Post History

# Python (3.6 and up), 32 bytes (unsigned integer I/O)

```
lambda x:int(f'{x:08b}'[::-1],2)
```

Input and output are `int` objects. This also reverses the bits in integers larger than 255, implicitly inferring their bit length from the number of bits needed to write them (even if that isn't a multiple of 8). It will raise an exception if given a negative number as input.

## Python (3.6 and up), 44 bytes (byte I/O)

```
lambda x:bytes([int(f'{x[0]:08b}'[::-1],2)])
```

Input and output are both actual `bytes` objects. This uses new-style string formatting to produce the bit-pattern directly (and takes advantage of the fact that indexing a `bytes` in 3.x gives an integer value), and uses the `bytes` constructor to produce a valid `bytes` object as the result (In Python 3.x, `chr` produces a string, not `bytes`).