Knuth-Morris-Pratt (KMP) is a linear time string matching algorithm. The algorithm avoids unnecessary comparison and computation of the transition function by using prefix (Π) function . Before starting the actual comparison between the characters of pattern and text, it computes the Π (pi) values for each character of the pattern using **prefix function** algorithm. The code is based on the algorithm described in CLRS book, but it checks the multiple occurrences of the pattern along with mismatch.

```
def KMP_Matcher(t,p):
n = len (t)
m = len (p)
pi = calculate_PI(p)
k = 0
for i in xrange(n):
while (k > 0 and p[k] != t[i]):
k = pi[k]
if (p[k] == t[i]):
k = k+1
if (k == m):
print "pattern",p + " found at index ", i-m+1
k = 0
else:
if (i == n-1):
print "Pattern", p + " not found"
pat = "monkey"
str = "five little monkey jumping on the bed. one monkey fell down and now there are just four monkey"
KMP_Matcher(str,pat)
pat = "donkey"
str = "five little monkey jumping on the bed. one monkey fell down and now there are just four monkey"
KMP_Matcher(str,pat)
```

**The output of the code is as follows**…

pattern monkey found at index 12

pattern monkey found at index 43

pattern monkey found at index 88

Pattern donkey not found