Why re.search(r'(ab*)','aaAaABBbbb',re.I) in python gives result 'a' instead of 'ABBbbb' though 're.I' is used?

Question

In python, re.search() checks for a match anywhere in the string (this is what Perl does by default).

So why don't we get output as 'ABBbbb' in Ex(1) as we found in Ex(2) and Ex(3) below.

Ex(1)

>>> s=re.search(r'(ab*)','aaAaABBbbb',re.I)
>>> print s.group()
a

Ex(2)

>>> s=re.search(r'(ab.*)','aaAaABBbbb',re.I)
>>> print s.group()
ABBbbb

Ex(3)

>>> s=re.search(r'(ab+)','aaAaABBbbb',re.I)
>>> print s.group()
ABBbbb

Answer 1

Example 1 is searching for a followed by zero or more b, ignoring case. This matches right at the beginning of the string. The regex engine will see that match and use it. It won't look for any other matches.

Example 2 is searching for ab followed by as much of the string as it can eat. Example 3 is searching for a following by at least one b. The difference is that each of these needs at least one b, while Example 1 does not.

Answer 2

search:

checks for a match anywhere in the string (this is what Perl does by default).

---

re.search(r'(ab*)', 'aaAaABBbbb', re.I)

This will try to match a (case ignored) followed by zero or more b. It find that match in the first a, since it's followed by zero b, and it returned it.

re.search(r'(ab.*)', 'aaAaABBbbb', re.I)

This one will try to match a, followed by b and then with anything (.* is greedy). It matches ABBbbb because it's the first sequence that matches the regex.

re.search(r'(ab+)', 'aaAaABBbbb', re.I)

Finally, this will match a, followed by at least one b (again, case ignored). That first match is ABBbbb, and it's returned.