2015-04-17 21:29:30 -03:00

141 lines
3.5 KiB
Python

"""
The `+` operator produces a `Vector` result.
>>> v1 = Vector(2, 4)
>>> v2 = Vector(2, 1)
>>> v1 + v2
Vector(4, 5)
We can also implemement the `*` operator to perform scalar multiplication
or elementwise multiplication.
>>> v = Vector(3, 4)
>>> abs(v)
5.0
>>> v * 3
Vector(9, 12)
>>> abs(v * 3)
15.0
>>> v25 = Vector(2, 5)
>>> v71 = Vector(7, 1)
>>> v71 * v25
Vector(14, 5)
A vector can be used in a boolean context, where it will be considered
_falsy_ if it has magnitude zero, otherwise it is _truthy_::
>>> bool(v)
True
>>> bool(Vector(0, 0))
False
Vectors can have n-dimensions::
>>> v3 = Vector(1, 2, 3)
>>> len(v3)
3
>>> v3
Vector(1, 2, 3)
>>> abs(v3) # doctest:+ELLIPSIS
3.74165738...
>>> v3 + Vector(4, 5, 6)
Vector(5, 7, 9)
>>> v3 * 5
Vector(5, 10, 15)
>>> v2 + v3
Traceback (most recent call last):
...
ValueError: Addition applies only to vectors of equal dimensions.
The `repr` of a Vector is produced with the help of the `reprlib.repr`
function, limiting the size of the output string:
>>> Vector(*range(100))
Vector(0, 1, 2, 3, 4, 5, ...)
Dot product is a scalar: the sum of the products of the corresponding
components of two vectors.
>>> v25 = Vector(2, 5)
>>> v71 = Vector(7, 1)
>>> v25.dot(v71)
19
>>> Vector(1, 2, 3).dot(Vector(4, 5, 6))
32
>>> Vector(1, 2, 3).dot(Vector(-2, 0, 5))
13
As described in PEP 465, starting with Python 3.5, `__matmul__` is
the special method for the new ``@`` operator, to be used the dot
product of vectors or matrix multiplication (as opposed to ``*``
which is intended for scalar or elementwise multiplication):
>>> # skip these tests on Python < 3.5
>>> v25 @ v71 # doctest:+SKIP
19
>>> v71 * v25
Vector(14, 5)
>>> Vector(1, 2, 3) @ Vector(-2, 0, 5) # doctest:+SKIP
13
"""
# BEGIN VECTOR_OPS
import math
import numbers
import reprlib
EQ_DIMENSIONS_MSG = '%s applies only to vectors of equal dimensions.'
class Vector:
"""An n-dimensional vector"""
def __init__(self, *components): # <1>
self._components = tuple(components) # <2>
def __repr__(self):
return 'Vector' + (reprlib.repr(self._components)) # <3>
def __iter__(self):
return iter(self._components) # <4>
def __abs__(self):
return math.sqrt(sum(comp*comp for comp in self)) # <5>
def __len__(self):
return len(self._components) # <6>
def __add__(self, other):
if len(self) != len(other):
raise ValueError(EQ_DIMENSIONS_MSG % 'Addition')
return Vector(*(a+b for a, b in zip(self, other))) # <7>
def __mul__(self, other):
if isinstance(other, numbers.Number):
return Vector(*(comp*other for comp in self)) # <8>
else:
return self.elementwise_mul(other) # <9>
def elementwise_mul(self, other):
if len(self) != len(other):
raise ValueError(EQ_DIMENSIONS_MSG %
'Elementwise multiplication')
return Vector(*(a*b for a, b in zip(self, other))) # <10>
def __bool__(self):
return any(self) # <11>
def dot(self, other):
if len(self) != len(other):
raise ValueError(EQ_DIMENSIONS_MSG %
'Dot product')
return sum(a*b for a, b in zip(self, other)) # <12>
__matmul__ = dot # support @ operator in Python 3.5
# END VECTOR_OPS