Read into a bytearray at an offset?

Question

How can I use the readinto() method call to an offset inside a bytearray, in the same way that struct.unpack_from works?

srgerg · Accepted Answer · 2011-11-26T04:36:28.217

19

You can use a memoryview to do the job. For example:

dest = bytearray(10) # all zero bytes
v = memoryview(dest)
ioObject.readinto(v[3:])
print(repr(dest))

Assuming that iObject.readinto(...) reads the bytes 1, 2, 3, 4, and 5 then this code prints:

bytearray(b'\x00\x00\x00\x01\x02\x03\x04\x05\x00\x00')

You can also use the memoryview object with struct.unpack_from and struct.pack_into. For example:

dest = bytearray(10) # all zero bytes
v = memoryview(dest)
struct.pack_into("2c", v[3:5], 0, b'\x07', b'\x08')
print(repr(dest))

This code prints

bytearray(b'\x00\x00\x00\x07\x08\x00\x00\x00\x00\x00')

edited Nov 26 '11 at 04:36

answered Nov 25 '11 at 02:25

srgerg

1

My intention is to read directly into the bytearray at an offset, and avoid all the intermediate copying. – Matt Joiner Nov 25 '11 at 02:38
I've edited my answer to include an example using the `memoryview` class, which I think does what you want. – srgerg Nov 25 '11 at 02:53
1

This memoryview form seems to accomplish what I wanted. If I did a similar thing with struct.unpack_into, and used a memoryview instead of an offset, would it be the same? If this is the case, please remove the rest of your answer to focus on this. – Matt Joiner Nov 25 '11 at 06:01
Thanks, I was checking whether I need to add this as an answer – Antti Haapala -- Слава Україні May 23 '16 at 05:08

1 Answers1