Question

I’m currently stuck on trying to make a naive algorithm which given a piece of a pattern e.g aabba
search for it in a text e.g abbbbaababaabbaaabbaa one letter at a time. It will compare a with the text if that is right then compares the next letter and if that’s wrong the whole pattern will shift one and compare a with b etc

we were give code example

print "Input text: ",
text = raw_input()
print "Input pattern: ",
pattern = raw_input()

index = text.find(pattern)
while index > -1:
    print index
    index = text.find(pattern, index+1)

but the find() function in python is too fast(I need a non optimized sort of algorithm, using while
and for loops statements I guess).

Any help appreciated,
Thanks

Answer 1

I guess here’s what you need, the following code does character-by-character comparison. You may also replace the calls to find by iterations over text which includes checks whether the first character of text matches the first character of pattern:

def my_find(text, pattern):
    '''Find the start index of a pattern string in a text.
    Return -1 if not found, and assume that pattern is not empty'''

    found = False
    current_start_index = text.find(pattern[0])
    index_text = current_start_index
    index_pattern = 0

    while not found and index_text + len(pattern) - 1 < len(text) and \
            current_start_index != -1:

        index_text += 1
        index_pattern += 1

        while index_text < len(text) and \
                index_pattern < len(pattern) and \
                text[index_text] == pattern[index_pattern]:

            if index_pattern == len(pattern) - 1:
                found = True
                break
            else:
                index_text += 1
                index_pattern += 1

        if not found:
            current_start_index = text.find(pattern[0],current_start_index + 1)
            index_text = current_start_index

    if found:
        return current_start_index
    else:
        -1