Finding all possible permutations of a given strin

Question

I have a string. I want to generate all permutations from that string, by changing the order of characters in it. For example, say:

x='stack'

what I want is a list like this,

l=['stack','satck','sackt'.......]

Currently I am iterating on the list cast of the string, picking 2 letters randomly and transposing them to form a new string, and adding it to set cast of l. Based on the length of the string, I am calculating the number of permutations possible and continuing iterations till set size reaches the limit. There must be a better way to do this.

Answer 1

Here's a slightly improved version of illerucis's code for returning a list of all permutations of a string s with distinct characters (not necessarily in lexicographic sort order), without using itertools:

def get_perms(s, i=0):
    """
    Returns a list of all (len(s) - i)! permutations t of s where t[:i] = s[:i].
    """
    # To avoid memory allocations for intermediate strings, use a list of chars.
    if isinstance(s, str):
        s = list(s)

    # Base Case: 0! = 1! = 1.
    # Store the only permutation as an immutable string, not a mutable list.
    if i >= len(s) - 1:
        return ["".join(s)]

    # Inductive Step: (len(s) - i)! = (len(s) - i) * (len(s) - i - 1)!
    # Swap in each suffix character to be at the beginning of the suffix.
    perms = get_perms(s, i + 1)
    for j in range(i + 1, len(s)):
        s[i], s[j] = s[j], s[i]
        perms.extend(get_perms(s, i + 1))
        s[i], s[j] = s[j], s[i]
    return perms

Answer 2

Here's a simple and straightforward recursive implementation;

def stringPermutations(s):
    if len(s) < 2:
        yield s
        return
    for pos in range(0, len(s)):
        char = s[pos]
        permForRemaining = list(stringPermutations(s[0:pos] + s[pos+1:]))
        for perm in permForRemaining:
            yield char + perm

Answer 3

The itertools module has a useful method called permutations(). The documentation says:

itertools.permutations(iterable[, r])

Return successive r length permutations of elements in the iterable.

If r is not specified or is None, then r defaults to the length of the iterable and all possible full-length permutations are generated.

Permutations are emitted in lexicographic sort order. So, if the input iterable is sorted, the permutation tuples will be produced in sorted order.

You'll have to join your permuted letters as strings though.

>>> from itertools import permutations
>>> perms = [''.join(p) for p in permutations('stack')]
>>> perms

['stack', 'stakc', 'stcak', 'stcka', 'stkac', 'stkca', 'satck', 'satkc', 'sactk', 'sackt', 'saktc', 'sakct', 'sctak', 'sctka', 'scatk', 'scakt', 'sckta', 'sckat', 'sktac', 'sktca', 'skatc', 'skact', 'skcta', 'skcat', 'tsack', 'tsakc', 'tscak', 'tscka', 'tskac', 'tskca', 'tasck', 'taskc', 'tacsk', 'tacks', 'taksc', 'takcs', 'tcsak', 'tcska', 'tcask', 'tcaks', 'tcksa', 'tckas', 'tksac', 'tksca', 'tkasc', 'tkacs', 'tkcsa', 'tkcas', 'astck', 'astkc', 'asctk', 'asckt', 'asktc', 'askct', 'atsck', 'atskc', 'atcsk', 'atcks', 'atksc', 'atkcs', 'acstk', 'acskt', 'actsk', 'actks', 'ackst', 'ackts', 'akstc', 'aksct', 'aktsc', 'aktcs', 'akcst', 'akcts', 'cstak', 'cstka', 'csatk', 'csakt', 'cskta', 'cskat', 'ctsak', 'ctska', 'ctask', 'ctaks', 'ctksa', 'ctkas', 'castk', 'caskt', 'catsk', 'catks', 'cakst', 'cakts', 'cksta', 'cksat', 'cktsa', 'cktas', 'ckast', 'ckats', 'kstac', 'kstca', 'ksatc', 'ksact', 'kscta', 'kscat', 'ktsac', 'ktsca', 'ktasc', 'ktacs', 'ktcsa', 'ktcas', 'kastc', 'kasct', 'katsc', 'katcs', 'kacst', 'kacts', 'kcsta', 'kcsat', 'kctsa', 'kctas', 'kcast', 'kcats']

If you find yourself troubled by duplicates, try fitting your data into a structure with no duplicates like a set:

>>> perms = [''.join(p) for p in permutations('stacks')]
>>> len(perms)
720
>>> len(set(perms))
360

Thanks to @pst for pointing out that this is not what we'd traditionally think of as a type cast, but more of a call to the set() constructor.

Answer 4

You can get all N! permutations without much code

def permutations(string, step = 0):

    # if we've gotten to the end, print the permutation
    if step == len(string):
        print "".join(string)

    # everything to the right of step has not been swapped yet
    for i in range(step, len(string)):

        # copy the string (store as array)
        string_copy = [character for character in string]

        # swap the current index with the step
        string_copy[step], string_copy[i] = string_copy[i], string_copy[step]

        # recurse on the portion of the string that has not been swapped yet (now it's index will begin with step + 1)
        permutations(string_copy, step + 1)

Answer 5

This program does not eliminate the duplicates, but I think it is one of the most efficient approaches:

s=raw_input("Enter a string: ")
print "Permutations :\n",s
size=len(s)
lis=list(range(0,size))
while(True):
    k=-1
    while(k>-size and lis[k-1]>lis[k]):
        k-=1
    if k>-size:
        p=sorted(lis[k-1:])
        e=p[p.index(lis[k-1])+1]
        lis.insert(k-1,'A')
        lis.remove(e)
        lis[lis.index('A')]=e
        lis[k:]=sorted(lis[k:])
        list2=[]
        for k in lis:
                list2.append(s[k])
        print "".join(list2)
    else:
                break

Answer 6

itertools.permutations is good, but it doesn't deal nicely with sequences that contain repeated elements. That's because internally it permutes the sequence indices and is oblivious to the sequence item values.

Sure, it's possible to filter the output of itertools.permutations through a set to eliminate the duplicates, but it still wastes time generating those duplicates, and if there are several repeated elements in the base sequence there will be lots of duplicates. Also, using a collection to hold the results wastes RAM, negating the benefit of using an iterator in the first place.

Fortunately, there are more efficient approaches. The code below uses the algorithm of the 14th century Indian mathematician Narayana Pandita, which can be found in the Wikipedia article on Permutation. This ancient algorithm is still one of the fastest known ways to generate permutations in order, and it is quite robust, in that it properly handles permutations that contain repeated elements.

def lexico_permute_string(s):
    ''' Generate all permutations in lexicographic order of string `s`

        This algorithm, due to Narayana Pandita, is from
        https://en.wikipedia.org/wiki/Permutation#Generation_in_lexicographic_order

        To produce the next permutation in lexicographic order of sequence `a`

        1. Find the largest index j such that a[j] < a[j + 1]. If no such index exists, 
        the permutation is the last permutation.
        2. Find the largest index k greater than j such that a[j] < a[k].
        3. Swap the value of a[j] with that of a[k].
        4. Reverse the sequence from a[j + 1] up to and including the final element a[n].
    '''

    a = sorted(s)
    n = len(a) - 1
    while True:
        yield ''.join(a)

        #1. Find the largest index j such that a[j] < a[j + 1]
        for j in range(n-1, -1, -1):
            if a[j] < a[j + 1]:
                break
        else:
            return

        #2. Find the largest index k greater than j such that a[j] < a[k]
        v = a[j]
        for k in range(n, j, -1):
            if v < a[k]:
                break

        #3. Swap the value of a[j] with that of a[k].
        a[j], a[k] = a[k], a[j]

        #4. Reverse the tail of the sequence
        a[j+1:] = a[j+1:][::-1]

for s in lexico_permute_string('data'):
    print(s)

output

aadt
aatd
adat
adta
atad
atda
daat
data
dtaa
taad
tada
tdaa

Of course, if you want to collect the yielded strings into a list you can do

list(lexico_permute_string('data'))

or in recent Python versions:

[*lexico_permute_string('data')]