Andrew Murray

Instead of using the "exif" part of the `info` dictionary, you can use `getexif()`. Within that data, you will want to get the [EXIF IFD](https://www.awaresystems.be/imaging/tiff/tifftags/privateifd/exif.html), pointed to by the [EXIF...

The EXIF IFD tag. See the "Code" value at https://www.awaresystems.be/imaging/tiff/tifftags/exififd.html Within the `getexif()`, there are a few sections, IFDs. There is [EXIF, GPS and Interoperability](https://www.awaresystems.be/imaging/tiff/tifftags/privateifd.html). We have limited support for...

Your question is no longer about EXIF data. It would be good to either create a new issue, or continue this in #6467. It shouldn't matter at what point in...

Regarding an image without "icc_profile" in the `info` dictionary, https://stackoverflow.com/questions/50641637/identify-colour-space-of-any-image-if-icc-profile-is-empty-pil appears helpful. You've also asked whether it is better to apply an ICC profile or to resize the image first....

@julin69 has that answered your questions?

@pankajsinghal does [this commit](https://github.com/rakeshvar/unicode_text_to_image_array/commit/061b14bb20f023ac6eef0187c0dc5722bc320e42) mean that your problem is resolved?

https://www.w3.org/International/questions/qa-scripts#:~:text=Ideographic%20languages%20(e.g.%20Japanese%2C%20Korean,occasionally%20written%20right%20to%20left. > Ideographic languages (e.g. Japanese, Korean, Chinese) are more flexible in their writing direction. They are generally written left-to-right, or vertically top-to-bottom (with the vertical lines proceeding from right...

Would you mind providing some detail on how you ran the static analyzer? What tool did you use specifically?

https://github.com/python-pillow/Pillow/pull/6326#issuecomment-1135809007 > I believe this change will cause undefined behaviour by reading deallocated memory. > > From https://docs.python.org/3/c-api/bytes.html#c.PyBytes_AsString (emphasis mine): > > > char *PyBytes_AsString([PyObject](https://docs.python.org/3/c-api/structures.html#c.PyObject) *o) > > Part of...

@DeusAres there isn't an open PR for this, so no, there isn't any work currently in progress. But that doesn't mean that such a change would be rejected. Feel free...