# HG changeset patch # User Henry S. Thompson # Date 1744227749 -3600 # Node ID a3d55cc7da184f086b7e8af2d4a4ba6fb7e9100b # Parent 70da637d1402bcab97e134f9cad55b5183ac5921 fix another long-tail bug diff -r 70da637d1402 -r a3d55cc7da18 lib/python/cc/warc.py --- a/lib/python/cc/warc.py Wed Apr 09 17:15:40 2025 +0100 +++ b/lib/python/cc/warc.py Wed Apr 09 20:42:29 2025 +0100 @@ -54,9 +54,10 @@ done: bool = bl < BUFSIZE while buf.startswith(b'\r\n',bp): bp+=2 - while not (done and bl == bp): + while not (done and bp <= bl): start_1: int = bp if not buf.startswith(b'WARC/1.0\r\n',bp): + breakpoint() raise ValueError("Not a WARC file? In %s at %s of %s (%s): %s[%s]"%(filename, bp,bl,fpos, (buf[bp:min(bl,bp+20)] if bp