Merge pull request #95 from ruffson/support_modern_jpeg

Support decoding of modern JPEG compression

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/ifd.rs

@@ -10,13 +10,14 @@ use tags::{Tag, Type};
 use {TiffError, TiffFormatError, TiffResult, TiffUnsupportedError};
 
 use self::Value::{
-    Ascii, Double, Float, List, Rational, RationalBig, SRational, SRationalBig, Signed, SignedBig,
+    Ascii, Byte, Double, Float, List, Rational, RationalBig, SRational, SRationalBig, Signed,
-    Unsigned, UnsignedBig,
+    SignedBig, Unsigned, UnsignedBig,
 };
 
 #[allow(unused_qualifications)]
 #[derive(Debug, Clone, PartialEq)]
 pub enum Value {
+    Byte(u8),
     Signed(i32),
     SignedBig(i64),
     Unsigned(u32),
@@ -34,6 +35,13 @@ pub enum Value {
 }
 
 impl Value {
+    pub fn into_u8(self) -> TiffResult<u8> {
+        match self {
+            Byte(val) => Ok(val),
+            val => Err(TiffError::FormatError(TiffFormatError::ByteExpected(val))),
+        }
+    }
+
     pub fn into_u32(self) -> TiffResult<u32> {
         match self {
             Unsigned(val) => Ok(val),
@@ -114,6 +122,23 @@ impl Value {
         }
     }
 
+    pub fn into_u8_vec(self) -> TiffResult<Vec<u8>> {
+        match self {
+            List(vec) => {
+                let mut new_vec = Vec::with_capacity(vec.len());
+                for v in vec {
+                    new_vec.push(v.into_u8()?)
+                }
+                Ok(new_vec)
+            }
+            Byte(val) => Ok(vec![val]),
+
+            val => Err(TiffError::FormatError(
+                TiffFormatError::UnsignedIntegerExpected(val),
+            )),
+        }
+    }
+
     pub fn into_i32_vec(self) -> TiffResult<Vec<i32>> {
         match self {
             List(vec) => {
@@ -426,6 +451,9 @@ impl Entry {
                         Ok(Ascii(string))
                     }
                 }
+                (Type::UNDEFINED, n) => self.decode_offset(n, bo, limits, decoder, |decoder| {
+                    Ok(Byte(u8::from(decoder.read_byte()?)))
+                }),
                 _ => Err(TiffError::UnsupportedError(
                     TiffUnsupportedError::UnsupportedDataType,
                 )),

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, JpegReader, LZWReader, PackBitsReader, SmartReader,
 };
 
 pub mod ifd;
@@ -165,6 +165,9 @@ pub struct Limits {
     /// The maximum size of any ifd value in bytes, the default is
     /// 1MiB.
     pub ifd_value_size: usize,
+    /// Maximum size for intermediate buffer which may be used to limit the amount of data read per
+    /// segment even if the entire image is decoded at once.
+    pub intermediate_buffer_size: usize,
     /// The purpose of this is to prevent all the fields of the struct from
     /// being public, as this would make adding new fields a major version
     /// bump.
@@ -175,6 +178,7 @@ impl Default for Limits {
     fn default() -> Limits {
         Limits {
             decoding_buffer_size: 256 * 1024 * 1024,
+            intermediate_buffer_size: 128 * 1024 * 1024,
             ifd_value_size: 1024 * 1024,
             _non_exhaustive: (),
         }
@@ -638,7 +642,9 @@ impl<R: Read + Seek> Decoder<R> {
         for _ in 0..num_tags {
             let (tag, entry) = match self.read_entry()? {
                 Some(val) => val,
-                None => continue, // Unknown data type in tag, skip
+                None => {
+                    continue;
+                } // Unknown data type in tag, skip
             };
             dir.insert(tag, entry);
         }
@@ -746,6 +752,11 @@ impl<R: Read + Seek> Decoder<R> {
         self.get_tag(tag)?.into_f64_vec()
     }
 
+    /// Tries to retrieve a tag and convert it to a 8bit vector.
+    pub fn get_tag_u8_vec(&mut self, tag: Tag) -> TiffResult<Vec<u8>> {
+        self.get_tag(tag)?.into_u8_vec()
+    }
+
     /// Decompresses the strip into the supplied buffer.
     /// Returns the number of bytes read.
     fn expand_strip<'a>(
@@ -923,6 +934,60 @@ impl<R: Read + Seek> Decoder<R> {
         Ok(())
     }
 
+    pub fn read_jpeg(&mut self) -> TiffResult<DecodingResult> {
+        let offsets = self.get_tag_u32_vec(Tag::StripOffsets)?;
+        let bytes = self.get_tag_u32_vec(Tag::StripByteCounts)?;
+
+        let jpeg_tables: Option<Vec<u8>> = match self.find_tag(Tag::JPEGTables) {
+            Ok(None) => None,
+            Ok(_) => Some(self.get_tag_u8_vec(Tag::JPEGTables)?),
+            Err(e) => return Err(e),
+        };
+
+        if offsets.len() == 0 {
+            return Err(TiffError::FormatError(TiffFormatError::RequiredTagEmpty(
+                Tag::StripOffsets,
+            )));
+        }
+        if offsets.len() != bytes.len() {
+            return Err(TiffError::FormatError(
+                TiffFormatError::InconsistentSizesEncountered,
+            ));
+        }
+
+        let mut res_img = Vec::with_capacity(offsets[0] as usize);
+
+        for (idx, offset) in offsets.iter().enumerate() {
+            if bytes[idx] as usize > self.limits.intermediate_buffer_size {
+                return Err(TiffError::LimitsExceeded);
+            }
+
+            self.goto_offset(*offset)?;
+            let jpeg_reader = JpegReader::new(&mut self.reader, bytes[idx], &jpeg_tables)?;
+            let mut decoder = jpeg::Decoder::new(jpeg_reader);
+
+            match decoder.decode() {
+                Ok(mut val) => res_img.append(&mut val),
+                Err(e) => {
+                    return match e {
+                        jpeg::Error::Io(io_err) => Err(TiffError::IoError(io_err)),
+                        jpeg::Error::Format(fmt_err) => {
+                            Err(TiffError::FormatError(TiffFormatError::Format(fmt_err)))
+                        }
+                        jpeg::Error::Unsupported(_) => Err(TiffError::UnsupportedError(
+                            TiffUnsupportedError::UnknownInterpretation,
+                        )),
+                        jpeg::Error::Internal(_) => Err(TiffError::UnsupportedError(
+                            TiffUnsupportedError::UnknownInterpretation,
+                        )),
+                    }
+                }
+            }
+        }
+
+        Ok(DecodingResult::U8(res_img))
+    }
+
     pub fn read_strip_to_buffer(&mut self, mut buffer: DecodingBuffer) -> TiffResult<()> {
         self.initialize_strip_decoder()?;
 
@@ -1045,6 +1110,10 @@ 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 {
+            return self.read_jpeg();
+        }
+
         self.initialize_strip_decoder()?;
         let rows_per_strip =
             usize::try_from(self.get_tag_u32(Tag::RowsPerStrip).unwrap_or(self.height))?;
@@ -1057,7 +1126,6 @@ impl<R: Read + Seek> Decoder<R> {
         for i in 0..usize::try_from(self.strip_count()?)? {
             self.read_strip_to_buffer(result.as_buffer(samples_per_strip * i))?;
         }
-
         Ok(result)
     }
 }

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 +318,79 @@ impl EndianReader for LZWReader {
     }
 }
 
+///
+/// ## JPEG Reader (for "new-style" JPEG format (TIFF compression tag 7))
+///
+
+pub(crate) struct JpegReader {
+    buffer: io::Cursor<Vec<u8>>,
+    byte_order: ByteOrder,
+}
+impl JpegReader {
+    /// Constructs new JpegReader wrapping a SmartReader.
+    /// Because JPEG compression in TIFF allows to save quantization and/or huffman tables in one
+    /// central location, the constructor accepts this data as `jpeg_tables` here containing either
+    /// or both.
+    /// These `jpeg_tables` are simply prepended to the remaining jpeg image data.
+    /// Because these `jpeg_tables` 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 it 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: &Option<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[..])?;
+
+        match jpeg_tables {
+            Some(tables) => {
+                let mut jpeg_data = tables.clone();
+                let truncated_length = jpeg_data.len() - 2;
+                jpeg_data.truncate(truncated_length);
+                jpeg_data.extend_from_slice(&mut segment[2..]);
+
+                Ok(JpegReader {
+                    buffer: io::Cursor::new(jpeg_data),
+                    byte_order: order,
+                })
+            }
+            None => Ok(JpegReader {
+                buffer: io::Cursor::new(segment),
+                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 +463,10 @@ impl EndianReader for PackBitsReader {
     }
 }
 
+///
+/// ## SmartReader Reader
+///
+
 /// Reader that is aware of the byte order.
 #[derive(Debug)]
 pub struct SmartReader<R>

src/error.rs

@@ -45,9 +45,12 @@ pub enum TiffFormatError {
     InvalidTagValueType(Tag),
     RequiredTagNotFound(Tag),
     UnknownPredictor(u16),
+    ByteExpected(Value),
     UnsignedIntegerExpected(Value),
     SignedIntegerExpected(Value),
     InflateError(InflateError),
+    Format(String),
+    RequiredTagEmpty(Tag),
     #[doc(hidden)]
     /// Do not match against this variant. It may get removed.
     __NonExhaustive,
@@ -69,6 +72,7 @@ impl fmt::Display for TiffFormatError {
             UnknownPredictor(ref predictor) => {
                 write!(fmt, "Unknown predictor “{}” encountered", predictor)
             }
+            ByteExpected(ref val) => write!(fmt, "Expected byte, {:?} found.", val),
             UnsignedIntegerExpected(ref val) => {
                 write!(fmt, "Expected unsigned integer, {:?} found.", val)
             }
@@ -76,6 +80,8 @@ impl fmt::Display for TiffFormatError {
                 write!(fmt, "Expected signed integer, {:?} found.", val)
             }
             InflateError(_) => write!(fmt, "Failed to decode inflate data."),
+            Format(ref val) => write!(fmt, "Invalid format: {:?}.", val),
+            RequiredTagEmpty(ref val) => write!(fmt, "Required tag {:?} was empty.", val),
             __NonExhaustive => unreachable!(),
         }
     }

src/lib.rs

@@ -6,6 +6,7 @@
 //! # Related Links
 //! * <https://www.adobe.io/open/standards/TIFF.html> - The TIFF specification
 
+extern crate jpeg;
 extern crate miniz_oxide;
 extern crate weezl;
 

src/tags.rs

@@ -111,6 +111,8 @@ pub enum Tag(u16) unknown("A private or extension tag") {
     Threshholding = 263, // TODO add support
     XResolution = 282,
     YResolution = 283,
+    // JPEG
+    JPEGTables = 347,
     // Advanced tags
     Predictor = 317,
     TileWidth = 322,
@@ -130,6 +132,7 @@ pub enum Type(u16) {
     LONG = 4,
     RATIONAL = 5,
     SBYTE = 6,
+    UNDEFINED = 7,
     SSHORT = 8,
     SLONG = 9,
     SRATIONAL = 10,
@@ -141,6 +144,8 @@ 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,
@@ -148,6 +153,8 @@ pub enum CompressionMethod(u16) {
     Fax4 = 4,
     LZW = 5,
     JPEG = 6,
+    // "Extended JPEG" or "new JPEG" style
+    ModernJPEG = 7,
     Deflate = 8,
     OldDeflate = 0x80B2,
     PackBits = 0x8005,