How to search for a string in text files?

The reason why you always got True has already been given, so I'll just offer another suggestion:

If your file is not too large, you can read it into a string, and just use that (easier and often faster than reading and checking line per line):

with open('example.txt') as f:
    if 'blabla' in f.read():
        print("true")

Another trick: you can alleviate the possible memory problems by using mmap.mmap() to create a "string-like" object that uses the underlying file (instead of reading the whole file in memory):

import mmap

with open('example.txt') as f:
    s = mmap.mmap(f.fileno(), 0, access=mmap.ACCESS_READ)
    if s.find('blabla') != -1:
        print('true')

NOTE: in python 3, mmaps behave like bytearray objects rather than strings, so the subsequence you look for with find() has to be a bytes object rather than a string as well, eg. s.find(b'blabla'):

#!/usr/bin/env python3
import mmap

with open('example.txt', 'rb', 0) as file, \
     mmap.mmap(file.fileno(), 0, access=mmap.ACCESS_READ) as s:
    if s.find(b'blabla') != -1:
        print('true')

You could also use regular expressions on mmap e.g., case-insensitive search: if re.search(br'(?i)blabla', s):

Here's another way to possibly answer your question using the find function which gives you a literal numerical value of where something truly is

open('file', 'r').read().find('')

in find write the word you want to find and 'file' stands for your file name

As Jeffrey Said, you are not checking the value of check(). In addition, your check() function is not returning anything. Note the difference:

def check():
    with open('example.txt') as f:
        datafile = f.readlines()
    found = False  # This isn't really necessary
    for line in datafile:
        if blabla in line:
            # found = True # Not necessary
            return True
    return False  # Because you finished the search without finding

Then you can test the output of check():

if check():
    print('True')
else:
    print('False')