Benchmarking regular expressions with Python

In order to benchmark our regex, we're going to measure the time a regex takes to execute. It's important to test them with different inputs, because with small inputs almost every regex is fast enough. However, with longer ones it can be a completely different beast, as we will see later in the section Backtracking.

First, we're going to create a small function to help us with this task:

>>> from time import clock as now
>>> def test(f, *args, **kargs):
        start = now()
        f(*args, **kargs)
        print "The function %s lasted: %f" %(f.__name__, now() - start)

So, we can test a regex using the following code:

>>> def alternation(text):
       pat = re.compile('spa(in|niard)')
       pat.search(text)
>>> test(alternation, "spain")
The function alternation lasted: 0.000009

Python comes with a built-in profiler http://docs.python.org/2/library/profile.html that we can also use to measure the time and the...