Group values based on range of number in python

Question

I have a list as follows:

[(220921998, 2426),
(220921999, 2427),
(220922000, 2428),
(220922001, 2429),
(220922563, 2991),
(220922564, 2992),
(220922565, 2993),
(220922566, 2994),
(220922575, 3003),
(220923958, 4386),
(220924161, 4589),
(220924170, 4598),
(220924171, 4599),
(220924172, 4600),
(220924173, 4601),
(220924912, 5340),
(220926340, 6768),
(220926341, 6769),
(220926342, 6770),
(220926343, 6771),
(220926344, 6772),
(220927052, 7480),
(220927053, 7481),
(220927054, 7482),
(220927055, 7483),
(220927056, 7484),
(220927069, 7497),
(220927071, 7499)]

I want to add a string to the list based on the second number. If second number in the list are within 20 or so of other second numbers, they will be assigned same 'project' name. See below:

[(220921998, 2426,project1),
(220921999, 2427,project1),
(220922000, 2428,project1),
(220922001, 2429,project1),
(220922563, 2991,project2),
(220922564, 2992,project2),
(220922565, 2993,project2),
(220922566, 2994,project2),
(220922575, 3003,project3),
(220923958, 4386,project4),
(220924161, 4589,project5),
(220924170, 4598,project5),
(220924171, 4599,project5),
(220924172, 4600,project5),
(220924173, 4601,project5),
(220924912, 5340,project6),
(220926340, 6768,project7),
(220926341, 6769,project7),
(220926342, 6770,project7),
(220926343, 6771,project7),
(220926344, 6772,project7),
(220927052, 7480,project8),
(220927053, 7481,project8),
(220927054, 7482,project8),
(220927055, 7483,project8),
(220927056, 7484,project8),
(220927069, 7497,project8),
(220927071, 7499,project8)]

I have tried groupby, but couldn't find a way to work it for range.Any help would be great. Thank you

Answer 1

Use itertools.groupby with a key function that remember the last item and check it with current item.

lst = [(220921998, 2426),
       (220921999, 2427),
       (220922000, 2428),
       (220922001, 2429),
       (220922563, 2991),
       (220922564, 2992),
       (220922565, 2993),
       (220922566, 2994),
       (220922575, 3003),
       (220923958, 4386),
       (220924161, 4589),
       ....]

class Delta:
    def __init__(self, delta):
        self.last = None
        self.delta = delta
        self.key = 1
    def __call__(self, value):
        if self.last is not None and abs(self.last - value[1]) > self.delta:
            # Compare with the last value (`self.last`)
            # If difference is larger than 20, advance to next project
            self.key += 1
        self.last = value[1]  # Remeber the last value.
        return self.key

import itertools
for key, grp in itertools.groupby(lst, key=Delta(20)):
    for tup in grp:
        print(tup + ('project{}'.format(key),))

---

If you use Python 3.x, you can use the following function instead (See nonlocal):

def Delta(delta):
    last = None
    key = 1
    def keyfunc(value):
        nonlocal last, key
        if last is not None and abs(last - value[1]) > delta:
            key += 1
        last = value[1]
        return key
    return keyfunc

Answer 2

using https://pypi.python.org/pypi/cluster/1.1.0b1

>>> import cluster
>>> cl = cluster.HierarchicalClustering(data, lambda x,y: abs(x[1]-y[1]))
>>> cl.getlevel(20)
[
 [(220926340, 6768), (220926341, 6769), (220926344, 6772), (220926342, 6770), 
  (220926343, 6771)], 

 [(220927052, 7480), (220927053, 7481), (220927056, 7484), 
  (220927054, 7482), (220927055, 7483), (220927069, 7497), (220927071, 7499)], 

 [(220921998, 2426), (220921999, 2427), (220922000, 2428), (220922001, 2429)], 

 [(220922575, 3003), (220922563, 2991), (220922564, 2992), (220922565, 2993), 
  (220922566, 2994)], 

 [(220924912, 5340)], 

 [(220923958, 4386)], 

 [(220924161, 4589), (220924170, 4598), (220924171, 4599), (220924172, 4600), 
  (220924173, 4601)]
]