Fix the intermittent UnicodeDecodeError when the debug log is truncated on an (multi-byte) unicode character by treating everything as bytes.
Also, remove the ignore_case option and there.search+re.escape wrap. All of this is unused and doesn't exist on raw byte strings.
93 | if self.is_wallet_compiled(): 94 | - lines_to_terminate_after.append('Verifying wallet') 95 | + lines_to_terminate_after.append(b'Verifying wallet') 96 | 97 | for terminate_line in lines_to_terminate_after: 98 | self.log.info(f"Starting node and will exit after line '{terminate_line}'")
nit, maybe clean up this log by converting the bytes back to a string, or by dropping the single quotes
2022-06-07T16:30:25.742000Z TestFramework (INFO): Starting node and will exit after line 'b'Validating signatures for all blocks''
2022-06-07T16:30:26.322000Z TestFramework (INFO): Starting node and will exit after line 'b'scheduler thread start''
2022-06-07T16:30:26.663000Z TestFramework (INFO): Starting node and will exit after line 'b'Starting HTTP server''
2022-06-07T16:30:27.137000Z TestFramework (INFO): Starting node and will exit after line 'b'Loading P2P addresses''
2022-06-07T16:30:27.558000Z TestFramework (INFO): Starting node and will exit after line 'b'Loading banlist''
2022-06-07T16:30:27.986000Z TestFramework (INFO): Starting node and will exit after line 'b'Loading block index''
2022-06-07T16:30:28.460000Z TestFramework (INFO): Starting node and will exit after line 'b'Switching active chainstate
...
thx, fixed
ACK
ACK fa74b63c01db412f6a4378cb669d89496a89d02e
ACK fa74b63c01db412f6a4378cb669d89496a89d02e
@MarcoFalke I think line 451 of test_node.py is now broken. log is now no longer a string, but bytes.
if time.time() >= time_end:
print_log = " - " + "\n - ".join(log.splitlines())
break
mypy detected this when I tried adding a typehint (List[bytes]) to the expected_msgs parameter of wait_for_debug_log.
This could work
if time.time() >= time_end:
print_log = " - " + "\n - ".join([f"\n - {line.decode()}" for line in log.splitlines()])
break