tsai
/
ImageSharp
mirror of https://github.com/SixLabors/ImageSharp
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
Browse Source

Decoder now doesn't break tests

pull/298/head
James Jackson-South 9 years ago
parent
ba8a5b34c8
commit
2f501eb1a4
3 changed files with 45 additions and 93 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 18
      src/ImageSharp/Formats/Jpeg/Port/Components/HuffmanTables.cs
  2. 117
      src/ImageSharp/Formats/Jpeg/Port/JpegDecoderCore.cs
  3. 3
      tests/ImageSharp.Tests/xunit.runner.json

18
src/ImageSharp/Formats/Jpeg/Port/Components/HuffmanTables.cs
View File

@ -13,9 +13,7 @@ namespace ImageSharp.Formats.Jpeg.Port.Components
/// </summary>
internal class HuffmanTables
{
private HuffmanBranch[] first;
private HuffmanBranch[] second;
private readonly HuffmanBranch[][] tables = new HuffmanBranch[4][];
/// <summary>
/// Gets or sets the table at the given index.
@ -27,23 +25,13 @@ namespace ImageSharp.Formats.Jpeg.Port.Components
[MethodImpl(MethodImplOptions.AggressiveInlining)]
get
{
if (index == 0)
{
return this.first;
}
return this.second;
return this.tables[index];
}
[MethodImpl(MethodImplOptions.AggressiveInlining)]
set
{
if (index == 0)
{
this.first = value;
}
this.second = value;
this.tables[index] = value;
}
}
}

117
src/ImageSharp/Formats/Jpeg/Port/JpegDecoderCore.cs
View File

@ -90,47 +90,32 @@ namespace ImageSharp.Formats.Jpeg.Port
public static FileMarker FindNextFileMarkerOld(Stream stream)
{
byte[] buffer = new byte[2];
while (true)
int value = stream.Read(buffer, 0, 2);
if (value == 0)
{
int value = stream.Read(buffer, 0, 2);
return new FileMarker(JpegConstants.Markers.EOI, (int)stream.Length, true);
}
if (value == 0)
{
return new FileMarker(JpegConstants.Markers.EOI, (int)stream.Length, true);
}
// According to Section B.1.1.2:
// "Any marker may optionally be preceded by any number of fill bytes, which are bytes assigned code 0xFF."
if (buffer[1] != JpegConstants.Markers.Prefix)
{
return new FileMarker((ushort)((buffer[0] << 8) | buffer[1]), (int)(stream.Position - 2));
}
while (buffer[0] != JpegConstants.Markers.Prefix)
while (buffer[1] == JpegConstants.Markers.Prefix)
{
int suffix = stream.ReadByte();
if (suffix == -1)
{
// Strictly speaking, this is a format error. However, libjpeg is
// liberal in what it accepts. As of version 9, next_marker in
// jdmarker.c treats this as a warning (JWRN_EXTRANEOUS_DATA) and
// continues to decode the stream. Even before next_marker sees
// extraneous data, jpeg_fill_bit_buffer in jdhuff.c reads as many
// bytes as it can, possibly past the end of a scan's data. It
// effectively puts back any markers that it overscanned (e.g. an
// "\xff\xd9" EOI marker), but it does not put back non-marker data,
// and thus it can silently ignore a small number of extraneous
// non-marker bytes before next_marker has a chance to see them (and
// print a warning).
// We are therefore also liberal in what we accept. Extraneous data
// is silently ignore
// This is similar to, but not exactly the same as, the restart
// mechanism within a scan (the RST[0-7] markers).
// Note that extraneous 0xff bytes in e.g. SOS data are escaped as
// "\xff\x00", and so are detected a little further down below.
buffer[0] = buffer[1];
value = stream.ReadByte();
if (value == -1)
{
return new FileMarker(JpegConstants.Markers.EOI, (int)stream.Length, true);
}
buffer[1] = (byte)value;
return new FileMarker(JpegConstants.Markers.EOI, (int)stream.Length, true);
}
return new FileMarker((ushort)((buffer[0] << 8) | buffer[1]), (int)(stream.Position - 2));
buffer[1] = (byte)value;
}
return new FileMarker((ushort)((buffer[0] << 8) | buffer[1]), (int)(stream.Position - 2));
}
/// <summary>
@ -185,45 +170,6 @@ namespace ImageSharp.Formats.Jpeg.Port
return new FileMarker(newMarker, newPos, true);
}
/// <summary>
/// Finds the next file marker within the byte stream. Used for testing. Slower as it only reads on byte at a time
/// </summary>
/// <param name="stream">The input stream</param>
/// <returns>The <see cref="FileMarker"/></returns>
public static FileMarker FindNextFileMarkerNew(Stream stream)
{
while (true)
{
int value = stream.ReadByte();
if (value == -1)
{
// We've reached the end of the stream
return new FileMarker(JpegConstants.Markers.EOI, (int)stream.Length, true);
}
byte prefix = (byte)value;
byte suffix = JpegConstants.Markers.Prefix;
// According to Section B.1.1.2:
// "Any marker may optionally be preceded by any number of fill bytes, which are bytes assigned code 0xFF."
while (prefix == JpegConstants.Markers.Prefix && suffix == JpegConstants.Markers.Prefix)
{
value = stream.ReadByte();
if (value == -1)
{
// We've reached the end of the stream
return new FileMarker(JpegConstants.Markers.EOI, (int)stream.Length, true);
}
suffix = (byte)value;
}
return new FileMarker((ushort)((prefix << 8) | suffix), (int)(stream.Position - 2));
}
}
/// <summary>
/// Decodes the image from the specified <see cref="Stream"/> and sets
/// the data to image.
@ -357,11 +303,12 @@ namespace ImageSharp.Formats.Jpeg.Port
break;
}
throw new ImageFormatException($"Unknown Marker {fileMarker.Marker} at {fileMarker.Position}");
// throw new ImageFormatException($"Unknown Marker {fileMarker.Marker} at {fileMarker.Position}");
break;
}
// Read on. TODO: Test this on damaged images.
fileMarker = FindNextFileMarkerNew(this.InputStream);
fileMarker = FindNextFileMarkerOld(this.InputStream);
}
this.width = this.frame.SamplesPerLine;
@ -644,7 +591,6 @@ namespace ImageSharp.Formats.Jpeg.Port
i += 17 + codeLengthSum;
// Everything I can discover indicates there's a max of two table per DC AC pair though this limits the index to 16?
this.BuildHuffmanTable(
huffmanTableSpec >> 4 == 0 ? this.dcHuffmanTables : this.acHuffmanTables,
huffmanTableSpec & 15,
@ -679,7 +625,23 @@ namespace ImageSharp.Formats.Jpeg.Port
int selectorsCount = this.InputStream.ReadByte();
for (int i = 0; i < selectorsCount; i++)
{
byte componentIndex = this.frame.ComponentIds[this.InputStream.ReadByte() - 1];
int index = -1;
int selector = this.InputStream.ReadByte();
foreach (byte id in this.frame.ComponentIds)
{
if (selector == id)
{
index = selector;
}
}
if (index < 0)
{
throw new ImageFormatException("Unknown component selector");
}
byte componentIndex = this.frame.ComponentIds[index];
ref FrameComponent component = ref this.frame.Components[componentIndex];
int tableSpec = this.InputStream.ReadByte();
component.DCHuffmanTableId = tableSpec >> 4;
@ -717,6 +679,7 @@ namespace ImageSharp.Formats.Jpeg.Port
/// <summary>
/// Builds the huffman tables
/// TODO: This is our bottleneck. We should use a faster algorithm with a LUT.
/// </summary>
/// <param name="tables">The tables</param>
/// <param name="index">The table index</param>

3
tests/ImageSharp.Tests/xunit.runner.json
View File

@ -1,3 +1,4 @@
{
"methodDisplay": "method"
"methodDisplay": "method",
"diagnosticMessages": true
}
Write Preview
Loading…
Cancel
Save