N
Nils Grimsmo
i always have trouble explaining why creation of multidimensional lists
is not as straight forward as it could be in python. list comprehensions
are ugly if you are new to the language. i really would like to see it
made easy. i propose using tuples in the same way as integers are now.
example:
0], [0, 0, 0, 0]]]
this is even less "mathematically bad" syntax than multiplication with
an integer, as multiplication of vectors is only well defined if they
are of the same length. [1] * 7 could be interpreted as a vector of
length one multiplied with a scalar 7 resulting in the vector [7]. (this
was not meant as a proposition to remove the current syntax.)
i would be easy to implement:
class mylist(list):
def __mul__(self, dims):
if dims.__class__ == (()).__class__:
if len(dims) > 1:
return [self.__mul__(dims[1:]) for i in range(dims[0])]
else:
return list.__mul__(self, dims[0])
else:
return list.__mul__(self, dims)
li = mylist()
li.append(0)
print li * (2,3,4)
what do you think?
klem fra nils
is not as straight forward as it could be in python. list comprehensions
are ugly if you are new to the language. i really would like to see it
made easy. i propose using tuples in the same way as integers are now.
example:
[[[0, 0, 0, 0], [0, 0, 0, 0], [0, 0, 0, 0]], [[0, 0, 0, 0], [0, 0, 0,>>> [0] * (2,3,4)
0], [0, 0, 0, 0]]]
this is even less "mathematically bad" syntax than multiplication with
an integer, as multiplication of vectors is only well defined if they
are of the same length. [1] * 7 could be interpreted as a vector of
length one multiplied with a scalar 7 resulting in the vector [7]. (this
was not meant as a proposition to remove the current syntax.)
i would be easy to implement:
class mylist(list):
def __mul__(self, dims):
if dims.__class__ == (()).__class__:
if len(dims) > 1:
return [self.__mul__(dims[1:]) for i in range(dims[0])]
else:
return list.__mul__(self, dims[0])
else:
return list.__mul__(self, dims)
li = mylist()
li.append(0)
print li * (2,3,4)
what do you think?
klem fra nils