Exceptions and Tracebacks

  • so far we’ve ignored situations where errors occurred, but real software needs to handle errors or unexpected conditions all the time

>>> value = ' Aquamarine Falcon '
>>> float( value )
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ValueError: could not convert string to float:  Aquamarine Falcon
  • when functions call other functions, the system creates a “stack” of “frames”, an uncaught error will, by default, print out a “traceback” of these frames

    • when something goes wrong, you use the traceback to help you find out where and what the problem was

    • in python the traceback is ordered from “top” to “bottom”, that is, the “frame” printed first in the traceback (“<stdin>” in the example below) is the “top level” caller

    • each frame is a function which was running (not yet complete) when the uncaught error was encountered

    • in python, the last line of the traceback is a string representation of the Exception which was raised, which generally attempts to be a useful description of what went wrong

>>> from functionarguments import *
>>> rows = split_rows( open('../sample_data.csv').read().splitlines()[1:] )
>>> first,second = extract_columns( rows, 1, -2 )
>>> first,second = extract_columns( rows, 30 )
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "functionarguments.py", line 15, in extract_columns
    result.append( extract_column( rows, column ))
  File "functionarguments.py", line 8, in extract_column
    result.append( row[column] )
IndexError: list index out of range
  • it is possible to catch these Exceptions in Python by using a special type of block around the code in which the exception may occur

>>> value = '  Aquamarine Falcon '
>>> try:
...     value = float( value )
... except ValueError as err:
...     value = value.strip()
>>> value
'Aquamarine Falcon'


We can catch multiple Exception types using except (ValueError,TypeError) as err instead.


The syntax for catching exceptions changes between Python 2.x and 3.x, in Python 3.x the syntax becomes except ValueError, TypeError as err


  • does your script fail if you point it at ../bad_sample_data.csv?

    • if not, congratulations; you pass

    • if so, what does the traceback tell you?

  • (if necessary) modify your moduleexercise.py so that it can parse ../bad_sample_data.csv as well as any file in the ../real_data/ directory

    • catch the case where the first column is a quoted, comma-separated name, convert the name to first last rather than last, first

    • assume that missing (numeric) values should be set equal to 0.0

    • assume that comments (lines starting with ‘#’) and blank lines should be ignored