Dev 240122 (#78)

.github/workflows/tests.yml

@@ -51,7 +51,7 @@ jobs:
         php bin/fileeye-mediaprobe --version
         # Remove the comment to the lines below during development, to let
         # re-compile the specification PHP file from the YAML files.
-        # php bin/fileeye-mediaprobe compile
+        php bin/fileeye-mediaprobe compile
 
     - name: Image file dumps
       continue-on-error: true

specs/Jpeg/Jpeg.yaml

@@ -1,6 +1,8 @@
 collection: Jpeg\Jpeg
 title: JPEG image
 class: FileEye\MediaProbe\Block\Jpeg
+parser: FileEye\MediaProbe\Parser\Jpeg\Jpeg
+writer: FileEye\MediaProbe\Writer\Jpeg\Jpeg
 DOMNode: jpeg
 items:
     0xC0:

specs/Media.yaml

@@ -1,4 +1,3 @@
 collection: Media
-class: FileEye\MediaProbe\Media
 DOMNode: media
 items: {}

specs/Tiff/Tiff.yaml

@@ -1,6 +1,8 @@
 collection: Tiff\Tiff
 title: 'TIFF image data'
 class: FileEye\MediaProbe\Block\Tiff
+parser: FileEye\MediaProbe\Parser\Tiff\Tiff
+writer: FileEye\MediaProbe\Writer\Tiff\Tiff
 DOMNode: tiff
 items:
     0:

src/Block/Exif/Exif.php

@@ -34,8 +34,13 @@ class Exif extends BlockBase
     protected function doParseData(DataElement $data): void
     {
         assert($this->debugInfo(['dataElement' => $data]));
-        if (Tiff::getTiffSegmentByteOrder($data, strlen(self::EXIF_HEADER)) !== null) {
-            $tiff = new ItemDefinition(CollectionFactory::get('Tiff\Tiff'));
+
+        $tiff = new ItemDefinition(
+            collection: CollectionFactory::get('Tiff\Tiff'),
+        );
+        $tiffParser = $tiff->collection->getPropertyValue('parser');
+
+        if ($tiffParser::getTiffSegmentByteOrder($data, strlen(self::EXIF_HEADER)) !== null) {
             $this->addBlock($tiff)->parseData($data, strlen(self::EXIF_HEADER), $data->getSize() - strlen(self::EXIF_HEADER));
         } else {
             // We store the data as normal JPEG content if it could not be

src/Block/Jpeg.php

@@ -2,15 +2,15 @@
 
 namespace FileEye\MediaProbe\Block;
 
-use FileEye\MediaProbe\Model\BlockBase;
 use FileEye\MediaProbe\Collection\CollectionFactory;
 use FileEye\MediaProbe\Data\DataElement;
 use FileEye\MediaProbe\Data\DataException;
+use FileEye\MediaProbe\Data\DataFormat;
 use FileEye\MediaProbe\Data\DataWindow;
 use FileEye\MediaProbe\Entry\Core\Undefined;
 use FileEye\MediaProbe\ItemDefinition;
-use FileEye\MediaProbe\Data\DataFormat;
 use FileEye\MediaProbe\MediaProbe;
+use FileEye\MediaProbe\Model\BlockBase;
 use FileEye\MediaProbe\Utility\ConvertBytes;
 
 /**
@@ -27,154 +27,4 @@ class Jpeg extends BlockBase
      * JPEG header.
      */
     const JPEG_HEADER = "\xFF\xD8\xFF";
-
-    /**
-     * {@inheritdoc}
-     */
-    protected function doParseData(DataElement $data): void
-    {
-        assert($this->debugInfo(['dataElement' => $data]));
-
-        // JPEG data is stored in big-endian format.
-        $data->setByteOrder(ConvertBytes::BIG_ENDIAN);
-
-        // Run through the data to parse the segments in the image. After each
-        // segment is parsed, the offset will be moved forward, and after the
-        // last segment we will terminate.
-        $offset = 0;
-        while ($offset < $data->getSize()) {
-            // Get the next JPEG segment id offset.
-            try {
-                $new_offset = $this->getJpegSegmentIdOffset($data, $offset);
-                $segment_id = $segment_id ?? 0;
-                if ($new_offset !== $offset) {
-                    // Add any trailing data from previous segment in a
-                    // RawData block.
-                    $this->error('Unexpected data found at end of JPEG segment {id}/{hexid} @ offset {offset}, size {size}', [
-                        'id' => $segment_id,
-                        'hexid' => '0x' . strtoupper(dechex($segment_id)),
-                        'offset' => $data->getAbsoluteOffset($offset),
-                        'size' => $new_offset - $offset,
-                    ]);
-                    $trail = new ItemDefinition(
-                        CollectionFactory::get('RawData', ['name' => 'trail']),
-                        DataFormat::BYTE,
-                        $offset
-                    );
-                    $this->addBlock($trail)->parseData($data, $offset, $new_offset - $offset);
-                }
-                $offset = $new_offset;
-            } catch (DataException $e) {
-                $this->error($e->getMessage());
-                return;
-            }
-
-            // Get the JPEG segment id.
-            $segment_id = $data->getByte($offset + 1);
-
-            // Warn if an unidentified segment is detected.
-            if (!in_array($segment_id, $this->getCollection()->listItemIds())) {
-                $this->warning('Invalid JPEG marker {id}/{hexid} found @ offset {offset}', [
-                    'id' => $segment_id,
-                    'hexid' => '0x' . strtoupper(dechex($segment_id)),
-                    'offset' => $data->getAbsoluteOffset($offset),
-                ]);
-            }
-
-            // Get the JPEG segment size.
-            $segment_collection = $this->getCollection()->getItemCollection($segment_id);
-            switch ($segment_collection->getPropertyValue('payload')) {
-                case 'none':
-                    // The data window size is the JPEG delimiter byte and the
-                    // segment identifier byte.
-                    $segment_size = 2;
-                    break;
-                case 'variable':
-                    // Read the length of the segment. The data window size
-                    // includes the JPEG delimiter byte, the segment identifier
-                    // byte and two bytes used to store the segment length.
-                    $segment_size = $data->getShort($offset + 2) + 2;
-                    break;
-                case 'fixed':
-                    // The data window size includes the JPEG delimiter byte
-                    // and the segment identifier byte.
-                    $segment_size = $segment_collection->getPropertyValue('components') + 2;
-                    break;
-                case 'scan':
-                    // In case of image scan segment, the window is to the end
-                    // of the data.
-                    $segment_size = null;
-                    break;
-            }
-
-            // Parse the MediaProbe JPEG segment data.
-            $segment_definition = new ItemDefinition($segment_collection);
-            $segment = $this->addBlock($segment_definition);
-            $segment->parseData($data, $offset, $segment_size);
-
-            // Position to end of the segment.
-            $offset += $segment->getSize();
-        }
-
-        // Fail if SOS is missing.
-        if (!$this->getElement("jpegSegment[@name='SOS']")) {
-            $this->error('Missing SOS (Start Of Scan) JPEG marker');
-        }
-
-        // Fail if EOI is missing.
-        if (!$this->getElement("jpegSegment[@name='EOI']")) {
-            $this->error('Missing EOI (End Of Image) JPEG marker');
-        }
-    }
-
-    /**
-     * Determines the offset where the next JPEG segment id is found.
-     *
-     * JPEG sections start with 0xFF. The first byte that is not 0xFF is a
-     * marker (hopefully).
-     *
-     * @param DataElement $data_element
-     *   The data element to be checked.
-     * @param int $offset
-     *   The starting offset in the data element.
-     *
-     * @return int
-     *   The found offset.
-     *
-     * @throws DataException
-     *   In case of marker not found.
-     */
-    protected function getJpegSegmentIdOffset(DataElement $data_element, int $offset): int
-    {
-        for ($i = $offset; $i < $offset + 128; $i++) {
-            if ($data_element->getByte($i) === Jpeg::JPEG_DELIMITER && $data_element->getByte($i + 1) !== Jpeg::JPEG_DELIMITER) {
-                return $i;
-            }
-        }
-        throw new DataException('JPEG marker not found @%d', $data_element->getAbsoluteOffset($offset));
-    }
-
-    /**
-     * Returns the MIME type of the image.
-     *
-     * @returns string
-     */
-    public function getMimeType(): string
-    {
-        return 'image/jpeg';
-    }
-
-    /**
-     * Determines if the data is a JPEG image.
-     *
-     * @param DataElement $data_element
-     *   The data element to be checked.
-     *
-     * @return bool
-     *   TRUE if the data element is a JPEG image.
-     */
-    public static function isDataMatchingFormat(DataElement $data_element): bool
-    {
-        return $data_element->getBytes(0, 3) === static::JPEG_HEADER;
-    }
 }

src/Block/Tiff.php

@@ -27,21 +27,9 @@ class Tiff extends BlockBase
 
     /**
      * The byte order of this TIFF segment.
-     *
-     * @var int
      */
     protected int $byteOrder;
 
-    /**
-     * Returns the MIME type of the image.
-     *
-     * @returns string
-     */
-    public function getMimeType(): string
-    {
-        return 'image/tiff';
-    }
-
     public function setByteOrder(int $byteOrder): self
     {
         $this->byteOrder = $byteOrder;
@@ -53,84 +41,6 @@ public function getByteOrder(): int
         return $this->byteOrder;
     }
 
-    /**
-     * {@inheritdoc}
-     */
-    protected function doParseData(DataElement $data): void
-    {
-        // Determine the byte order of the TIFF data.
-        $this->setByteOrder(self::getTiffSegmentByteOrder($data));
-        $data->setByteOrder($this->getByteOrder());
-
-        assert($this->debugInfo(['dataElement' => $data]));
-
-        // Starting IFD will be at offset 4 (2 bytes for byte order + 2 for header).
-        $ifd_offset = $data->getLong(4);
-
-        // If the offset to first IFD is higher than 8, then there may be an
-        // image scan (TIFF) in between. Store that in a RawData block.
-        if ($ifd_offset > 8) {
-            $scan = new ItemDefinition(
-                CollectionFactory::get('RawData', ['name' => 'scan']),
-                DataFormat::BYTE,
-                $ifd_offset - 8
-            );
-            $this->addBlock($scan)->parseData($data, 8, $ifd_offset - 8);
-        }
-
-        // Loops through IFDs. In fact we should only have IFD0 and IFD1.
-        for ($i = 0; $i <= 1; $i++) {
-            // Check data is accessible, warn otherwise.
-            if ($ifd_offset >= $data->getSize() || $ifd_offset + 4 > $data->getSize()) {
-                $this->warning(
-                    'Could not determine number of entries for {item}, overflow',
-                    ['item' => $this->getCollection()->getItemCollection($i)->getPropertyValue('name')]
-                );
-                continue;
-            }
-
-            // Find number of tags in IFD and warn if not enough data to read them.
-            $ifd_tags_count = $data->getShort($ifd_offset);
-            if ($ifd_offset + $ifd_tags_count * 4 > $data->getSize()) {
-                $this->warning(
-                    'Invalid data for {item}',
-                    ['item' => $this->getCollection()->getItemCollection($i)->getPropertyValue('name')]
-                );
-                continue;
-            }
-
-            // Create and load the IFDs. Note that the data element cannot
-            // be split in windows since any pointer will refer to the
-            // entire segment space.
-            $ifd_class = $this->getCollection()->getItemCollection($i)->getPropertyValue('class');
-            $ifd_item = new ItemDefinition($this->getCollection()->getItemCollection($i), DataFormat::LONG, $ifd_tags_count, $ifd_offset, 0, $i);
-            $ifd = new $ifd_class($ifd_item, $this);
-            try {
-                $ifd->parseData($data);
-            } catch (DataException $e) {
-                $this->error('Error processing {ifd_name}: {msg}.', [
-                    'ifd_name' => $this->getCollection()->getItemCollection($i)->getPropertyValue('name'),
-                    'msg' => $e->getMessage(),
-                ]);
-                continue;
-            }
-
-            // Offset to next IFD.
-            $ifd_offset = $data->getLong($ifd_offset + $ifd_tags_count * 12 + 2);
-
-            // If next IFD offset is 0 we are finished.
-            if ($ifd_offset === 0) {
-                break;
-            }
-
-            // IFD1 shouldn't link further.
-            if ($i === 1) {
-                $this->error('IFD1 should not link to another IFD');
-                break;
-            }
-        }
-    }
-
     /**
      * {@inheritdoc}
      */
@@ -181,54 +91,6 @@ public function toBytes($order = ConvertBytes::LITTLE_ENDIAN, $offset = 0): stri
         return $bytes;
     }
 
-    /**
-     * Determines if the data is a TIFF image.
-     *
-     * @param DataElement $data_element
-     *   The data element to be checked.
-     *
-     * @return bool
-     */
-    public static function isDataMatchingFormat(DataElement $data_element): bool
-    {
-        return static::getTiffSegmentByteOrder($data_element) !== null;
-    }
-
-    /**
-     * Returns the byte order of a TIFF segment.
-     *
-     * @return int|null
-     *   The byte order of the TIFF segment in case data is a TIFF block, null
-     *   otherwise.
-     */
-    public static function getTiffSegmentByteOrder(DataElement $data_element, int $offset = 0): ?int
-    {
-        // There must be at least 8 bytes available: 2 bytes for the byte
-        // order, 2 bytes for the TIFF header, and 4 bytes for the offset to
-        // the first IFD.
-        if ($data_element->getSize() - $offset < 8) {
-            return null;
-        }
-
-        // Byte order.
-        $order_string = $data_element->getBytes($offset, 2);
-        if ($order_string === 'II') {
-            $order = ConvertBytes::LITTLE_ENDIAN;
-        } elseif ($order_string === 'MM') {
-            $order = ConvertBytes::BIG_ENDIAN;
-        } else {
-            return null;
-        }
-
-        // Verify the TIFF header.
-        $magic_string = $data_element->getBytes($offset + 2, 2);
-        if (ConvertBytes::toShort($magic_string, $order) !== self::TIFF_HEADER) {
-            return null;
-        }
-
-        return $order;
-    }
-
     public function collectInfo(array $context = []): array
     {
         $info = [];

src/Command/CompileCommand.php

@@ -41,6 +41,7 @@ protected function execute(InputInterface $input, OutputInterface $output): int
     {
         $compiler = new SpecCompiler();
         $compiler->compile($input->getArgument('spec-dir'), $input->getArgument('resource-dir'));
+        $output->writeln('Compile OK');
         return(0);
     }
 }

src/ItemDefinition.php

@@ -25,7 +25,7 @@ class ItemDefinition
      *   The sequence of the item on its parent list.
      */
     public function __construct(
-        protected CollectionInterface $collection,
+        public readonly CollectionInterface $collection,
         protected int $format = DataFormat::BYTE,
         protected int $valuesCount = 1,
         protected int $dataOffset = 0,