Problem with copynone using libjpeg-turbo 3.0.0

[erikogabrielsson]

Hi,

I noticed that when using libjpeg-turbo 3.0.0 markers are removed regardless of copynone being set:

libjpeg-turbo 3.0.0:

turbo = TurboJPEG(r"C:\tools\libjpeg-turbo64-3.0.0\bin\turbojpeg.dll")
cropped = turbo.crop_multiple(tile_2, [(0, 0, 32, 32)], copynone=False)[0]
print(cropped[0:30])
>> b'\xff\xd8\xff\xe0\x00\x10JFIF\x00\x01\x01\x00\x00\x01\x00\x01\x00\x00\xff\xdb\x00C\x00\x06\x04\x05\x06\x05'

libjpeg-turbo 2.1.*

turbo = TurboJPEG(r"C:\tools\libjpeg-turbo64\bin\turbojpeg.dll")
cropped = turbo.crop_multiple(tile_2, [(0, 0, 32, 32)], copynone=False)[0]
print(cropped[0:30])
>> b'\xff\xd8\xff\xe0\x00\x10JFIF\x00\x01\x01\x00\x00\x01\x00\x01\x00\x00\xff\xfe\x00)Hamama'

On 2.1.* I get the FF-FE comment marker followed by the comment, on 3.0.0 it is missing.

Setting copynone=True on 2.1.* produces the desired output (no comment marker). Also jpegtran in 3.0.0 seems to work with the different copy-switches.

[lilohuang]

I looked at the release notes for libjpeg-turbo 3.0.0, and it appears that there's a known issue where metadata cannot be copied correctly. Could you please share your input file? This might help speed up the process. I'm not certain if I can completely resolve this problem. One more thing, PyTurboJPEG should already be utilizing the mechanism for automatic reallocation of the JPEG destination buffer...

---

https://github.com/libjpeg-turbo/libjpeg-turbo/releases/tag/3.0.0

---

[erikogabrielsson]

Here is the image I tested with. It is derived from CMU-1.ndpi (CC0-1.0 licence) using OpenTile.

[erikogabrielsson]

When viewing the file in JpegSnoop the comment is the only extra metadata, and it is of length 41.

[erikogabrielsson]

I have investigated this further. For some reason if I use tj3Transform instead of tjTransform I get the comment marker included in the result for copynone=False, see this branch..

[lilohuang]

Hi @erikogabrielsson ,

I apologize for not having had the opportunity to delve into this due to my daily work commitments. I am pleased to learn that you have found a solution, and I have reviewed your branch. It seems ideal to use getattr for checking tj3Transform availability. Could you kindly create a pull request so that I can incorporate your contribution into the mainstream?

Thanks,
Lilo

[lilohuang]

Hi @erikogabrielsson ,

Your code change has been merged.

Thanks,
Lilo

[erikogabrielsson]

Thank you @lilohuang. Your work with pyTurboJPEG has saved a lot of work so happy to contribute.