Skip to content

Instantly share code, notes, and snippets.

Created September 8, 2015 21:32
Show Gist options
  • Save anonymous/490c74a6473425fdaa88 to your computer and use it in GitHub Desktop.
Save anonymous/490c74a6473425fdaa88 to your computer and use it in GitHub Desktop.
Python coding task: "iter_sample" @http://codepad.org/9NdtMCJh
"""
Programming task
================
Implement the method iter_sample below to make the Unit test pass. iter_sample
is supposed to peek at the first n elements of an iterator, and determine the
minimum and maximum values (using their comparison operators) found in that
sample. To make it more interesting, the method is supposed to return an
iterator which will return the same exact elements that the original one would
have yielded, i.e. the first n elements can't be missing.
You may make use of Python's standard library. Python 3 is allowed, even though
it's not supported by codepad apparently.
Create your solution as a private fork, and send us the URL.
"""
from itertools import count
import unittest
def iter_sample(it, n):
"""
Peek at the first n elements of an iterator, and determine the min and max
values. Preserve all elements in the iterator!
@param it: Iterator, potentially infinite
@param n: Number of elements to peek off the iterator
@return: Tuple of minimum, maximum (in sample), and an iterator that yields
all elements that would have been yielded by the original iterator.
"""
if n == 0:
return None, None, None
count = n
iter_elements = []
iter_elements.append(next(it)) #store elements from the iterator for later use
min_val = iter_elements.index(0) #set it to the first value
max_val = iter_elements.index(0) #set it to the first value
for value in it:
iter_elements.append(value)
if min_val > value:
min_val = value
elif value > max_val:
max_val = value
count = count - 1
if count == 1:
break
it = (val for gen in (iter_elements, it) for val in gen) #flattening of 2 lists helps us to maintain the iterator
return min_val, max_val, it
class StreamSampleTestCase(unittest.TestCase):
def test_smoke(self):
# sample only the first 10 elements of a range of length 100
it = iter(range(100))
min_val, max_val, new_it = iter_sample(it, 10)
self.assertEqual(0, min_val)
self.assertEqual(9, max_val)
# all elements are still there:
self.assertEqual(list(range(100)), list(new_it))
def test_sample_all(self):
# sample more elements than there are - no error raised
# now we now the global maximum!
it = iter(range(100))
min_val, max_val, new_it = iter_sample(it, 1000)
self.assertEqual(0, min_val)
self.assertEqual(99, max_val)
self.assertEqual(list(range(100)), list(new_it))
def test_infinite_stream(self):
# and guess what - it also works with infinite iterators
it = count(0)
min_val, max_val, _ = iter_sample(it, 10)
self.assertEqual(0, min_val)
self.assertEqual(9, max_val)
if __name__ == "__main__":
unittest.main()
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment