Difference betwen the [a-z][a-z]* and [a-z]+ regular expressions

Question

What is the difference between the following regular expressions. For me they are both the same

1. [a-z][a-z]* Vs [a-z]+
2. [a-z][a-z]* Vs [a-z]*[a-z]

Answer 1

These regexes are identical, as you thought.

#1:

[a-zA-Z]  # exactly one alphabetic char
[a-zA-Z]* # 0 to infinite alphabetic chars

versus

[a-zA-Z]+ # 1 to infinite alphabetic chars

One is just 1 + [0, \infinity] = [1, \infinity], the other [1, \infinity].

Further comments

#2 works similarly, all you're doing in each case is taking one example of the repeated character (in your case, [a-zA-Z], out of the repeated character command, * or +.

The answer below that points out that the more readable version is preferred is right on target. There is absolutely no reason to do something like [a-zA-Z]*[a-zA-Z] or [a-zA-Z][a-zA-Z]*, since ultimatley they're both just [a-zA-Z]+.

TL;DR

All are the same, and anytime you're repeating two identical commands in a row in a regex, you're doing something wrong.

Update:

$ python -m timeit -s "import re" "re.search(r'[a-zA-Z]*[a-
zA-Z]', '2323hfjfkf 23023493')"
1000000 loops, best of 3: 1.14 usec per loop

$ python -m timeit -s "import re" "re.search(r'[a-zA-Z]+',
'2323hfjfkf 23023493')"
1000000 loops, best of 3: 1 usec per loop

$ python -m timeit -s "import re" "re.search(r'[a-zA-Z][a-z
A-Z]*', '2323hfjfkf 23023493')"
1000000 loops, best of 3: 0.956 usec per loop

Turns out that [a-zA-Z][a-zA-Z]* is marginally faster than using [a-zA-Z]+. I'm a little surprised, but frankly I don't think the loss in readability is worth the .05 microsecond gain in efficiency.

Answer 2

Functionally all these regular expressions are identical.

Using the + quantifier, though, may be problematic in some cases, because depending on the parser and its settings it may or it may not need to be escaped (\+) in order to retain its special meaning. That is why some people avoid using + and prefer the more explicit XX* form, in order to keep their regular expressions more portable.

As far as Java is concerned, though, + always retains its special meaning, unless escaped.

Answer 3

Yes, all four are totally equal regular expressions. [a-z]+ is the simplest one and should be chosen for readability issues.

Answer 4

You're right that [a-zA-Z][a-zA-Z]* and [a-zA-Z]+ match all of the same strings so in that respect there's no difference. There's one main advantage [a-zA-Z]+ has over the other which is that it's more readable (readability counts!).

Answer 5

Both are the same check out Pattern Reluctant quantifiers. [a-zA-Z]+ is more readable for yourself and others.

Answer 6

[a-zA-Z][a-zA-Z]* Vs [a-zA-Z]*[a-zA-Z]

I think the main difference between this regular expression is that first expression will be done early than the second one. Because tree-walk for match for [a-zA-Z][a-zA-Z]* consist of steps less than another part of the expression.