Jpeg color space deduction fix #2177
Conversation
|
There were 3 failing tiff tests, but I believe it was an issue with the Tiff JpegDecompressor. Hopefully this 1959bf6 fixes it now. |
I believe it's not not that simple :D TIFF with PhotometricInterpretation=RGB can have YCbCr signature at jpeg decoding level which would lead to invalid results. |
| @@ -506,90 +506,48 @@ public void Dispose() | |||
| /// Returns the correct colorspace based on the image component count and the jpeg frame component id's. | |||
There was a problem hiding this comment.
The comment is no longer valid, maybe change it like this:
| /// Returns the correct colorspace based on the image component count and the jpeg frame component id's. | |
| /// Returns the correct colorspace based on the image component count and adobe APP14 marker's color transform flag. |
| { | ||
| if (componentCount == 1) |
There was a problem hiding this comment.
I have found a jpeg image with an app14 marker which is gray, should that not be possible according to the spec?
There was a problem hiding this comment.
Adobe spec doesn't say anything about grayscale images but I think we can include it as decoding it doesn't violate the spec :)
|
I've implemented tests for the specific method which resolves encoded jpeg color space. Not only it's easier to maintain, it is also much faster and there's no need to find a ton of images with different setups. This is ready to be merged after the review. |
| [InlineData(3, JpegConstants.Adobe.ColorTransformYCbCr, JpegColorSpace.YCbCr)] | ||
| [InlineData(4, JpegConstants.Adobe.ColorTransformUnknown, JpegColorSpace.Cmyk)] | ||
| [InlineData(4, JpegConstants.Adobe.ColorTransformYcck, JpegColorSpace.Ycck)] | ||
| internal void DeduceJpegColorSpaceAdobeMarker_ShouldReturnValidColorSpace(byte componentCount, byte adobeFlag, JpegColorSpace expectedColorSpace) |
There was a problem hiding this comment.
This is all very nice 👍
Prerequisites
Description
Fixes #2174