/**
* Extract Metadata from a PDF file
*
* @since 2.10
*
* @param string full path to the desired file
*
* @return array ( 'xmp' => array( key => value ), 'pdf' => array( key => value ) ) for each metadata field, in string format
*/
public static function mla_extract_pdf_metadata($file_name)
{
$xmp = array();
$metadata = array();
self::$pdf_indirect_objects = NULL;
$chunksize = 16384;
if (!file_exists($file_name)) {
return array('xmp' => $xmp, 'pdf' => $metadata);
}
$filesize = filesize($file_name);
$file_offset = $chunksize < $filesize ? $filesize - $chunksize : 0;
$tail = file_get_contents($file_name, false, NULL, $file_offset);
//error_log( __LINE__ . " MLAPDF::mla_extract_pdf_metadata( {$file_name}, {$file_offset} ) tail = \r\n" . MLAData::mla_hex_dump( $tail ), 0 );
if (0 == $file_offset) {
$header = substr($tail, 0, 128);
} else {
$header = file_get_contents($file_name, false, NULL, 0, 128);
}
//error_log( __LINE__ . " MLAPDF::mla_extract_pdf_metadata( {$file_name}, {$file_offset} ) header = \r\n" . MLAData::mla_hex_dump( $header ), 0 );
if ('%PDF-' == substr($header, 0, 5)) {
$metadata['PDF_Version'] = substr($header, 1, 7);
$metadata['PDF_VersionNumber'] = substr($header, 5, 3);
}
/*
* Find the xref and (optional) trailer
*/
$match_count = preg_match_all('/startxref[\\x00-\\x20]+(\\d+)[\\x00-\\x20]+\\%\\%EOF/', $tail, $matches, PREG_OFFSET_CAPTURE);
if (0 == $match_count) {
/* translators: 1: ERROR tag 2: path and file */
error_log(sprintf(_x('%1$s: File "%2$s", startxref not found.', 'error_log', 'media-library-assistant'), __('ERROR', 'media-library-assistant'), $path), 0);
return array('xmp' => $xmp, 'pdf' => $metadata);
}
$startxref = (int) $matches[1][$match_count - 1][0];
$trailer_dictionaries = self::_extract_pdf_trailer($file_name, $startxref);
//error_log( __LINE__ . " MLAPDF::mla_extract_pdf_metadata trailer_dictionaries = " . var_export( $trailer_dictionaries, true ), 0 );
if (is_array($trailer_dictionaries)) {
$info_reference = NULL;
foreach ($trailer_dictionaries as $trailer_dictionary) {
if (isset($trailer_dictionary['Info'])) {
$info_reference = $trailer_dictionary['Info'];
break;
}
}
//error_log( __LINE__ . " MLAPDF::mla_extract_pdf_metadata info_reference = " . var_export( $info_reference, true ), 0 );
if (isset($info_reference)) {
$info_object = self::_find_pdf_indirect_dictionary($file_name, $info_reference['object'], $info_reference['generation']);
/*
* Handle single or multiple Info instances
*/
$info_objects = array();
if ($info_object) {
if (1 == $info_object['count']) {
$info_objects[] = $info_object;
} else {
for ($index = 0; $index < $info_object['count']; $index++) {
$info_objects[] = self::_find_pdf_indirect_dictionary($file_name, $info_reference['object'], $info_reference['generation'], $index);
}
}
}
//error_log( __LINE__ . " MLAPDF::mla_extract_pdf_metadata info_objects = " . var_export( $info_objects, true ), 0 );
foreach ($info_objects as $info_object) {
$info_dictionary = self::_parse_pdf_dictionary($info_object['content'], 0);
//error_log( __LINE__ . " MLAPDF::mla_extract_pdf_metadata info_dictionary = " . var_export( $info_dictionary, true ), 0 );
unset($info_dictionary['/length']);
foreach ($info_dictionary as $name => $value) {
if ('string' == $value['type']) {
$prefix = substr($value['value'], 0, 2);
if ('D:' == $prefix) {
$metadata[$name] = MLAData::mla_parse_pdf_date($value['value']);
} elseif (chr(0xfe) . chr(0xff) == $prefix) {
$metadata[$name] = self::_parse_pdf_UTF16BE($value['value']);
} else {
$metadata[$name] = $value['value'];
}
} else {
$metadata[$name] = $value['value'];
}
}
// each info entry
}
// foreach Info object
/*
* Remove spurious "Filter" dictionaries
*/
unset($metadata['Filter']);
unset($metadata['Length']);
unset($metadata['Length1']);
}
// found Info reference
//error_log( __LINE__ . ' MLAPDF::mla_extract_pdf_metadata pdf metadata = ' . var_export( $metadata, true ), 0 );
/*
* Look for XMP Metadata
*/
$root_reference = NULL;
//error_log( __LINE__ . " MLAPDF::mla_extract_pdf_metadata info_dictionary = " . var_export( $info_dictionary, true ), 0 );
foreach ($trailer_dictionaries as $trailer_dictionary) {
if (isset($trailer_dictionary['Root'])) {
$root_reference = $trailer_dictionary['Root'];
break;
}
}
//error_log( __LINE__ . " MLAPDF::mla_extract_pdf_metadata root_reference = " . var_export( $root_reference, true ), 0 );
if (isset($root_reference)) {
$root_object = self::_find_pdf_indirect_dictionary($file_name, $root_reference['object'], $root_reference['generation']);
//error_log( __LINE__ . " MLAPDF::mla_extract_pdf_metadata root_object = " . var_export( $root_object, true ), 0 );
if ($root_object) {
$root_dictionary = self::_parse_pdf_dictionary($root_object['content'], 0);
//error_log( __LINE__ . " MLAPDF::mla_extract_pdf_metadata root_dictionary = " . var_export( $root_dictionary, true ), 0 );
unset($root_dictionary['/length']);
if (isset($root_dictionary['Metadata'])) {
$xmp_object = self::_find_pdf_indirect_dictionary($file_name, $root_dictionary['Metadata']['object'], $root_dictionary['Metadata']['generation']);
//error_log( __LINE__ . " MLAPDF::mla_extract_pdf_metadata xmp_object = " . var_export( $xmp_object, true ), 0 );
$xmp = MLAData::mla_parse_xmp_metadata($file_name, $xmp_object['start'] + $xmp_object['length']);
if (is_array($xmp)) {
$metadata = array_merge($metadata, $xmp);
} else {
$xmp = array();
$xmp = MLAData::mla_parse_xmp_metadata($file_name, 0);
//error_log( __LINE__ . ' MLAPDF::mla_extract_pdf_metadata recovered xmp = ' . var_export( $xmp, true ), 0 );
}
}
// found Metadata reference
}
// found Root object
}
// found Root reference
}
// found trailer_dictionaries
//error_log( __LINE__ . ' MLAPDF::mla_extract_pdf_metadata pdf = ' . var_export( $metadata, true ), 0 );
//error_log( __LINE__ . ' MLAPDF::mla_extract_pdf_metadata xmp = ' . var_export( $xmp, true ), 0 );
return array('xmp' => $xmp, 'pdf' => $metadata);
}
/**
* Fetch and filter IPTC and EXIF, XMP or PDF metadata for an image attachment
*
* @since 0.90
*
* @param int post ID of attachment
* @param string optional; if $post_id is zero, path to the image file.
*
* @return array Meta data variables, IPTC and EXIF or PDF
*/
public static function mla_fetch_attachment_image_metadata($post_id, $path = '')
{
$results = array('post_id' => $post_id, 'mla_iptc_metadata' => array(), 'mla_exif_metadata' => array(), 'mla_xmp_metadata' => array(), 'mla_pdf_metadata' => array());
if (0 != $post_id) {
$path = get_attached_file($post_id);
}
if (!empty($path)) {
if ('pdf' == strtolower(pathinfo($path, PATHINFO_EXTENSION))) {
$pdf_metadata = MLAPDF::mla_extract_pdf_metadata($path);
$results['mla_xmp_metadata'] = $pdf_metadata['xmp'];
$results['mla_pdf_metadata'] = $pdf_metadata['pdf'];
return $results;
}
$size = getimagesize($path, $info);
if (is_callable('iptcparse')) {
if (!empty($info['APP13'])) {
//set_error_handler( 'MLAData::mla_IPTC_EXIF_error_handler' );
$iptc_values = iptcparse($info['APP13']);
//restore_error_handler();
if (!empty(MLAData::$mla_IPTC_EXIF_errors)) {
$results['mla_iptc_errors'] = MLAData::$mla_IPTC_EXIF_errors;
MLAData::$mla_IPTC_EXIF_errors = array();
error_log(__('ERROR', 'media-library-assistant') . ': ' . '$results[mla_iptc_errors] = ' . var_export($results['mla_exif_errors'], true), 0);
}
if (!is_array($iptc_values)) {
$iptc_values = array();
}
foreach ($iptc_values as $key => $value) {
if (in_array($key, array('1#000', '1#020', '1#022', '1#120', '1#122', '2#000', '2#200', '2#201'))) {
$value = unpack('nbinary', $value[0]);
$results['mla_iptc_metadata'][$key] = (string) $value['binary'];
} elseif (1 == count($value)) {
$results['mla_iptc_metadata'][$key] = $value[0];
} else {
$results['mla_iptc_metadata'][$key] = $value;
}
}
// foreach $value
}
// ! empty
}
// iptcparse
if (is_callable('exif_read_data') && in_array($size[2], array(IMAGETYPE_JPEG, IMAGETYPE_TIFF_II, IMAGETYPE_TIFF_MM))) {
//set_error_handler( 'MLAData::mla_IPTC_EXIF_error_handler' );
$results['mla_exif_metadata'] = $exif_data = @exif_read_data($path);
//restore_error_handler();
if (!empty(MLAData::$mla_IPTC_EXIF_errors)) {
$results['mla_exif_errors'] = MLAData::$mla_IPTC_EXIF_errors;
MLAData::$mla_IPTC_EXIF_errors = array();
error_log(__('ERROR', 'media-library-assistant') . ': ' . '$results[mla_exif_errors] = ' . var_export($results['mla_exif_errors'], true), 0);
}
}
// exif_read_data
$results['mla_xmp_metadata'] = self::mla_parse_xmp_metadata($path, 0);
if (NULL == $results['mla_xmp_metadata']) {
$results['mla_xmp_metadata'] = array();
}
}
/*
* Expand EXIF Camera-related values:
*
* ExposureBiasValue
* ExposureTime
* Flash
* FNumber
* FocalLength
* ShutterSpeed from ExposureTime
*/
$new_data = array();
if (isset($exif_data['FNumber'])) {
if (false !== ($value = self::_rational_to_string($exif_data['FNumber'], '%1$d', '%1$d/%2$d', '%1$.1f'))) {
$new_data['FNumber'] = $value;
}
}
// FNumber
if (isset($exif_data['ExposureBiasValue'])) {
$fragments = array_map('intval', explode('/', $exif_data['ExposureBiasValue']));
if (!is_null($fragments[1])) {
$numerator = $fragments[0];
$denominator = $fragments[1];
// Clean up some common format issues, e.g. 4/6, 2/4
while (0 == ($numerator & 0x1) && 0 == ($denominator & 0x1)) {
$numerator = $numerator >> 1;
$denominator = $denominator >> 1;
}
// Remove excess precision
if ($denominator > $numerator && 1000 < $numerator && 1000 < $denominator) {
$exif_data['ExposureBiasValue'] = sprintf('%1$+.3f', $numerator / $denominator);
} else {
$fragments[0] = $numerator;
$fragments[1] = $denominator;
$exif_data['ExposureBiasValue'] = $numerator . '/' . $denominator;
}
}
if (false !== ($value = self::_rational_to_string($exif_data['ExposureBiasValue'], '%1$+d', '%1$+d/%2$d', '%1$+.2f'))) {
$new_data['ExposureBiasValue'] = $value;
}
}
// ExposureBiasValue
if (isset($exif_data['Flash'])) {
$value = absint($exif_data['Flash']);
if ($value & 0x1) {
$new_data['Flash'] = __('Yes', 'media-library-assistant');
} else {
$new_data['Flash'] = __('No', 'media-library-assistant');
}
}
// Flash
if (isset($exif_data['FocalLength'])) {
if (false !== ($value = self::_rational_to_string($exif_data['FocalLength'], '%1$d', '%1$d/%2$d', '%1$.2f'))) {
$new_data['FocalLength'] = $value;
}
}
// FocalLength
if (isset($exif_data['ExposureTime'])) {
if (false !== ($value = self::_rational_to_string($exif_data['ExposureTime'], '%1$d', '%1$d/%2$d', '%1$.2f'))) {
$new_data['ExposureTime'] = $value;
}
}
// ExposureTime
/*
* ShutterSpeed in "1/" format, from ExposureTime
* Special logic for "fractional shutter speed" values 1.3, 1.5, 1.6, 2.5
*/
if (isset($exif_data['ExposureTime'])) {
$fragments = array_map('intval', explode('/', $exif_data['ExposureTime']));
if (!is_null($fragments[1] && $fragments[0])) {
if (1 == $fragments[1]) {
$new_data['ShutterSpeed'] = $new_data['ExposureTime'] = sprintf('%1$d', $fragments[0]);
} elseif (0 != $fragments[1]) {
$value = $fragments[0] / $fragments[1];
if (0 < $value && 1 > $value) {
// Convert to "1/" value for shutter speed
if (1 == $fragments[0]) {
$new_data['ShutterSpeed'] = $new_data['ExposureTime'];
} else {
$test = (double) number_format(1.0 / $value, 1, '.', '');
if (in_array($test, array(1.3, 1.5, 1.6, 2.5))) {
$new_data['ShutterSpeed'] = '1/' . number_format(1.0 / $value, 1, '.', '');
} else {
$new_data['ShutterSpeed'] = '1/' . number_format(1.0 / $value, 0, '.', '');
}
}
} else {
$new_data['ShutterSpeed'] = $new_data['ExposureTime'] = sprintf('%1$.2f', $value);
}
}
// fractional value
}
// valid denominator and non-zero numerator
}
// ShutterSpeed
if (isset($exif_data['UndefinedTag:0xA420'])) {
$new_data['ImageUniqueID'] = $exif_data['UndefinedTag:0xA420'];
}
if (isset($exif_data['UndefinedTag:0xA430'])) {
$new_data['CameraOwnerName'] = $exif_data['UndefinedTag:0xA430'];
}
if (isset($exif_data['UndefinedTag:0xA431'])) {
$new_data['BodySerialNumber'] = $exif_data['UndefinedTag:0xA431'];
}
if (isset($exif_data['UndefinedTag:0xA432']) && is_array($exif_data['UndefinedTag:0xA432'])) {
$array = $new_data['LensSpecification'] = $exif_data['UndefinedTag:0xA432'];
if (isset($array[0])) {
if (false !== ($value = self::_rational_to_string($array[0], '%1$d', '%1$d/%2$d', '%1$.2f'))) {
$new_data['LensMinFocalLength'] = $value;
}
}
if (isset($array[1])) {
if (false !== ($value = self::_rational_to_string($array[1], '%1$d', '%1$d/%2$d', '%1$.2f'))) {
$new_data['LensMaxFocalLength'] = $value;
}
}
if (isset($array[2])) {
if (false !== ($value = self::_rational_to_string($array[2], '%1$d', '%1$d/%2$d', '%1$.1f'))) {
$new_data['LensMinFocalLengthFN'] = $value;
}
}
if (isset($array[3])) {
if (false !== ($value = self::_rational_to_string($array[3], '%1$d', '%1$d/%2$d', '%1$.1f'))) {
$new_data['LensMaxFocalLengthFN'] = $value;
}
}
}
if (isset($exif_data['UndefinedTag:0xA433'])) {
$new_data['LensMake'] = $exif_data['UndefinedTag:0xA433'];
}
if (isset($exif_data['UndefinedTag:0xA434'])) {
$new_data['LensModel'] = $exif_data['UndefinedTag:0xA434'];
}
if (isset($exif_data['UndefinedTag:0xA435'])) {
$new_data['LensSerialNumber'] = $exif_data['UndefinedTag:0xA435'];
}
if (!empty($new_data)) {
$results['mla_exif_metadata']['CAMERA'] = $new_data;
}
/*
* Expand EXIF GPS values
*/
$new_data = array();
if (isset($exif_data['GPSVersion'])) {
$new_data['Version'] = sprintf('%1$d.%2$d.%3$d.%4$d', ord($exif_data['GPSVersion'][0]), ord($exif_data['GPSVersion'][1]), ord($exif_data['GPSVersion'][2]), ord($exif_data['GPSVersion'][3]));
}
if (isset($exif_data['GPSLatitudeRef'])) {
$new_data['LatitudeRef'] = $exif_data['GPSLatitudeRef'];
$new_data['LatitudeRefS'] = 'N' == $exif_data['GPSLatitudeRef'] ? '' : '-';
$ref = $new_data['LatitudeRef'];
$refs = $new_data['LatitudeRefS'];
} else {
$ref = '';
$refs = '';
}
if (isset($exif_data['GPSLatitude'])) {
$rational = $exif_data['GPSLatitude'];
$new_data['LatitudeD'] = $degrees = self::_rational_to_decimal($rational[0]);
$new_data['LatitudeM'] = $minutes = self::_rational_to_decimal($rational[1]);
$new_data['LatitudeS'] = sprintf('%1$01.4f', $seconds = self::_rational_to_decimal($rational[2]));
$decimal_minutes = $minutes + $seconds / 60;
$decimal_degrees = $decimal_minutes / 60;
$new_data['Latitude'] = sprintf('%1$dd %2$d\' %3$01.4f" %4$s', $degrees, $minutes, $seconds, $ref);
$new_data['LatitudeDM'] = sprintf('%1$d %2$01.4f', $degrees, $decimal_minutes);
$new_data['LatitudeDD'] = sprintf('%1$01f', $degrees + $decimal_degrees);
$new_data['LatitudeMinDec'] = substr($new_data['LatitudeDM'], strpos($new_data['LatitudeDM'], ' ') + 1);
$new_data['LatitudeDegDec'] = substr($new_data['LatitudeDD'], strpos($new_data['LatitudeDD'], '.'));
$new_data['LatitudeSDM'] = $refs . $new_data['LatitudeDM'];
$new_data['LatitudeSDD'] = $refs . $new_data['LatitudeDD'];
$new_data['LatitudeDM'] = $new_data['LatitudeDM'] . $ref;
$new_data['LatitudeDD'] = $new_data['LatitudeDD'] . $ref;
}
if (isset($exif_data['GPSLongitudeRef'])) {
$new_data['LongitudeRef'] = $exif_data['GPSLongitudeRef'];
$new_data['LongitudeRefS'] = 'E' == $exif_data['GPSLongitudeRef'] ? '' : '-';
$ref = $new_data['LongitudeRef'];
$refs = $new_data['LongitudeRefS'];
} else {
$ref = '';
$refs = '';
}
if (isset($exif_data['GPSLongitude'])) {
$rational = $exif_data['GPSLongitude'];
$new_data['LongitudeD'] = $degrees = self::_rational_to_decimal($rational[0]);
$new_data['LongitudeM'] = $minutes = self::_rational_to_decimal($rational[1]);
$new_data['LongitudeS'] = sprintf('%1$01.4f', $seconds = self::_rational_to_decimal($rational[2]));
$decimal_minutes = $minutes + $seconds / 60;
$decimal_degrees = $decimal_minutes / 60;
$new_data['Longitude'] = sprintf('%1$dd %2$d\' %3$01.4f" %4$s', $degrees, $minutes, $seconds, $ref);
$new_data['LongitudeDM'] = sprintf('%1$d %2$01.4f', $degrees, $decimal_minutes);
$new_data['LongitudeDD'] = sprintf('%1$01f', $degrees + $decimal_degrees);
$new_data['LongitudeMinDec'] = substr($new_data['LongitudeDM'], strpos($new_data['LongitudeDM'], ' ') + 1);
$new_data['LongitudeDegDec'] = substr($new_data['LongitudeDD'], strpos($new_data['LongitudeDD'], '.'));
$new_data['LongitudeSDM'] = $refs . $new_data['LongitudeDM'];
$new_data['LongitudeSDD'] = $refs . $new_data['LongitudeDD'];
$new_data['LongitudeDM'] = $new_data['LongitudeDM'] . $ref;
$new_data['LongitudeDD'] = $new_data['LongitudeDD'] . $ref;
}
if (isset($exif_data['GPSAltitudeRef'])) {
$new_data['AltitudeRef'] = sprintf('%1$d', ord($exif_data['GPSAltitudeRef'][0]));
$new_data['AltitudeRefS'] = '0' == $new_data['AltitudeRef'] ? '' : '-';
$refs = $new_data['AltitudeRefS'];
} else {
$refs = '';
}
if (isset($exif_data['GPSAltitude'])) {
$new_data['Altitude'] = sprintf('%1$s%2$01.4f', $refs, $meters = self::_rational_to_decimal($exif_data['GPSAltitude']));
$new_data['AltitudeFeet'] = sprintf('%1$s%2$01.2f', $refs, $meters * 3.280839895013);
}
if (isset($exif_data['GPSTimeStamp'])) {
$rational = $exif_data['GPSTimeStamp'];
$new_data['TimeStampH'] = sprintf('%1$02d', $hours = self::_rational_to_decimal($rational[0]));
$new_data['TimeStampM'] = sprintf('%1$02d', $minutes = self::_rational_to_decimal($rational[1]));
$new_data['TimeStampS'] = sprintf('%1$02d', $seconds = self::_rational_to_decimal($rational[2]));
$new_data['TimeStamp'] = sprintf('%1$02d:%2$02d:%3$02d', $hours, $minutes, $seconds);
}
if (isset($exif_data['GPSDateStamp'])) {
$parts = explode(':', $exif_data['GPSDateStamp']);
$new_data['DateStampY'] = $parts[0];
$new_data['DateStampM'] = $parts[1];
$new_data['DateStampD'] = $parts[2];
$new_data['DateStamp'] = $exif_data['GPSDateStamp'];
}
if (isset($exif_data['GPSMapDatum'])) {
$new_data['MapDatum'] = $exif_data['GPSMapDatum'];
}
if (!empty($new_data)) {
$results['mla_exif_metadata']['GPS'] = $new_data;
}
/*
* Expand EXIF array values - replaced by mla_find_array_element MLA v2.13
* /
foreach ( $results['mla_exif_metadata'] as $exif_key => $exif_value ) {
if ( is_array( $exif_value ) ) {
foreach ( $exif_value as $key => $value ) {
$results['mla_exif_metadata'][ $exif_key . '.' . $key ] = $value;
}
} // is_array
} // */
return $results;
}