On Mon, 03 Dec 2012 12:37:42 -0500, wrw wrote: > So far in my experience with Python, it's error messages have been > clear, concise, and quite good at fingering my errors. However, the > message below has me stumped. The routine in question has been running > for weeks with no problems, then yesterday I got the following: > > Traceback (most recent call last): > File "./Connection_Monitor.py", line 146, in <module> > Google_up, Google_summary, Google_RTT, Google_stddev = Google.connection_test() > File "/Users/wrw/Dev/Python/Connection_Monitor/Version2.2/WorkingCopy/ network.py", > line 101, in connection_test > # > IndexError: list index out of range
Are you running Python with the -x option? If so, then I understand that the line number (and subsequent line of code) reported in tracebacks may sometimes be off by one. If not, or if the discrepancy is more than one line, I would start investigating the following scenario: * there is a mismatch between the source code .py file and the byte-code .pyc file; * but the .pyc file's "last modified date" is newer than the .py file; * the .pyc file is executed (as is usual); * but when an error occurs and Python looks up the source file, it gets something which no longer corresponds to the byte-code being executed. By the way, in connection_test you have the following: > def connection_test(self): > found_0 = '0 packets received' in ping_result > found_1 = '1 packets received' in ping_result > if found_0 == True or found_1 == True: [...] Of course that's not correct. You should write: if (found_0 == True or found_1 == True) == True: No wait, I mean: if ((found_0 == True or found_1 == True) == True) == True: No wait! I meant this: if ((found_0 == True == True == True == True == ... # Help! or found_1 == True == True == True == True == ... # Where do I stop? ) == True) == True == True == True == True ... : *wink* Of course the whole thing is silly. The right way to test booleans for their boolean value is just: if found_0 or found_1: "flag == True" is the same as "flag". -- Steven -- http://mail.python.org/mailman/listinfo/python-list
