Christoph Gohlke

Could be related to this: the [current wheels on PyPI](https://pypi.org/project/pyopencl/#files) contain many superfluous files from the source distribution that get installed into the Python root folder (at least on Windows)....

> the python `tifffile` package tells me that the `tag` 32781 is malformed The value of tag `32781` is not a null terminated 7-bit `ASCII` string but a sequence of...

There is currently no API for deleting tags, just changing tag types and values. It should be easy to add in principle, but would leave many class instances out of...

A sample file was posted at [forum.image.sc](https://forum.image.sc/t/trouble-generating-ome-tiffs-in-the-right-shape/89191/4). See also the [comments](https://forum.image.sc/t/trouble-generating-ome-tiffs-in-the-right-shape/89191/12) regarding the file format.

Tifffile does not do any conversion or remapping. `tif.pages[0].tags['ColorMap'].value` is a numpy array of dtype `uint16`. The conversion in the example code is `int.from_bytes(r[i])`, which is wrong in general but...

The files are shared on Zenodo: https://zenodo.org/records/8046636

> JPEG8_DECODE: Not a JPEG file: starts with 0xff 0x0a The file is a JPEG XL image and imagecodecs was built without the JPEGXL codec.

Does using a sequence of ImagejRoi not work for you? For example: https://github.com/cgohlke/roifile/blob/c966c36777d4974a93257f03e49046d4b300e1a8/roifile_demo.py#L49-L57

There should probably be a little example for reading/writing TIFF in the main docstring.

In #119, `filterwarnings = ["error"]` was disabled for pytest because it broke debugging tests in VSCode due to a bug in Pylance.