如何合并元组列表

Question

I have two lists of tuples like this:

x1 = [('A', 3), ('B', 4), ('C', 5)]

x2 = [('B', 4), ('C', 5), ('D', 6)]

I want to merge the two lists as a new one x3 so that the values in the list are added.

x3 = [('A', 3), ('B', 8), ('C', 10),('D',6)]

Could you please show me how I can do this?

Answer 1

You can create a dictionary and then loop over the values in each list, and either adding to the current value for each key in the dictionary, or setting the value equal to the current value if no value currently exists. Afterwards you can cast back to a list.

For example:

full_dict = {}
for x in [x1, x2]:
    for key, value in x:
        full_dict[key] = full_dict.get(key, 0) + value # add to the current value, if none found then use 0 as current value

x3 = list(full_dict.items())

Result for x3 :

[('A', 3), ('B', 8), ('C', 10), ('D', 6)]

Answer 2

Since you tag pandas

df1,df2=pd.DataFrame(x1),pd.DataFrame(x2)
pd.concat([df1,df2]).groupby(0,as_index=False).sum().apply(tuple,1).tolist()
[('A', 3), ('B', 8), ('C', 10), ('D', 6)]

Answer 3

Use collections.Counter

>>> x1 = [('A', 3), ('B', 4), ('C', 5)]
>>> x2 = [('B', 4), ('C', 5), ('D', 6)]
>>> from collections import Counter
>>> c1 = Counter(dict(x1))
>>> c2 = Counter(dict(x2))
>>> c3 = c1 + c2   
>>> c3.items()
dict_items([('A', 3), ('B', 8), ('C', 10), ('D', 6)])

Just create a Counter for each list, you can do this by turning the list of tuples into an intermediate dict. Then just add the Counters.

To get the list back just call the items method of the counter. Technically what is returned is a dict_items object, if it really matters that its a list then just call list on c3.items() ie list(c3.items()) .

An alternative 1 liner using groupby and heapq.merge.

from heapq import merge
from itertools import groupby
from operator import itemgetter

x1 = [('A', 3), ('B', 4), ('C', 5)]
x2 = [('B', 4), ('C', 5), ('D', 6)]

result = [(a, sum(t[1] for t in b)) for a, b in groupby(merge(x1, x2), key=itemgetter(0))]
print(result)

heapq.merge is roughly equivalent to sorted(itertools.chain(*iterables)) . It basically provides a sorted, merged list of the tuples, ready to be fed to groupby.

Answer 4

I think the following code works

dict_x1=dict(x1)
for item in x2:
  if item[0] in dictx1:
    dictx1[item[0]] += item[1]
  else:
    dictx1[item[0] = item[1]

Answer 5

You can also solve this with itertools

from itertools import groupby
from operator import itemgetter

first = itemgetter(0)
second = itemgetter(1)
res = {}
data = x1 + x2

for _, g in groupby(sorted(data, key=first), key=first):
    group = list(g)
    k = first(first(group))
    v = sum(second(gr) for gr in group)
    res[k] = v

print(sorted(res.items(), key=first))

[('A', 3), ('B', 8), ('C', 10), ('D', 6)]