Python: POSIX character class in regex?

Question

How can I search for, say, a sequence of 10 isprint characters in a given string in Python?

With GNU grep, I would simply do grep [[:print:]]{10}

score 12 · Accepted Answer · answered Aug 10 '15 at 09:01

12

Since POSIX is not supported by Python re module, you have to emulate it with the help of character class.

You can use the one from the regular-expressions.info and add a limiting quantifier {10}:

[\x20-\x7E]{10}

Alternatively, you can use Matthew Barnett regex module that claims to support POSIX character classes (POSIX character classes are supported.).

answered Aug 10 '15 at 09:01

Wiktor Stribiżew

This character class worked for me in Python 3 `[\`~!@#$%^&*()_=+\[\]{}\\\|;:\"\'<>.,/?]` when using inside the `re.sub()` method – Allan Tsai Feb 02 '20 at 13:20
1

@Iota, that ``[`~!@#$%^&*()_=+\[\]{}\\\|;:\"\'<>.,/?]`` only matches ASCII punctuation, it has nothing to do with the concept of "printable chars". So, if you were to use a POSIX character class, it would be `[[:punct:]]`. To match punctuation in Python, you can use `[^\w\s]`, although there are better and more precise patterns. – Wiktor Stribiżew Feb 02 '20 at 13:25
1

My mistake! I misread the `[[:print]]` class as `[[:punct]]`. Appreciate your correction. – Allan Tsai Feb 04 '20 at 06:19
The regex in the answer will not match Unicode non-ASCII characters like grep (GNU grep) does. – pabouk - Ukraine stay strong Apr 05 '22 at 13:41
2

@pabouk-Ukrainestaystrong Then see the bottom of the answer. Just install the PyPi regex module (`pip install regex` in the console/terminal) and then use `import regex` and `pattern = regex.compile(r'[[:print:]]{10}')`. – Wiktor Stribiżew Apr 05 '22 at 13:49

1 Answers1