Convert from '_io.BytesIO' to a bytes-like object in python3.6?

Question

I am using this function to uncompress the body of a HTTP response if it is compressed with gzip, compress or deflate.

def uncompress_body(self, compression_type, body):
    if compression_type == 'gzip' or compression_type == 'compress':
        return zlib.decompress(body)
    elif compression_type == 'deflate':
        compressor = zlib.compressobj(9, zlib.DEFLATED, -zlib.MAX_WBITS)
        compressed = compressor.compress(body)
        compressed += compressor.flush()
        return base64.b64encode(compressed)

    return body

However python throws this error message.

TypeError: a bytes-like object is required, not '_io.BytesIO'

on this line:

return zlib.decompress(body)

Essentially, how do I convert from '_io.BytesIO' to a bytes-like object?

score 88 · Accepted Answer · answered Jan 10 '19 at 22:28

88

It's a file-like object. Read them:

>>> b = io.BytesIO(b'hello')
>>> b.read()
b'hello'

If the data coming in from body is too large to read into memory, you'll want to refactor your code and use zlib.decompressobj instead of zlib.decompress.

answered Jan 10 '19 at 22:28

wim

338,267
99
616
750

When you overthink a basic solution :D – Cyril N. Jan 07 '21 at 13:29

score 57 · Answer 2 · answered Sep 19 '19 at 07:46

In case you write into the object first, make sure to reset the stream before reading:

>>> b = io.BytesIO()
>>> image = PIL.Image.open(path_to_image)
>>> image.save(b, format='PNG')
>>> b.seek(0)
>>> b.read()
b'\x89PNG\r\n\x1a\n\x00\x00\x00\rIHDR\x00\x00\x06\xcf\x00\x00\x03W\x08\x02\x00'

or directly get the data with getvalue

>>> b.getvalue()
b'\x89PNG\r\n\x1a\n\x00\x00\x00\rIHDR\x00\x00\x06\xcf\x00\x00\x03W\x08\x02\x00'

score 5 · Answer 3 · answered Mar 28 '22 at 13:09

5

b = io.BytesIO()
image = PIL.Image.open(path_to_image) # ie 'image.png'
image.save(b, format='PNG')
b.getbuffer().tobytes() # b'\x89PNG\r\n\x1a\n\x00 ...

answered Mar 28 '22 at 13:09

Nwawel A Iroume

1,249
3
21
42

2

Please add context as to *why* this is an answer. – S3DEV Mar 28 '22 at 15:24
1

One upside to this answer is that you don't need to reset the stream with `b.seek` to get the full bytes that were written to it. – Chris Collett Apr 28 '22 at 20:20

score 1 · Answer 4 · answered Jun 06 '23 at 11:56

I know its late for the answer but if anyone finds into same issue then the below code snippet might be useful.

from io import BytesIO


data = BytesIO(b'Some data')

bytes_data = data.getvalue()

print(bytes_data, type(bytes_data)

Output

b'Some data' <class 'bytes'>

Convert from '_io.BytesIO' to a bytes-like object in python3.6?

4 Answers4

Linked