view contrib/fuzz/fm1readmarkers_corpus.py @ 50400:95acba2c29f6

encoding: avoid quadratic time complexity when json-encoding non-UTF8 strings

Apparently the code uses "+=" with a bytes object, which is linear-time, so the
whole encoding is quadratic-time. This patch makes us use a bytearray object,
instead, which has a(n amortized-)constant-time append operation.

The encoding is still not particularly fast, but at least a 10MB file
takes tens of seconds, not many hours to encode.

import argparse
import zipfile

ap = argparse.ArgumentParser()
ap.add_argument("out", metavar="some.zip", type=str, nargs=1)
args = ap.parse_args()

with zipfile.ZipFile(args.out[0], "w", zipfile.ZIP_STORED) as zf:
    zf.writestr(
        'smallish_obsstore',
        (
            # header: fm1readmarkers should start at offset 1, and
            # read until byte 597.
            '1\x00597\x00'
            # body of obsstore file
            '\x01\x00\x00\x00vA\xd7\x02+C\x1a<)\x01,\x00\x00\x01\x03\x03\xe6'
            '\x92\xde)x\x16\xd1Xph\xc7\xa7[\xe5\xe2\x1a\xab\x1e6e\xaf\xc2\xae'
            '\xe7\xbc\x83\xe1\x88\xa5\xda\xce>O\xbd\x04\xe9\x03\xc4o\xeb\x03'
            '\x01\t\x05\x04\x1fef18operationamenduserAugie Fackler <raf@duri'
            'n42.com>\x00\x00\x00vA\xd7\x02-\x8aD\xaf-\x01,\x00\x00\x01\x03\x03'
            '\x17*\xca\x8f\x9e}i\xe0i\xbb\xdf\x9fb\x03\xd2XG?\xd3h\x98\x89\x1a'
            '=2\xeb\xc3\xc5<\xb3\x9e\xcc\x0e;#\xee\xc3\x10ux\x03\x01\t\x05\x04'
            '\x1fef18operationamenduserAugie Fackler <[e-mail]>\x00\x00'
            '\x00vA\xd7\x02Mn\xd9%\xea\x01,\x00\x00\x01\x03\x03\x98\x89\x1a='
            '2\xeb\xc3\xc5<\xb3\x9e\xcc\x0e;#\xee\xc3\x10ux\xe0*\xcaT\x86Z8J'
            '\x85)\x97\xff7\xcc)\xc1\x7f\x19\x0c\x01\x03\x01\t\x05\x04\x1fef'
            '18operationamenduserAugie Fackler <[e-mail]>\x00\x00\x00'
            'yA\xd7\x02MtA\xbfj\x01,\x00\x00\x01\x03\x03\xe0*\xcaT\x86Z8J\x85'
            ')\x97\xff7\xcc)\xc1\x7f\x19\x0c\x01\x00\x94\x01\xa9\n\xf80\x92\xa3'
            'j\xc5X\xb1\xc9:\xd51\xb8*\xa9\x03\x01\t\x08\x04\x1fef11operatio'
            'nhistedituserAugie Fackler <[e-mail]>\x00\x00\x00yA\xd7\x02'
            'MtA\xd4\xe1\x01,\x00\x00\x01\x03\x03"\xa5\xcb\x86\xb6\xf4\xbaO\xa0'
            'sH\xe7?\xcb\x9b\xc2n\xcfI\x9e\x14\xf0D\xf0!\x18DN\xcd\x97\x016\xa5'
            '\xef\xa06\xcb\x884\x8a\x03\x01\t\x08\x04\x1fef14operationhisted'
        ),
    )