def shortest_term_span(positions):
"""
Given a list of positions in a corpus, returns
the shortest span of words that contain all query terms.
"""
# Initialize our list of lists where each list corresponds
# to the locations within the document for a term
indices = [0] * len(positions)
min_window = window = get_window(positions, indices)
# Iteratively moving the minimum index forward finds us our
# minimum span
while True:
min_index = get_min_index(positions, window)
if min_index is None:
return sorted(min_window)
indices[min_index] += 1
window = get_window(positions, indices)
if list_range(min_window) > list_range(window):
min_window = window
if list_range(min_window) == len(positions):
return sorted(min_window)
def test_get_window(self):
self.assertEquals(
get_window([[1, 2, 3], [4, 5, 6], [7, 8, 9]], [0, 1, 2]),
[1, 5, 9])
def test_get_window(self):
self.assertEquals(
get_window([[1, 2, 3], [4, 5, 6], [7, 8, 9]],
[0, 1, 2]), [1, 5, 9])
