I want to create a number of lists from a single list. The lists will either have 1 value or 3 based on if there is data from a user. However, since the data is attached to a specific user, the order must be maintained.
For example:
data = ['no data', 'choice 1', 'choice 4', 'choice 2', 'no data', 'choice 1', ...etc]
I want the output to be:
list1 = ['no data']
list2 = ['choice 1', 'choice4', 'choice 2']
list3 = ['no data']
list4 = ['choice 1'...]
How would I achieve this?
Unfortunately, itertools.groupby won't work if you need to chunk runs of non-'no-data' into groups of 3, or as singltons for each individual 'no-data'. So, here's a something else thrown together:
In [40]: def group_lists(data):
...: final = []
...: contains_data = False
...: temp = []
...: for sub in data:
...: if sub == 'no data':
...: if contains_data:
...: final.append(temp)
...: temp = []
...: final.append([sub])
...: contains_data = False
...: else:
...: final.append([sub])
...: else:
...: contains_data = True
...: if len(temp) < 3:
...: temp.append(sub)
...: else:
...: final.append(temp)
...: temp = []
...: contains_data = False
...: return final
...:
In [41]: data = ['choice 1', 'choice 2', 'choice 1', 'choice 1', 'choice 4', 'choice 1', 'no data']
In [42]: group_lists(data)
Out[42]: [['choice 1', 'choice 2', 'choice 1'], ['choice 4', 'choice 1'], ['no data']]
In [43]: data2 = ['choice 1', 'choice 2', 'choice 1', 'choice 1', 'choice 4', 'choice 1', 'choice1', 'no data']
In [44]: group_lists(data2)
Out[44]:
[['choice 1', 'choice 2', 'choice 1'],
['choice 4', 'choice 1', 'choice1'],
['no data']]
Use itertools.groupby:
In [15]: data = ['no data', 'choice 1', 'choice 4', 'choice 2', 'no data', 'choice 1']
...:
In [16]: import itertools
In [17]: grouped = [list(g) for _, g in itertools.groupby(data, lambda s: s == 'no data')]
In [18]: grouped[0]
Out[18]: ['no data']
In [19]: grouped[1]
Out[19]: ['choice 1', 'choice 4', 'choice 2']
In [20]: grouped[2]
Out[20]: ['no data']
In [21]: grouped[3]
Out[21]: ['choice 1']
Unpacking that rather unwieldy one-liner:
In [26]: lists = []
...: for _, g in itertools.groupby(data, lambda s: s == 'no data'):
...: lists.append(list(g))
...:
In [27]: lists[0]
Out[27]: ['no data']
In [28]: lists[1]
Out[28]: ['choice 1', 'choice 4', 'choice 2']
In [29]: lists[2]
Out[29]: ['no data']
In [30]: lists[3]
Out[30]: ['choice 1']
So, the result is a lists of lists:
In [32]: lists
Out[32]: [['no data'], ['choice 1', 'choice 4', 'choice 2'], ['no data'], ['choice 1']]
This seems to work and is fairly easy to understand. It doesn't create separate variables, but instead creates a list-of-lists named lists.
data = ['no data', 'choice 1', 'choice 4', 'choice 2', 'no data', 'choice 1', '...etc']
lists = []
tmp = []
for elem in data:
if elem != 'no data':
tmp.append(elem)
else:
if tmp:
lists.append(tmp)
lists.append([elem])
tmp = []
if tmp:
lists.append(tmp)
for i, sublist in enumerate(lists, start=1):
print('list{}: {}'.format(i, sublist))
Output:
list1: ['no data']
list2: ['choice 1', 'choice 4', 'choice 2']
list3: ['no data']
list4: ['choice 1', '...etc']