Implement bytes.startswith in mypyc

[esarp]

Implements bytes.startswith in mypy. Potentially could be more efficient without relying on memcmp but not sure.

Tested with the following benchmark code, which shows a ~6.3x performance improvement compared to standard Python:

import time

def bench(prefix: bytes, a: list[bytes], n: int) -> int:
    i = 0
    for x in range(n):
        for b in a:
            if b.startswith(prefix):
                i += 1
    return i


a = [b"foo", b"barasdfsf", b"foobar", b"ab", b"asrtert", b"sertyeryt"]
n = 5 * 1000 * 1000
prefix = b"foo"

bench(prefix, a, n)

t0 = time.time()
bench(prefix, a, n)
td = time.time() - t0
print(f"{td}s")

Output:

$ python /tmp/bench.py
1.0015509128570557s
$ python -c 'import bench'
0.154998779296875s

[github-actions]

According to mypy_primer, this change doesn't affect type check results on a corpus of open source code. ✅

[BobTheBuidler]

maybe this if check should go above the 2 PyBytes_AS_STRING lines? We can exit without those calls if the check returns true

[esarp]

Good call, updated. I split the checks around each PyBytes_GET_SIZE call a bit further to optimize for the empty-arg case. Probably won't save a ton but I don't think it makes it unreadable

[JukkaL]

Test with bytearray 1) as the receiver object and 2) the argument. This way we will also test the slow path.

[esarp]

Added a few checks to cover those as well