'Differ','IS_CHARACTER_JUNK', 'IS_LINE_JUNK', 'context_diff',
'unified_diff', 'HtmlDiff', 'Match']
-import heapq
+from heapq import nlargest as _nlargest
from collections import namedtuple as _namedtuple
Match = _namedtuple('Match', 'a b size')
result.append((s.ratio(), x))
# Move the best scorers to head of list
- result = heapq.nlargest(n, result)
+ result = _nlargest(n, result)
# Strip scores for the best n matches
return [x for score, x in result]