Searching a text file to find words matching a pattern

[snippsat]

It's in Swedish so there is of course åäö in the file and that's a problem as well.

A lot have change regard Unicode,it was one the biggest changes moving to Python 3(as mention bye @heiner55 you should use Python 3).
In Python 3 open() has build in encoding parameter.
So the simple rule is to keep it UTF-8 in and out when reading a file.
Inside Python 3 is all strings sequences of Unicode character,if not encode in or Python 3 do not not recognize encoding it will be bytes (b'hello').
Python 3 will not guess as Python 2 do.

So if borrow code from @heiner55 it look like this:

import re

with open('ss.txt', encoding='utf-8') as f:
    for line in f:
        line = line.strip()
        if re.match(r"h...g..", line) and len(line)==7:
            print(line)

There is no need for # -*- coding: utf-8 -*- in Python 3,because UTF-8 is default.

In Python 2 it would look like this,same rule UTF-8 in and out.
But has to use a library io or codecs and # -*- coding: utf-8 -*- because Python 2 has ASCII default encoding.

# -*- coding: utf-8 -*-
import re
import io

with io.open('ss.txt', encoding='utf-8') as f:
    for line in f:
        line = line.strip()
        if re.match(r"h...g..", line) and len(line)==7:
            print(line)