Dan
2004-06-20 14:46:35 UTC
t9.in: 5#
t9.out: Invalid character at position 2. Expected 0-9A-Z
Since we know the range of valid characters in base 5 is 0-4, shouldn't
the error message say "Expected 0-4"?
And Troy replied that we shouldn't change the error messages to readt9.out: Invalid character at position 2. Expected 0-9A-Z
Since we know the range of valid characters in base 5 is 0-4, shouldn't
the error message say "Expected 0-4"?
"Expected 0-4" but t9.out now reads "Invalid character at position 2.
Expected 0-4.".
So I'm a little confused as to which is the correct behaviour at this point.
Is the test case wrong? Or have the requirements changed?
I also am curious about how error messages are going to be marked. As this
test case has (I presume) an error in it, are the error messages in the
other test cases valid? Is it possible in the future that all posted test
cases can output exactly what is expected from that input? We find it very
difficult to write code that will be automarked when we do not have proper
test cases.
I also have another question: how robust is the automarking? ie: If I
have two spaces somewhere in an error message and you are expecting one,
will that cause my output to be invalid? I'm sure there are other cases
where this could be an issue as well....
Thanks,
Dan