Reputation:
Assuming there is a list with sublists like this
[[2013, 'Patric', 'M', 1356], [2013, 'Helena', 'F', 202], [2013, 'Patric', 'F', 6],[1993, 'Patric', 'F', 7]......]
which is an output of def list_of_names()
where 2013 is year, M is gender and 1356 is number of M births etc.
And I want to create a dictionary which outputs the name as a key and values as tuples (year, number_of_males,number_of_females) . So for example:
{ .. ’Patric’:[... , (1993, 0, 7), (2013, 1356, 6), ... ], ... }.
Technically 1993 is year, 0 is number of males and 7 is number of females and the tuples should be arranged in order of the years.
and I'm stuck on how to add this info into a dictionary
def name_Index(names):
d = dict()
L = readNames() #the list with from previous def which outputs different names and info as above
newlist = []
for sublist in L:
Upvotes: 1
Views: 100
Reputation: 135
I didn't understand why you are taking names as an argument of name_Index function and then calling readNames, there must be some necessity required for your work. Hence, i just put a dummy readNames function and sent None as argument to name_Index. Using class is a good technique to solve complicated data structures. Btw, nicely written question i must admit.
def readNames ():
return [[2013, 'Patric', 'M', 1356], [2013, 'Helena', 'F', 202], [2013, 'Patric', 'F', 6],[1993, 'Patric', 'F', 7]]
class YearOb(object):
def __init__(self):
self.male = 0
self.female = 0
def add_birth_data(self, gender, birth_count):
if gender == "M":
self.male += birth_count
else:
self.female += birth_count
class NameOb(object):
def __init__(self):
self.yearobs = dict()
def add_record(self, year, gender, birth_count):
if year not in self.yearobs:
self.yearobs[year]=YearOb()
self.yearobs[year].add_birth_data(gender, birth_count)
def get_as_list(self):
list_data = []
for year, yearob in self.yearobs.items():
list_data.append((year, yearob.male, yearob.female))
return list_data
def name_Index(names):
d = dict()
L = readNames() #the list with from previous def which outputs different names and info as above
newlist = []
for sublist in L:
name = sublist[1]
if name not in d:
d[name]=NameOb()
d[name].add_record(sublist[0], sublist[2], sublist[3])
for name, nameob in d.items():
d[name] = nameob.get_as_list()
return d
print(name_Index(None))
Upvotes: 0
Reputation: 6387
The most convenient will be to use collections.defauldict
. It returns dictionary-like object, that returns default value, if it doesn't find key. In your case, you use a list
as default value, and in your loop you append tuples to it:
from collections import defaultdict
names = [ [2013, 'Patric', 'M', 1356],
[2013, 'Helena', 'F', 202],
[2013, 'Patric', 'F', 6],
[1993, 'Patric', 'F', 7] ]
def name_Index(data):
# name => year => sex
d = defaultdict(lambda: defaultdict(lambda: {'F': 0, 'M': 0}))
for year, name, sex, births in data:
d[name][year][sex] += births
# if you are fine with defauldict result: return d
# else collect results into tuples:
result = {}
for name, data in d.items():
result[name] = [(year, c['M'], c['F']) for year, c in data.items()]
return result
print name_Index(names)
# {'Helena': [(2013, 0, 202)], 'Patric': [(1993, 0, 7), (2013, 1356, 6)]}
Upvotes: 0
Reputation: 82470
This was my attempt at the problem:
from collections import defaultdict, namedtuple
from itertools import groupby
data = [[2013, 'Patric', 'M', 1356],
[2013, 'Helena', 'F', 202],
[2013, 'Patric', 'F', 6],
[1993, 'Patric', 'F', 7]]
names = defaultdict(list)
datum = namedtuple('datum', 'year gender number')
for k, g in groupby(data, key=lambda x: x[1]):
for l in g:
year, name, gender, number = l
names[k].append(datum(year, gender, number))
final_dict = defaultdict(list)
for n in names:
for k, g in groupby(names[n], lambda x: x.year):
males = 0
females = 0
for l in g:
if l.gender == 'M':
males += l.number
elif l.gender == 'F':
females += l.number
final_dict[n].append((k, males, females))
print(final_dict)
Upvotes: 1
Reputation: 2120
from collections import defaultdict
def list_of_names():
return [[2013, 'Patric', 'M', 1356],
[2013, 'Helena', 'F', 202],
[2013, 'Patric', 'F', 6],
[1993, 'Patric', 'F', 7]]
def name_Index():
tmp = defaultdict(lambda:defaultdict(lambda: [0,0]))
for year, name, sex, N in list_of_names():
i = 0 if sex == 'M' else 1
tmp[name][year][i] += N
d = {}
for name, entries in tmp.items():
d[name] = [(year, M, F) for (year, (M,F)) in entries.items()]
return d
print name_Index()
Upvotes: 1