Matthias Andreas Benkard | 7b2a3a1 | 2021-08-16 10:57:25 +0200 | [diff] [blame] | 1 | <?php |
| 2 | |
| 3 | /* |
| 4 | * This file is part of the Symfony package. |
| 5 | * |
| 6 | * (c) Fabien Potencier <fabien@symfony.com> |
| 7 | * |
| 8 | * For the full copyright and license information, please view the LICENSE |
| 9 | * file that was distributed with this source code. |
| 10 | */ |
| 11 | |
| 12 | namespace Symfony\Component\Translation\Loader; |
| 13 | |
| 14 | use Symfony\Component\Config\Resource\FileResource; |
| 15 | use Symfony\Component\Config\Util\Exception\InvalidXmlException; |
| 16 | use Symfony\Component\Config\Util\Exception\XmlParsingException; |
| 17 | use Symfony\Component\Config\Util\XmlUtils; |
| 18 | use Symfony\Component\Translation\Exception\InvalidResourceException; |
| 19 | use Symfony\Component\Translation\Exception\NotFoundResourceException; |
| 20 | use Symfony\Component\Translation\Exception\RuntimeException; |
| 21 | use Symfony\Component\Translation\MessageCatalogue; |
| 22 | use Symfony\Component\Translation\Util\XliffUtils; |
| 23 | |
| 24 | /** |
| 25 | * XliffFileLoader loads translations from XLIFF files. |
| 26 | * |
| 27 | * @author Fabien Potencier <fabien@symfony.com> |
| 28 | */ |
| 29 | class XliffFileLoader implements LoaderInterface |
| 30 | { |
| 31 | /** |
| 32 | * {@inheritdoc} |
| 33 | */ |
Matthias Andreas Benkard | 1ba5381 | 2022-12-27 17:32:58 +0100 | [diff] [blame^] | 34 | public function load(mixed $resource, string $locale, string $domain = 'messages'): MessageCatalogue |
Matthias Andreas Benkard | 7b2a3a1 | 2021-08-16 10:57:25 +0200 | [diff] [blame] | 35 | { |
| 36 | if (!class_exists(XmlUtils::class)) { |
| 37 | throw new RuntimeException('Loading translations from the Xliff format requires the Symfony Config component.'); |
| 38 | } |
| 39 | |
| 40 | if (!$this->isXmlString($resource)) { |
| 41 | if (!stream_is_local($resource)) { |
| 42 | throw new InvalidResourceException(sprintf('This is not a local file "%s".', $resource)); |
| 43 | } |
| 44 | |
| 45 | if (!file_exists($resource)) { |
| 46 | throw new NotFoundResourceException(sprintf('File "%s" not found.', $resource)); |
| 47 | } |
| 48 | |
| 49 | if (!is_file($resource)) { |
| 50 | throw new InvalidResourceException(sprintf('This is neither a file nor an XLIFF string "%s".', $resource)); |
| 51 | } |
| 52 | } |
| 53 | |
| 54 | try { |
| 55 | if ($this->isXmlString($resource)) { |
| 56 | $dom = XmlUtils::parse($resource); |
| 57 | } else { |
| 58 | $dom = XmlUtils::loadFile($resource); |
| 59 | } |
Matthias Andreas Benkard | 1ba5381 | 2022-12-27 17:32:58 +0100 | [diff] [blame^] | 60 | } catch (\InvalidArgumentException|XmlParsingException|InvalidXmlException $e) { |
Matthias Andreas Benkard | 7b2a3a1 | 2021-08-16 10:57:25 +0200 | [diff] [blame] | 61 | throw new InvalidResourceException(sprintf('Unable to load "%s": ', $resource).$e->getMessage(), $e->getCode(), $e); |
| 62 | } |
| 63 | |
| 64 | if ($errors = XliffUtils::validateSchema($dom)) { |
| 65 | throw new InvalidResourceException(sprintf('Invalid resource provided: "%s"; Errors: ', $resource).XliffUtils::getErrorsAsString($errors)); |
| 66 | } |
| 67 | |
| 68 | $catalogue = new MessageCatalogue($locale); |
| 69 | $this->extract($dom, $catalogue, $domain); |
| 70 | |
| 71 | if (is_file($resource) && class_exists(FileResource::class)) { |
| 72 | $catalogue->addResource(new FileResource($resource)); |
| 73 | } |
| 74 | |
| 75 | return $catalogue; |
| 76 | } |
| 77 | |
| 78 | private function extract(\DOMDocument $dom, MessageCatalogue $catalogue, string $domain) |
| 79 | { |
| 80 | $xliffVersion = XliffUtils::getVersionNumber($dom); |
| 81 | |
| 82 | if ('1.2' === $xliffVersion) { |
| 83 | $this->extractXliff1($dom, $catalogue, $domain); |
| 84 | } |
| 85 | |
| 86 | if ('2.0' === $xliffVersion) { |
| 87 | $this->extractXliff2($dom, $catalogue, $domain); |
| 88 | } |
| 89 | } |
| 90 | |
| 91 | /** |
| 92 | * Extract messages and metadata from DOMDocument into a MessageCatalogue. |
| 93 | */ |
| 94 | private function extractXliff1(\DOMDocument $dom, MessageCatalogue $catalogue, string $domain) |
| 95 | { |
| 96 | $xml = simplexml_import_dom($dom); |
| 97 | $encoding = $dom->encoding ? strtoupper($dom->encoding) : null; |
| 98 | |
| 99 | $namespace = 'urn:oasis:names:tc:xliff:document:1.2'; |
| 100 | $xml->registerXPathNamespace('xliff', $namespace); |
| 101 | |
| 102 | foreach ($xml->xpath('//xliff:file') as $file) { |
| 103 | $fileAttributes = $file->attributes(); |
| 104 | |
| 105 | $file->registerXPathNamespace('xliff', $namespace); |
| 106 | |
| 107 | foreach ($file->xpath('.//xliff:trans-unit') as $translation) { |
| 108 | $attributes = $translation->attributes(); |
| 109 | |
| 110 | if (!(isset($attributes['resname']) || isset($translation->source))) { |
| 111 | continue; |
| 112 | } |
| 113 | |
| 114 | $source = isset($attributes['resname']) && $attributes['resname'] ? $attributes['resname'] : $translation->source; |
| 115 | // If the xlf file has another encoding specified, try to convert it because |
| 116 | // simple_xml will always return utf-8 encoded values |
| 117 | $target = $this->utf8ToCharset((string) ($translation->target ?? $translation->source), $encoding); |
| 118 | |
| 119 | $catalogue->set((string) $source, $target, $domain); |
| 120 | |
| 121 | $metadata = [ |
| 122 | 'source' => (string) $translation->source, |
| 123 | 'file' => [ |
| 124 | 'original' => (string) $fileAttributes['original'], |
| 125 | ], |
| 126 | ]; |
| 127 | if ($notes = $this->parseNotesMetadata($translation->note, $encoding)) { |
| 128 | $metadata['notes'] = $notes; |
| 129 | } |
| 130 | |
| 131 | if (isset($translation->target) && $translation->target->attributes()) { |
| 132 | $metadata['target-attributes'] = []; |
| 133 | foreach ($translation->target->attributes() as $key => $value) { |
| 134 | $metadata['target-attributes'][$key] = (string) $value; |
| 135 | } |
| 136 | } |
| 137 | |
| 138 | if (isset($attributes['id'])) { |
| 139 | $metadata['id'] = (string) $attributes['id']; |
| 140 | } |
| 141 | |
| 142 | $catalogue->setMetadata((string) $source, $metadata, $domain); |
| 143 | } |
| 144 | } |
| 145 | } |
| 146 | |
| 147 | private function extractXliff2(\DOMDocument $dom, MessageCatalogue $catalogue, string $domain) |
| 148 | { |
| 149 | $xml = simplexml_import_dom($dom); |
| 150 | $encoding = $dom->encoding ? strtoupper($dom->encoding) : null; |
| 151 | |
| 152 | $xml->registerXPathNamespace('xliff', 'urn:oasis:names:tc:xliff:document:2.0'); |
| 153 | |
| 154 | foreach ($xml->xpath('//xliff:unit') as $unit) { |
| 155 | foreach ($unit->segment as $segment) { |
| 156 | $attributes = $unit->attributes(); |
| 157 | $source = $attributes['name'] ?? $segment->source; |
| 158 | |
| 159 | // If the xlf file has another encoding specified, try to convert it because |
| 160 | // simple_xml will always return utf-8 encoded values |
| 161 | $target = $this->utf8ToCharset((string) ($segment->target ?? $segment->source), $encoding); |
| 162 | |
| 163 | $catalogue->set((string) $source, $target, $domain); |
| 164 | |
| 165 | $metadata = []; |
| 166 | if (isset($segment->target) && $segment->target->attributes()) { |
| 167 | $metadata['target-attributes'] = []; |
| 168 | foreach ($segment->target->attributes() as $key => $value) { |
| 169 | $metadata['target-attributes'][$key] = (string) $value; |
| 170 | } |
| 171 | } |
| 172 | |
| 173 | if (isset($unit->notes)) { |
| 174 | $metadata['notes'] = []; |
| 175 | foreach ($unit->notes->note as $noteNode) { |
| 176 | $note = []; |
| 177 | foreach ($noteNode->attributes() as $key => $value) { |
| 178 | $note[$key] = (string) $value; |
| 179 | } |
| 180 | $note['content'] = (string) $noteNode; |
| 181 | $metadata['notes'][] = $note; |
| 182 | } |
| 183 | } |
| 184 | |
| 185 | $catalogue->setMetadata((string) $source, $metadata, $domain); |
| 186 | } |
| 187 | } |
| 188 | } |
| 189 | |
| 190 | /** |
| 191 | * Convert a UTF8 string to the specified encoding. |
| 192 | */ |
| 193 | private function utf8ToCharset(string $content, string $encoding = null): string |
| 194 | { |
| 195 | if ('UTF-8' !== $encoding && !empty($encoding)) { |
| 196 | return mb_convert_encoding($content, $encoding, 'UTF-8'); |
| 197 | } |
| 198 | |
| 199 | return $content; |
| 200 | } |
| 201 | |
| 202 | private function parseNotesMetadata(\SimpleXMLElement $noteElement = null, string $encoding = null): array |
| 203 | { |
| 204 | $notes = []; |
| 205 | |
| 206 | if (null === $noteElement) { |
| 207 | return $notes; |
| 208 | } |
| 209 | |
| 210 | /** @var \SimpleXMLElement $xmlNote */ |
| 211 | foreach ($noteElement as $xmlNote) { |
| 212 | $noteAttributes = $xmlNote->attributes(); |
| 213 | $note = ['content' => $this->utf8ToCharset((string) $xmlNote, $encoding)]; |
| 214 | if (isset($noteAttributes['priority'])) { |
| 215 | $note['priority'] = (int) $noteAttributes['priority']; |
| 216 | } |
| 217 | |
| 218 | if (isset($noteAttributes['from'])) { |
| 219 | $note['from'] = (string) $noteAttributes['from']; |
| 220 | } |
| 221 | |
| 222 | $notes[] = $note; |
| 223 | } |
| 224 | |
| 225 | return $notes; |
| 226 | } |
| 227 | |
| 228 | private function isXmlString(string $resource): bool |
| 229 | { |
| 230 | return 0 === strpos($resource, '<?xml'); |
| 231 | } |
| 232 | } |