LZMADecompressor not working correctly - python-3.x

Question

I am trying to use lzma to compress and decompress some data in memory. I know that the following approach works:
import lzma
s = 'Lorem ipsum dolor'
bytes_in = s.encode('utf-8')
print(s)
print(bytes_in)
# Compress
bytes_out = lzma.compress(data=bytes_in, format=lzma.FORMAT_XZ)
print(bytes_out)
# Decompress
bytes_decomp = lzma.decompress(data=bytes_out, format=lzma.FORMAT_XZ)
print(bytes_decomp)
The output is:
Lorem ipsum dolor
b'Lorem ipsum dolor'
b'\xfd7zXZ\x00\x00\x04\xe6\xd6\xb4F\x02\x00!\x01\x16\x00\x00\x00t/\xe5\xa3\x01\x00\x10Lorem ipsum dolor\x00\x00\x00\x00\xddq\x8e\x1d\x82\xc8\xef\xad\x00\x01)\x112\np\x0e\x1f\xb6\xf3}\x01\x00\x00\x00\x00\x04YZ'
b'Lorem ipsum dolor'
However, I notice that using lzma.LZMACompressor gives different results. With the following code:
import lzma
s = 'Lorem ipsum dolor'
bytes_in = s.encode('utf-8')
print(s)
print(bytes_in)
# Compress
lzc = lzma.LZMACompressor(format=lzma.FORMAT_XZ)
lzc.compress(bytes_in)
bytes_out = lzc.flush()
print(bytes_out)
# Decompress
bytes_decomp = lzma.decompress(data=bytes_out, format=lzma.FORMAT_XZ)
print(bytes_decomp)
I get this output:
Lorem ipsum dolor
b'Lorem ipsum dolor'
b'\x01\x00\x10Lorem ipsum dolor\x00\x00\x00\x00\xddq\x8e\x1d\x82\xc8\xef\xad\x00\x01)\x112\np\x0e\x1f\xb6\xf3}\x01\x00\x00\x00\x00\x04YZ'
And then the program fails on line 18 with _lzma.LZMAError: Input format not supported by decoder.
I have 3 questions here:
How come the output for lzma.compress is so much longer than lzma.LZMACompressor.compress even though it seemingly does the same thing?
In the second example, why does the decompressor complain about invalid format?
How can I get the second example to decompress correctly?

Answer 1

On your second example you're dropping a part of the compressed stream, and bytes_out only gets the flush part. On the other hand, that works:
lzc = lzma.LZMACompressor(format=lzma.FORMAT_XZ)
bytes_out = lzc.compress(bytes_in) + lzc.flush()
print(bytes_out)
note that the first example is really equivalent since source for lzma.compress is:
def compress(data, format=FORMAT_XZ, check=-1, preset=None, filters=None):
"""Compress a block of data.
Refer to LZMACompressor's docstring for a description of the
optional arguments *format*, *check*, *preset* and *filters*.
For incremental compression, use an LZMACompressor instead.
"""
comp = LZMACompressor(format, check, preset, filters)
return comp.compress(data) + comp.flush()