Add initial support for modern JPEG compression

Some things still need to be done:
- Check RGB images
- Check small images
- Check huffman-only jpeg tables
- Check empty jpeg tables
- Add tests

Cargo.toml

@@ -18,6 +18,7 @@ exclude = ["tests/images/*", "tests/fuzz_images/*"]
 
 [dependencies]
 weezl = "0.1.0"
+jpeg = { package = "jpeg-decoder", version = "0.1.17", default-features = false }
 
 [dependencies.miniz_oxide]
 version = "0.4.1"

src/decoder/mod.rs

@@ -9,7 +9,7 @@ use self::ifd::Directory;
 use tags::{CompressionMethod, PhotometricInterpretation, Predictor, SampleFormat, Tag, Type};
 
 use self::stream::{
-    ByteOrder, DeflateReader, EndianReader, LZWReader, PackBitsReader, SmartReader,
+ByteOrder, DeflateReader, EndianReader, LZWReader, PackBitsReader, SmartReader, JpegReader
 };
 
 pub mod ifd;
@@ -710,6 +710,32 @@ impl<R: Read + Seek> Decoder<R> {
         }
     }
 
+    pub fn get_all_tags(&mut self) -> Vec<(Tag, ifd::Value)> {
+        let mut all_tags = Vec::new();
+
+        let dir = match &self.ifd {
+            Some(d) => d,
+            None => {
+                println!("No tags found");
+                return all_tags;
+            }
+        };
+        let keys = dir.keys().cloned().collect::<Vec<Tag>>();
+        // go through all tags
+        for key in keys {
+            let tag_value = match self.get_tag(key) {
+                Ok(val) => val,
+                _ => {
+                    println!("Could not unwrap {:?}", key);
+                    continue;
+                }
+            };
+            all_tags.push((key, tag_value));
+        }
+
+        all_tags
+    }
+
     /// Tries to retrieve a tag and convert it to the desired type.
     pub fn get_tag_u32(&mut self, tag: Tag) -> TiffResult<u32> {
         self.get_tag(tag)?.into_u32()
@@ -928,6 +954,56 @@ impl<R: Read + Seek> Decoder<R> {
         Ok(())
     }
 
+    pub fn read_jpeg_tiles(&mut self) -> TiffResult<DecodingResult> {
+        let tile_offsets = self.get_tag_u32_vec(Tag::TileOffsets)?;
+        let tile_bytes = self.get_tag_u32_vec(Tag::TileByteCounts)?;
+        let jpeg_tables: Vec<u8> = self.get_tag_u8_vec(Tag::JPEGTables)?;
+
+        let mut res_img = Vec::with_capacity(tile_offsets[0] as usize);
+
+        for (idx, offset) in tile_offsets.iter().enumerate() {
+
+            self.goto_offset(*offset)?;
+            let jpeg_reader = JpegReader::new(&mut self.reader, tile_bytes[idx], &jpeg_tables)?;
+            let mut decoder = jpeg::Decoder::new(jpeg_reader);
+
+            match decoder.decode() {
+                Err(e) => {
+                    panic!("The ERror was: {}", e);
+                },
+                Ok(mut val) => res_img.append(&mut val),
+            }
+        }
+
+        Ok(DecodingResult::U8(res_img))
+
+    }
+
+    pub fn read_jpeg_strips(&mut self) -> TiffResult<DecodingResult> {
+        let strip_offsets = self.get_tag_u32_vec(Tag::StripOffsets)?;
+        let strip_bytes = self.get_tag_u32_vec(Tag::StripByteCounts)?;
+        let jpeg_tables: Vec<u8> = self.get_tag_u8_vec(Tag::JPEGTables)?;
+
+        let mut res_img = Vec::with_capacity(strip_offsets[0] as usize);
+
+        for (idx, offset) in strip_offsets.iter().enumerate() {
+
+            self.goto_offset(*offset)?;
+            let jpeg_reader = JpegReader::new(&mut self.reader, strip_bytes[idx], &jpeg_tables)?;
+            let mut decoder = jpeg::Decoder::new(jpeg_reader);
+
+            match decoder.decode() {
+                Err(e) => {
+                    panic!("Error when reading jpeg tiles: {}", e);
+                },
+                Ok(mut val) => res_img.append(&mut val),
+            }
+        }
+
+        Ok(DecodingResult::U8(res_img))
+    }
+
+
     pub fn read_strip_to_buffer(&mut self, mut buffer: DecodingBuffer) -> TiffResult<()> {
         self.initialize_strip_decoder()?;
 
@@ -1050,6 +1126,14 @@ impl<R: Read + Seek> Decoder<R> {
 
     /// Decodes the entire image and return it as a Vector
     pub fn read_image(&mut self) -> TiffResult<DecodingResult> {
+
+        if self.compression_method == CompressionMethod::ModernJPEG {
+            match self.find_tag(Tag::TileOffsets) {
+                Ok(None) | Err(_) => return self.read_jpeg_strips(),
+                Ok(_) => return self.read_jpeg_tiles(),
+            }
+        }
+
         self.initialize_strip_decoder()?;
         let rows_per_strip =
             usize::try_from(self.get_tag_u32(Tag::RowsPerStrip).unwrap_or(self.height))?;

src/decoder/stream.rs

@@ -187,6 +187,14 @@ pub trait EndianReader: Read {
     }
 }
 
+///
+/// # READERS
+///
+
+///
+/// ## Deflate Reader
+///
+
 /// Reader that decompresses DEFLATE streams
 pub struct DeflateReader {
     buffer: io::Cursor<Vec<u8>>,
@@ -233,6 +241,10 @@ impl EndianReader for DeflateReader {
     }
 }
 
+///
+/// ## LZW Reader
+///
+
 /// Reader that decompresses LZW streams
 pub struct LZWReader {
     buffer: io::Cursor<Vec<u8>>,
@@ -306,6 +318,76 @@ impl EndianReader for LZWReader {
     }
 }
 
+///
+/// ## JPEG Reader (for "new-style" JPEG format (TIFF compression tag 7))
+///
+
+pub struct JpegReader {
+    buffer: io::Cursor<Vec<u8>>,
+    byte_order: ByteOrder, // for jpeg image data always ...,
+}
+// Only tested with qtables in jpegtables, but it contain both qtables and huffman tables or either
+impl JpegReader {
+    /// Constructs new JpegReader wrapping a SmartReader.
+    /// Because Jpeg compression in TIFF allows to save quantization and/or huffman tables in one
+    /// central loacation, the constructor accepts this data as `jpeg_tables` here containing either
+    /// both.
+    /// These `jpeg_tables` are simply prepended to the remaining jpeg image data.
+    /// Because these jpegtables start with a `SOI` (HEX: `0xFFD8`) or __start of image__ marker
+    /// which is also at the beginning of the remaining jpeg image data and would
+    /// confuse the jpeg renderer, one of these has to be taken off. In this case the first two
+    /// bytes of the remaining jpeg data is removed because this follows `jpeg_tables`.
+    /// Similary, `jpeg_tables` ends with a `EOI` (HEX: `0xFFD9`) or __end of image__ marker,
+    /// this has to be removed as well (last two bytes of `jpeg_tables`).
+
+    pub fn new<R>(
+        reader: &mut SmartReader<R>,
+        length: u32,
+        jpeg_tables: &Vec<u8>
+    ) -> io::Result<JpegReader>
+    where
+        R: Read + Seek,
+    {
+        let order = reader.byte_order;
+
+        // Read jpeg image data
+        let mut segment = vec![0; length as usize];
+        reader.read_exact(&mut segment[..])?;
+
+        let mut jpeg_data = jpeg_tables.clone();
+        jpeg_data.truncate(jpeg_data.len() - 2);
+
+        jpeg_data.extend_from_slice(&mut segment[2..]);
+
+        Ok(
+            JpegReader {
+                buffer: io::Cursor::new(jpeg_data),
+                byte_order: order,
+            },
+        )
+    }
+}
+
+impl Read for JpegReader {
+    // #[inline]
+    fn read(&mut self, buf: &mut [u8]) -> io::Result<usize> {
+        self.buffer.read(buf)
+    }
+}
+
+impl EndianReader for JpegReader {
+    #[inline(always)]
+    fn byte_order(&self) -> ByteOrder {
+        self.byte_order
+    }
+}
+
+
+
+///
+/// ## PackBits Reader
+///
+
 /// Reader that unpacks Apple's `PackBits` format
 pub struct PackBitsReader {
     buffer: io::Cursor<Vec<u8>>,
@@ -378,6 +460,10 @@ impl EndianReader for PackBitsReader {
     }
 }
 
+///
+/// ## SmartReader Reader
+///
+
 /// Reader that is aware of the byte order.
 #[derive(Debug)]
 pub struct SmartReader<R>

src/lib.rs

@@ -8,6 +8,7 @@
 
 extern crate miniz_oxide;
 extern crate weezl;
+extern crate jpeg;
 
 mod bytecast;
 pub mod decoder;

src/tags.rs

@@ -111,6 +111,13 @@ pub enum Tag(u16) unknown("A private or extension tag") {
     Threshholding = 263, // TODO add support
     XResolution = 282,
     YResolution = 283,
+    // JPEG
+    TileWidth = 322,
+    TileLength = 323,
+    TileOffsets = 324,
+    TileByteCounts = 325,
+    SampleFormat = 339,
+    JPEGTables = 347,
     // Advanced tags
     Predictor = 317,
     TileWidth = 322,
@@ -142,6 +149,8 @@ pub enum Type(u16) {
 }
 
 tags! {
+/// See [TIFF compression tags](https://www.awaresystems.be/imaging/tiff/tifftags/compression.html)
+/// for reference.
 pub enum CompressionMethod(u16) {
     None = 1,
     Huffman = 2,
@@ -149,6 +158,8 @@ pub enum CompressionMethod(u16) {
     Fax4 = 4,
     LZW = 5,
     JPEG = 6,
+    // "Extended JPEG" or "new JPEG" style
+    ModernJPEG = 7,
     Deflate = 8,
     OldDeflate = 0x80B2,
     PackBits = 0x8005,