python - Reduce list of lists if entries match -

i have list in python looks like

[['boy','121','is male child'],['boy','121','is male'],['boy','121','is child'],['girl','122','is female child'],['girl','122','is child']] 

i want reduce list based on first 2 entries in each list, get

[['boy','121',is male child, male, child'],['girl','122','is female child','is child']] 

is there way efficiently without creating dummy list?

as more pythonic way such task can use dictionary :

>>> li=[['boy','121','is male child'],['boy','121','is male'],['boy','121','is child'],['girl','122','is female child'],['girl','122','is child']] >>>  >>> d={} >>>  >>> i,j,k in li: ...   d.setdefault((i,j),[]).append(k) ...  >>> d {('boy', '121'): ['is male child', 'is male', 'is child'], ('girl', '122'): ['is female child', 'is child']} 

setdefault(key[, default])

if key in dictionary, return value. if not, insert key value of default , return default. default defaults none.

and if want have elements within 1 container can loop on items , convert value tuple ad key:

>>> [i+tuple(j) i,j in d.items()] [('boy', '121', 'is male child', 'is male', 'is child'), ('girl', '122', 'is female child', 'is child')] 

as @jonrsharpe says more elegant way can use collections.defaultdict :

>>> collections import defaultdict >>>  >>> d=defaultdict(list) >>> i,j,k in li: ...   d[i,j].append(k) ...  >>> d defaultdict(<type 'list'>, {('boy', '121'): ['is male child', 'is male', 'is child'], ('girl', '122'): ['is female child', 'is child']})