In release 0.11.0, although flush() is now supported on decompressobj, it returns None unlike zlib.decompressobj which returns empty string.

From the python documentation:

Decompress.flush([length])
All pending input is processed, and a string containing the remaining uncompressed output is returned. After calling flush(), the decompress() method cannot be called again; the only realistic action is to delete the object.

The optional parameter length sets the initial size of the output buffer.

This behavior makes special handling necessary for zstandard decompressor when using the return value directly in things like file object's write(), which do not accept None.

Can we consider returning empty string instead?

Originally posted by @vnair81 in #65 (comment)