def range_closest(ranges, b, left=True):
"""
Returns the range that's closest to the given position. Notice that the
behavior is to return ONE closest range to the left end (if left is True).
This is a SLOW method.
>>> ranges = [("1", 30, 40), ("1", 33, 35), ("1", 10, 20)]
>>> b = ("1", 22, 25)
>>> range_closest(ranges, b)
('1', 10, 20)
>>> range_closest(ranges, b, left=False)
('1', 33, 35)
>>> b = ("1", 2, 5)
>>> range_closest(ranges, b)
"""
from jcvi.utils.orderedcollections import SortedCollection
key = (lambda x: x) if left else (lambda x: (x[0], x[2], x[1]))
rr = SortedCollection(ranges, key=key)
try:
if left:
s = rr.find_le(b)
assert key(s) <= key(b), (s, b)
else:
s = rr.find_ge(b)
assert key(s) >= key(b), (s, b)
except ValueError:
s = None
return s
def range_closest(ranges, b, left=True):
"""
Returns the range that's closest to the given position. Notice that the
behavior is to return ONE closest range to the left end (if left is True).
This is a SLOW method.
>>> ranges = [("1", 30, 40), ("1", 33, 35), ("1", 10, 20)]
>>> b = ("1", 22, 25)
>>> range_closest(ranges, b)
('1', 10, 20)
>>> range_closest(ranges, b, left=False)
('1', 33, 35)
>>> b = ("1", 2, 5)
>>> range_closest(ranges, b)
"""
from jcvi.utils.orderedcollections import SortedCollection
key = (lambda x: x) if left else (lambda x: (x[0], x[2], x[1]))
rr = SortedCollection(ranges, key=key)
try:
if left:
s = rr.find_le(b)
assert key(s) <= key(b), (s, b)
else:
s = rr.find_ge(b)
assert key(s) >= key(b), (s, b)
except ValueError:
s = None
return s