… and avoid assertEmpty() on arrays, in favor of a much more strict assertSame( [] ). Change-Id: I20266b0b1fc38a3a87666ba1b0793cb2b37d94a9
134 lines
5 KiB
PHP
134 lines
5 KiB
PHP
<?php
|
|
/**
|
|
* @todo Could use a test of extended XMP segments. Hard to find programs that
|
|
* create example files, and creating my own in vim probably wouldn't
|
|
* serve as a very good "test". (Adobe photoshop probably creates such files
|
|
* but it costs money). The implementation of it currently in MediaWiki is based
|
|
* solely on reading the standard, without any real world test files.
|
|
*
|
|
* @group Media
|
|
* @covers JpegMetadataExtractor
|
|
*/
|
|
class JpegMetadataExtractorTest extends MediaWikiIntegrationTestCase {
|
|
|
|
protected $filePath;
|
|
|
|
protected function setUp() : void {
|
|
parent::setUp();
|
|
|
|
$this->filePath = __DIR__ . '/../../data/media/';
|
|
}
|
|
|
|
/**
|
|
* We also use this test to test padding bytes don't
|
|
* screw stuff up
|
|
*
|
|
* @param string $file Filename
|
|
*
|
|
* @dataProvider provideUtf8Comment
|
|
*/
|
|
public function testUtf8Comment( $file ) {
|
|
$res = JpegMetadataExtractor::segmentSplitter( $this->filePath . $file );
|
|
$this->assertEquals( [ 'UTF-8 JPEG Comment — ¼' ], $res['COM'] );
|
|
}
|
|
|
|
public static function provideUtf8Comment() {
|
|
return [
|
|
[ 'jpeg-comment-utf.jpg' ],
|
|
[ 'jpeg-padding-even.jpg' ],
|
|
[ 'jpeg-padding-odd.jpg' ],
|
|
];
|
|
}
|
|
|
|
/** The file is iso-8859-1, but it should get auto converted */
|
|
public function testIso88591Comment() {
|
|
$res = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'jpeg-comment-iso8859-1.jpg' );
|
|
$this->assertEquals( [ 'ISO-8859-1 JPEG Comment - ¼' ], $res['COM'] );
|
|
}
|
|
|
|
/** Comment values that are non-textual (random binary junk) should not be shown.
|
|
* The example test file has a comment with a 0x5 byte in it which is a control character
|
|
* and considered binary junk for our purposes.
|
|
*/
|
|
public function testBinaryCommentStripped() {
|
|
$res = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'jpeg-comment-binary.jpg' );
|
|
$this->assertSame( [], $res['COM'] );
|
|
}
|
|
|
|
/* Very rarely a file can have multiple comments.
|
|
* Order of comments is based on order inside the file.
|
|
*/
|
|
public function testMultipleComment() {
|
|
$res = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'jpeg-comment-multiple.jpg' );
|
|
$this->assertEquals( [ 'foo', 'bar' ], $res['COM'] );
|
|
}
|
|
|
|
public function testXMPExtraction() {
|
|
$res = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'jpeg-xmp-psir.jpg' );
|
|
$expected = file_get_contents( $this->filePath . 'jpeg-xmp-psir.xmp' );
|
|
$this->assertEquals( $expected, $res['XMP'] );
|
|
}
|
|
|
|
public function testPSIRExtraction() {
|
|
$res = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'jpeg-xmp-psir.jpg' );
|
|
$expected = '50686f746f73686f7020332e30003842494d04040000000'
|
|
. '000181c02190004746573741c02190003666f6f1c020000020004';
|
|
$this->assertEquals( $expected, bin2hex( $res['PSIR'][0] ) );
|
|
}
|
|
|
|
public function testXMPExtractionNullChar() {
|
|
$res = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'jpeg-xmp-nullchar.jpg' );
|
|
$expected = file_get_contents( $this->filePath . 'jpeg-xmp-psir.xmp' );
|
|
$this->assertEquals( $expected, $res['XMP'] );
|
|
}
|
|
|
|
public function testXMPExtractionAltAppId() {
|
|
$res = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'jpeg-xmp-alt.jpg' );
|
|
$expected = file_get_contents( $this->filePath . 'jpeg-xmp-psir.xmp' );
|
|
$this->assertEquals( $expected, $res['XMP'] );
|
|
}
|
|
|
|
public function testIPTCHashComparisionNoHash() {
|
|
$segments = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'jpeg-xmp-psir.jpg' );
|
|
$res = JpegMetadataExtractor::doPSIR( $segments['PSIR'][0] );
|
|
|
|
$this->assertEquals( 'iptc-no-hash', $res );
|
|
}
|
|
|
|
public function testIPTCHashComparisionBadHash() {
|
|
$segments = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'jpeg-iptc-bad-hash.jpg' );
|
|
$res = JpegMetadataExtractor::doPSIR( $segments['PSIR'][0] );
|
|
|
|
$this->assertEquals( 'iptc-bad-hash', $res );
|
|
}
|
|
|
|
public function testIPTCHashComparisionGoodHash() {
|
|
$segments = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'jpeg-iptc-good-hash.jpg' );
|
|
$res = JpegMetadataExtractor::doPSIR( $segments['PSIR'][0] );
|
|
|
|
$this->assertEquals( 'iptc-good-hash', $res );
|
|
}
|
|
|
|
public function testExifByteOrder() {
|
|
$res = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'exif-user-comment.jpg' );
|
|
$expected = 'BE';
|
|
$this->assertEquals( $expected, $res['byteOrder'] );
|
|
}
|
|
|
|
public function testInfiniteRead() {
|
|
// test file truncated right after a segment, which previously
|
|
// caused an infinite loop looking for the next segment byte.
|
|
// Should get past infinite loop and throw in wfUnpack()
|
|
$this->expectException( MWException::class );
|
|
$res = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'jpeg-segment-loop1.jpg' );
|
|
}
|
|
|
|
public function testInfiniteRead2() {
|
|
// test file truncated after a segment's marker and size, which
|
|
// would cause a seek past end of file. Seek past end of file
|
|
// doesn't actually fail, but prevents further reading and was
|
|
// devolving into the previous case (testInfiniteRead).
|
|
$this->expectException( MWException::class );
|
|
$res = JpegMetadataExtractor::segmentSplitter( $this->filePath . 'jpeg-segment-loop2.jpg' );
|
|
}
|
|
}
|