mirrors/cpython
1
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-12-11 03:20:01 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 10

Remove the random=None nonsense from sample() before it gets set in stone.

Browse source 
It was once available so that faster generators could be substituted.  Now,
that is less necessary and preferrably done via subclassing.

Also, clarified and shortened the comments for sample().
This commit is contained in:
Raymond Hettinger 2003-01-04 05:20:33 +00:00
parent 950cdacfa5
commit 8b9aa8dbba
1 changed files with 11 additions and 18 deletions
Download patch file Download diff file Expand all files Collapse all files

27
Lib/random.py
View file

@ -207,7 +207,7 @@ class Random(CoreGenerator):
j = int(random() * (i+1)) j = int(random() * (i+1))
x[i], x[j] = x[j], x[i] x[i], x[j] = x[j], x[i]
def sample(self, population, k, random=None, int=int): def sample(self, population, k, int=int):
"""Chooses k unique random elements from a population sequence. """Chooses k unique random elements from a population sequence.
Returns a new list containing elements from the population while Returns a new list containing elements from the population while
@ -223,29 +223,22 @@ class Random(CoreGenerator):
To choose a sample in a range of integers, use xrange as an argument. To choose a sample in a range of integers, use xrange as an argument.
This is especially fast and space efficient for sampling from a This is especially fast and space efficient for sampling from a
large population: sample(xrange(10000000), 60) large population: sample(xrange(10000000), 60)
Optional arg random is a 0-argument function returning a random
float in [0.0, 1.0); by default, the standard random.random.
""" """
# Sampling without replacement entails tracking either potential # Sampling without replacement entails tracking either potential
# selections (the pool) or previous selections. # selections (the pool) in a list or previous selections in a
# dictionary.
# Pools are stored in lists which provide __getitem__ for selection # When the number of selections is small compared to the population,
# and provide a way to remove selections. But each list.remove() # then tracking selections is efficient, requiring only a small
# rebuilds the entire list, so it is better to rearrange the list, # dictionary and an occasional reselection. For a larger number of
# placing non-selected elements at the head of the list. Tracking # selections, the pool tracking method is preferred since the list takes
# the selection pool is only space efficient with small populations. # less space than the dictionary and it doesn't suffer from frequent
# reselections.
# Previous selections are stored in dictionaries which provide
# __contains__ for detecting repeat selections. Discarding repeats
# is efficient unless most of the population has already been chosen.
# So, tracking selections is fast only with small sample sizes.
n = len(population) n = len(population)
if not 0 <= k <= n: if not 0 <= k <= n:
raise ValueError, "sample larger than population" raise ValueError, "sample larger than population"
if random is None:
random = self.random random = self.random
result = [None] * k result = [None] * k
if n < 6 * k: # if n len list takes less space than a k len dict if n < 6 * k: # if n len list takes less space than a k len dict
@ -253,7 +246,7 @@ class Random(CoreGenerator):
for i in xrange(k): # invariant: non-selected at [0,n-i) for i in xrange(k): # invariant: non-selected at [0,n-i)
j = int(random() * (n-i)) j = int(random() * (n-i))
result[i] = pool[j] result[i] = pool[j]
pool[j] = pool[n-i-1] pool[j] = pool[n-i-1] # move non-selected item into vacancy
else: else:
selected = {} selected = {}
for i in xrange(k): for i in xrange(k):