |
| 240 | throw new RuntimeException( |
| 241 | 'Required <' . $name . '> tag not found inside <' . $element->tagName . '> tag' |
| 242 | ); |
| 243 | } |
| 244 | } |
| 245 | |
| 246 | $node = $elements->item( 0 ); |
| 247 | return $node->textContent; |
| 248 | } |
| 249 | |
| 250 | /** |
| 251 | * @param DOMElement $element |
| 252 | * @param string $name |
| 253 | * |
| 254 | * @return bool |
| 255 | */ |
| 256 | private function hasChild( DOMElement $element, $name ) { |
| 257 | return $this->getChildText( $element, $name, null ) !== null; |
| 258 | } |
| 259 | |
| 260 | /** |
| 261 | * @param Exception $ex |
| 262 | */ |
| 263 | private function handleException( Exception $ex ) { |
| 264 | if ( $this->exceptionCallback ) { |
| 265 | call_user_func( $this->exceptionCallback, $ex ); |
| 266 | } else { |
| 267 | wfLogWarning( $ex->getMessage() ); |
| 268 | } |
| 269 | } |
| 270 | |
| 271 | } |
| 272 | |
| 273 | /** @deprecated class alias since 1.41 */ |
| 274 | class_alias( SiteImporter::class, 'SiteImporter' ); |