What's the easiest way to use a linked list in python? In scheme, a linked list is defined simply by '(1 2 3 4 5). Python's lists, [1, 2, 3, 4, 5], and tuples, (1, 2, 3, 4, 5), are not, in fact, linked lists, and linked lists have some nice properties such as constant-time concatenation, and being able to reference separate parts of them. Make them immutable and they are really easy to work with!
- 216,039
- 159
- 467
- 667
-
13This might help you visualize it.. http://pythontutor.com/visualize.html#code=a+%3D+(1,+(2,+(3,+None)))%0Ab+%3D+a%5B1%5D&mode=display&cumulative=false&py=2&curInstr=2 – Dec 09 '12 at 07:41
-
@user1889082 awesome! it really helps me to understand some python concept – Jerry An Dec 19 '20 at 14:19
29 Answers
For some needs, a deque may also be useful. You can add and remove items on both ends of a deque at O(1) cost.
from collections import deque
d = deque([1,2,3,4])
print d
for x in d:
print x
print d.pop(), d
- 9,234
- 5
- 46
- 65
- 37,567
- 15
- 65
- 82
-
20While `deque` is a useful data type, it is not a linked list (though it is implemented using doubly linked list at C level). So it answers the question "what would you use *instead of* linked lists in Python?" and in that case the first answer should be (for some needs) an ordinary Python list (it is also not a linked list). – jfs Oct 19 '13 at 20:26
-
4@J.F.Sebastian: I almost agree with you :) I think the question this answers is rather: "What's the pythonic way to solve a problem that uses a linked list in other languages". It's not that linked lists aren't useful, it's just that problems where a deque doesn't work is very rare. – Emil Stenström Oct 20 '13 at 20:46
-
9It has nothing to do with "Pythonic": a linked list is a different data structure than a deque, and across the various operations the two support, they have different running times. – Thanatos Apr 07 '14 at 20:48
-
@Thanatos could you elaborate on the functional difference between a deque and a doubly linked list? At a glance they seem nearly identical. – dimo414 Jul 18 '14 at 20:25
-
6@dimo414: Linked lists typically prohibit indexing (no `linked_list[n]`) because it would be O(n). Dequeues allow it, and perform it in O(1). However, linked lists, given an iterator into the list, can allow O(1) insertion and removal, whereas deques cannot (it's O(n), like a vector). (Except at the front and end, where both deques and linked lists are both O(1). (though the deque is likely amortized O(1). The linked list is not.) – Thanatos Jul 19 '14 at 06:39
-
@Thanatos: I don't see how `deque` would allow O(1) index access. A simple `timeit` benchmark would show that [the time depends on `n`](http://goo.gl/oaTtLz). – jfs Jul 30 '15 at 22:00
-
@Thanatos The python docs specifically state that `deque` index access is `O(n)` in the middle. It behaves like a linked list in almost every way, even if the name is different. – Mad Physicist Oct 29 '15 at 16:02
-
3@MadPhysicist *"It [deque] behaves like a linked list in almost every way, even if the name is different."* — it is either wrong or meaningless: it is wrong because linked lists may provide different guarantees for time complexities e.g., you can remove an element (known position) from a linked list in O(1) while deque doesn't promise it (it is `O(n)`). If "almost every way" allows to ignore the difference in big O then your statement is meaningless because we could use a Python builtin list as a deque if it weren't for pop(0), insert(0,v) big O guarantees. – jfs Sep 13 '16 at 11:20
-
deque does not support insert/delete a node. The `delete` method performs a search for values and thus is O(n). – boh Oct 12 '17 at 14:39
I wrote this up the other day
#! /usr/bin/env python
class Node(object):
def __init__(self):
self.data = None # contains the data
self.next = None # contains the reference to the next node
class LinkedList:
def __init__(self):
self.cur_node = None
def add_node(self, data):
new_node = Node() # create a new node
new_node.data = data
new_node.next = self.cur_node # link the new node to the 'previous' node.
self.cur_node = new_node # set the current node to the new one.
def list_print(self):
node = self.cur_node # cant point to ll!
while node:
print node.data
node = node.next
ll = LinkedList()
ll.add_node(1)
ll.add_node(2)
ll.add_node(3)
ll.list_print()
- 5,154
- 2
- 38
- 55
- 38,929
- 21
- 58
- 60
-
how would you be able to go through the list and search for a specific node with specific data? – locoboy Aug 25 '11 at 17:17
-
1@locoboy the code to do that would be similar in logic to the code in `list_print()`. – Dennis Dec 11 '13 at 19:28
-
1
Here is some list functions based on Martin v. Löwis's representation:
cons = lambda el, lst: (el, lst)
mklist = lambda *args: reduce(lambda lst, el: cons(el, lst), reversed(args), None)
car = lambda lst: lst[0] if lst else lst
cdr = lambda lst: lst[1] if lst else lst
nth = lambda n, lst: nth(n-1, cdr(lst)) if n > 0 else car(lst)
length = lambda lst, count=0: length(cdr(lst), count+1) if lst else count
begin = lambda *args: args[-1]
display = lambda lst: begin(w("%s " % car(lst)), display(cdr(lst))) if lst else w("nil\n")
where w = sys.stdout.write
Although doubly linked lists are famously used in Raymond Hettinger's ordered set recipe, singly linked lists have no practical value in Python.
I've never used a singly linked list in Python for any problem except educational.
Thomas Watnedal suggested a good educational resource How to Think Like a Computer Scientist, Chapter 17: Linked lists:
A linked list is either:
- the empty list, represented by None, or
a node that contains a cargo object and a reference to a linked list.
class Node: def __init__(self, cargo=None, next=None): self.car = cargo self.cdr = next def __str__(self): return str(self.car) def display(lst): if lst: w("%s " % lst) display(lst.cdr) else: w("nil\n")
-
37You say: You've never used a singly linked list in Python for any problem except educational. That's good for you :-) But I can assure you: There ARE problems in the real world where a linked list will provide an ideal solution :-) That's why I scanned StackOverflow for linked lists in the first place :-) – Regis May Jan 27 '17 at 21:30
-
1@RegisMay: there was a long list of comments where your point and others were discussed. Unfortunately, they are deleted. I don't want to rehash the same arguments again. You could try to ask a moderator to undo the deletion. – jfs Jan 27 '17 at 21:34
-
Thank you for your reply, Sebastian. Undelete will not be necessary. It's enough if my comment (or a similar one) will remain visible. Otherwise some beginners might think linked lists are purely academic. Which they aren't: They solve problems other list variants simply can't. These problems might by very specific, but these problems do exist. In real world. – Regis May Jan 27 '17 at 21:49
-
9@RegisMay: would you mind providing a link to a specific practical code example? (note: it should be "a singly linked list in Python" "In real world": describe the benefits for your example e.g., readability, performance or other "practical value" of your choosing). I've made a similar request in the past: in 8 years, zero links except for doubly linked lists used in Raymond Hettinger's ordered set recipe--perhaps, it might be explained that only programmers new to Python read this question--your input would be valuable and highly appreciated. – jfs Jan 27 '17 at 22:02
-
3Oh, sorry. I'm not a native English speaker and confused "a singly linked list" with "a single linked list". Nevertheless I require a (double) linked list - which doesn't exist in python. A deque doesn't help as I need direct access to each single element without iterating over all elements. My goal: I want to implement a cache. Nevertheless: If my imperfection in the English language renders my comments out of place please delete these comments. Sorry for any inconvenience. – Regis May Jan 29 '17 at 19:00
-
@RegisMay, you seem to be implying that there are scenarios where a custom linked list implementation would offer advantages over python's built-in lists. Could you please give us an example of such a scenario? – Digio Mar 27 '17 at 08:49
-
5One practical advantage of a singly linked list over doubly linked lists or arrays (which Python uses internally for lists) is that two linked lists can share a tail. This is very useful for dynamic algorithms that require saved values from previous iterations where sharing list tails can reduce memory complexity from quadratic to linear and eliminate time overhead due to copying. – saolof Jun 25 '17 at 12:08
-
1@saolof: Do you have any example in Python? (I'm not talking about linked lists used at C level) – jfs Jun 25 '17 at 13:47
-
1@jfs: Sharing a (often very short) tail is the basis of the usual [disjoint-set forest](https://en.wikipedia.org/wiki/Disjoint-set_data_structure) implementation. – Davis Herring Sep 19 '17 at 05:05
-
2@DavisHerring: I don't see Python code there. Please, read [my comment above (from Jan 27)](https://stackoverflow.com/questions/280243/python-linked-list/283630?noredirect=1#comment70989964_283630). Here's a practical [code example that uses the concept of disjoint-set](https://stackoverflow.com/a/13716804/4279) -- notice: the implementation uses ordinary Python dicts to implement union, find operations. – jfs Sep 19 '17 at 09:29
-
1@jfs: The link wasn't for Python code; the ideas are not Python-specific. You can always store an object's data in a separate `dict` (writing `attr[x]` instead of `x.attr`). If you have more than one attribute, however, packaging it as an object is usually nicer. – Davis Herring Sep 19 '17 at 13:55
-
2
-
1@jfs: Sorry: I spoke too tersely. I don't have it to post, but since I used a linked list to implement a DSF _in Python_, I mentioned it as an example _problem_ that should (probably) be a linked list in any language that supports them. (The `dict` at your link has to be a global variable with weak references, for instance, if the structure is an implementation detail.) Python is not special here: the easy `dict` syntax is nice, but the (admittedly rare) applications of linked lists are about efficiency and aliasing, not syntax. – Davis Herring Sep 19 '17 at 14:11
-
1@DavisHerring please, do demonstrate the efficiency in pure Python with a specific *code example* – jfs Sep 19 '17 at 14:48
-
@RegisMay: A simple example where a native linked list improves performance of an algorithm is for the standard longest increasing subsequence problem. Basically, when you extend an existing increasing subsequence, if you use regular Python lists, you end up doing an O(n) copy to extend the subsequence. If you use singly linked lists, you simply create a new node having a value of the current element and set it's next pointer to point to the remainder of the subsequence. So it's functioning as a predecessor pointer. Thus O(n) copy turns to O(1) copy. – Gino Sep 26 '17 at 17:08
-
1@Gino: I'm sure there are many algorithms that might benefit theoretically from using a singly linked list. Please, do notice that I've asked for a [**code example**](https://stackoverflow.com/questions/280243/python-linked-list/283630?noredirect=1#comment70989964_283630). Here's [`max_increasing_subseq_length(numbers)` function in pure Python (O(n log n) -- no linked lists)](https://ru.stackoverflow.com/a/623894/23044) – jfs Sep 26 '17 at 17:30
-
1My post was in response to your original assertion that "I've never used a singly linked list in Python for any problem except educational." The longest increasing subsequence problem *can* be implemented using lists, but if you study the algorithm, you'll see that it's actually using arrays to *simulate* a linked list to achieve O(1) copies (per my previous post). For clarity and maintainability purposes in real world apps, it's probably better to simply use linked lists. So, for clarity and maintainability in the real world, this is an example of linked lists being important. – Gino Sep 26 '17 at 17:32
-
1@Gino: if it is important, it shouldn't be so hard to provide a "real world" code example? – jfs Sep 26 '17 at 17:41
-
For a code example, see https://rosettacode.org/wiki/Longest_increasing_subsequence#Python -- In the code, the P list is actually a simulated linked list and is used to reconstruct the actual longest increasing subsequence. – Gino Sep 26 '17 at 17:43
-
1
-
It looks like "longest increasing subsequence" is useful in statistics (real world apps). "Finding the longest increasing subsequence can be very useful not only at the Google/Yahoo/Facebook interview, but also in various fields of statistics." -- https://dzone.com/articles/algorithm-week-longest – Gino Sep 26 '17 at 17:48
-
1@Gino: again. Show "real world" (using your terminology) **code example**. For example, `collections.OrderedDict` from stdlib is implemented using a [doubly linked list](https://github.com/python/cpython/blob/c740e4fe8a9bc5815dc18c38d7f7600b128c3c51/Lib/collections/__init__.py#L71) (though actually C implementation (`_collections.OrderedDict`) is used if it is available). It is wildly used e.g., in `requests` module. – jfs Sep 26 '17 at 20:37
-
3That rosettacode link *was* a real world example, which uses a simulated linked list in place of an actual linked list. Take a look at it, rewrite it to use an actual linked list, for improved clarity and readability, and there you have the real world example of a linked list being used to improve existing code. And, secondly, the longest increasing subsequence algorithm is used in the real world, in statistics, so there you have it. Q.E.D. :). Beyond that, let's just agree to disagree. :) – Gino Sep 26 '17 at 21:06
-
1@Gino I don't believe in alternatives facts and the fact is you have failed to provide a code example from a real world product. – jfs Sep 26 '17 at 21:15
-
1BTW, spend a little time to study the longest increasing sequence algorithm. Prior to that, I had actually not seen how valuable a linked list could be, and I probably would have agreed with your original assertion, that linked lists were not needed in Python. If you study the algorithm, you'll see how it turns the copy-list aspect from an O(n) time complexity operation into an O(1) time complexity operation. Pretty cool stuff. :). – Gino Sep 26 '17 at 21:19
-
1More fuel for the fire: here is a question that actually needs a singly linked list for an optimal solution: https://stackoverflow.com/q/47293793/2988730. I have not posted my answer yet, but I can assure you that there is a real world application for this problem. – Mad Physicist Nov 15 '17 at 21:37
-
1@MadPhysicist "needs" is a too strong word. I've used [stack-based approach](https://stackoverflow.com/q/27652073/4279) successfully for similar problems (I don't know whether it is applicable here) – jfs Nov 15 '17 at 21:48
-
@jfs. I will post the answer in a few minutes. I could not think of anything that would run in under `O(m+n)` without using a linked list, (`m` is the number of rectangles, `n` is the number of overlaps). I would be very interested to see your algorithm. – Mad Physicist Nov 15 '17 at 21:50
-
@jfs I do have an example where a linked list is useful, imagine if you try to iterate trough a list where you always want to start from a different position but go trough all the elements. ex: First iteration [1,2,3,4] , second iteration [2,3,4,1], third [3,4,1,2]. with a linked list you can easily do this out of the box... without having to move any element,.. – Jbeat Feb 06 '20 at 03:02
-
@Jbeat: would you mind providing a specific code example? ([as described in my comment near the top](https://stackoverflow.com/questions/280243/python-linked-list/283630#comment70989964_283630)) – jfs Feb 06 '20 at 16:59
-
As already noted, singly linked list have the feature of being persistent data structures, i.e. you can concurrently modify them at will without locks (and share the tail). That is something Python's list cannot do—unless you always copy the whole list (bad for memory). No, I will not provide a code example because that useful property already holds on the theoretical level and implementing it in _any_ language won't change that. You can find singly linked lists in the Linux kernel and you bet these people did their research. Of course not in Python, but as said, the language doesn't matter. – primfaktor Oct 20 '21 at 06:48
-
@primfaktor: in theory there is no difference between the theory and practice -- in practice, there is ;) Python is not C. Different trade-offs, different (though overlapping) application domains. Unless you can give a Python code example from a real world product, you just proving the point. – jfs Oct 20 '21 at 15:54
The accepted answer is rather complicated. Here is a more standard design:
L = LinkedList()
L.insert(1)
L.insert(1)
L.insert(2)
L.insert(4)
print L
L.clear()
print L
It is a simple LinkedList class based on the straightforward C++ design and Chapter 17: Linked lists, as recommended by Thomas Watnedal.
class Node:
def __init__(self, value = None, next = None):
self.value = value
self.next = next
def __str__(self):
return 'Node ['+str(self.value)+']'
class LinkedList:
def __init__(self):
self.first = None
self.last = None
def insert(self, x):
if self.first == None:
self.first = Node(x, None)
self.last = self.first
elif self.last == self.first:
self.last = Node(x, None)
self.first.next = self.last
else:
current = Node(x, None)
self.last.next = current
self.last = current
def __str__(self):
if self.first != None:
current = self.first
out = 'LinkedList [\n' +str(current.value) +'\n'
while current.next != None:
current = current.next
out += str(current.value) + '\n'
return out + ']'
return 'LinkedList []'
def clear(self):
self.__init__()
- 1
- 1
- 15,845
- 19
- 82
- 102
-
9I like this answer. One nit, I believe that ```X is None``` is preferred over ```==```. http://stackoverflow.com/a/2988117/1740227 – mateor Aug 16 '14 at 04:17
-
Is the second branch of `insert` not a particular case of the third, so that you can entirely remove the `elif` clause? – Jaime Jan 23 '16 at 04:49
Immutable lists are best represented through two-tuples, with None representing NIL. To allow simple formulation of such lists, you can use this function:
def mklist(*args):
result = None
for element in reversed(args):
result = (element, result)
return result
To work with such lists, I'd rather provide the whole collection of LISP functions (i.e. first, second, nth, etc), than introducing methods.
- 120,633
- 17
- 193
- 234
Here's a slightly more complex version of a linked list class, with a similar interface to python's sequence types (ie. supports indexing, slicing, concatenation with arbitrary sequences etc). It should have O(1) prepend, doesn't copy data unless it needs to and can be used pretty interchangably with tuples.
It won't be as space or time efficient as lisp cons cells, as python classes are obviously a bit more heavyweight (You could improve things slightly with "__slots__ = '_head','_tail'" to reduce memory usage). It will have the desired big O performance characteristics however.
Example of usage:
>>> l = LinkedList([1,2,3,4])
>>> l
LinkedList([1, 2, 3, 4])
>>> l.head, l.tail
(1, LinkedList([2, 3, 4]))
# Prepending is O(1) and can be done with:
LinkedList.cons(0, l)
LinkedList([0, 1, 2, 3, 4])
# Or prepending arbitrary sequences (Still no copy of l performed):
[-1,0] + l
LinkedList([-1, 0, 1, 2, 3, 4])
# Normal list indexing and slice operations can be performed.
# Again, no copy is made unless needed.
>>> l[1], l[-1], l[2:]
(2, 4, LinkedList([3, 4]))
>>> assert l[2:] is l.next.next
# For cases where the slice stops before the end, or uses a
# non-contiguous range, we do need to create a copy. However
# this should be transparent to the user.
>>> LinkedList(range(100))[-10::2]
LinkedList([90, 92, 94, 96, 98])
Implementation:
import itertools
class LinkedList(object):
"""Immutable linked list class."""
def __new__(cls, l=[]):
if isinstance(l, LinkedList): return l # Immutable, so no copy needed.
i = iter(l)
try:
head = i.next()
except StopIteration:
return cls.EmptyList # Return empty list singleton.
tail = LinkedList(i)
obj = super(LinkedList, cls).__new__(cls)
obj._head = head
obj._tail = tail
return obj
@classmethod
def cons(cls, head, tail):
ll = cls([head])
if not isinstance(tail, cls):
tail = cls(tail)
ll._tail = tail
return ll
# head and tail are not modifiable
@property
def head(self): return self._head
@property
def tail(self): return self._tail
def __nonzero__(self): return True
def __len__(self):
return sum(1 for _ in self)
def __add__(self, other):
other = LinkedList(other)
if not self: return other # () + l = l
start=l = LinkedList(iter(self)) # Create copy, as we'll mutate
while l:
if not l._tail: # Last element?
l._tail = other
break
l = l._tail
return start
def __radd__(self, other):
return LinkedList(other) + self
def __iter__(self):
x=self
while x:
yield x.head
x=x.tail
def __getitem__(self, idx):
"""Get item at specified index"""
if isinstance(idx, slice):
# Special case: Avoid constructing a new list, or performing O(n) length
# calculation for slices like l[3:]. Since we're immutable, just return
# the appropriate node. This becomes O(start) rather than O(n).
# We can't do this for more complicated slices however (eg [l:4]
start = idx.start or 0
if (start >= 0) and (idx.stop is None) and (idx.step is None or idx.step == 1):
no_copy_needed=True
else:
length = len(self) # Need to calc length.
start, stop, step = idx.indices(length)
no_copy_needed = (stop == length) and (step == 1)
if no_copy_needed:
l = self
for i in range(start):
if not l: break # End of list.
l=l.tail
return l
else:
# We need to construct a new list.
if step < 1: # Need to instantiate list to deal with -ve step
return LinkedList(list(self)[start:stop:step])
else:
return LinkedList(itertools.islice(iter(self), start, stop, step))
else:
# Non-slice index.
if idx < 0: idx = len(self)+idx
if not self: raise IndexError("list index out of range")
if idx == 0: return self.head
return self.tail[idx-1]
def __mul__(self, n):
if n <= 0: return Nil
l=self
for i in range(n-1): l += self
return l
def __rmul__(self, n): return self * n
# Ideally we should compute the has ourselves rather than construct
# a temporary tuple as below. I haven't impemented this here
def __hash__(self): return hash(tuple(self))
def __eq__(self, other): return self._cmp(other) == 0
def __ne__(self, other): return not self == other
def __lt__(self, other): return self._cmp(other) < 0
def __gt__(self, other): return self._cmp(other) > 0
def __le__(self, other): return self._cmp(other) <= 0
def __ge__(self, other): return self._cmp(other) >= 0
def _cmp(self, other):
"""Acts as cmp(): -1 for self<other, 0 for equal, 1 for greater"""
if not isinstance(other, LinkedList):
return cmp(LinkedList,type(other)) # Arbitrary ordering.
A, B = iter(self), iter(other)
for a,b in itertools.izip(A,B):
if a<b: return -1
elif a > b: return 1
try:
A.next()
return 1 # a has more items.
except StopIteration: pass
try:
B.next()
return -1 # b has more items.
except StopIteration: pass
return 0 # Lists are equal
def __repr__(self):
return "LinkedList([%s])" % ', '.join(map(repr,self))
class EmptyList(LinkedList):
"""A singleton representing an empty list."""
def __new__(cls):
return object.__new__(cls)
def __iter__(self): return iter([])
def __nonzero__(self): return False
@property
def head(self): raise IndexError("End of list")
@property
def tail(self): raise IndexError("End of list")
# Create EmptyList singleton
LinkedList.EmptyList = EmptyList()
del EmptyList
- 113,117
- 28
- 106
- 111
-
I guess it's not so surprising, but this 8 year old (!) example does not work with python 3 :) – Andy Hayden Mar 14 '17 at 07:18
-
1Please provide explanation for __new__ and a just a bit of overall explanation. – anukalp Aug 08 '17 at 02:02
llist — Linked list datatypes for Python
llist module implements linked list data structures. It supports a doubly linked list, i.e. dllist and a singly linked data structure sllist.
dllist objects
This object represents a doubly linked list data structure.
first
First dllistnode object in the list. None if list is empty.
last
Last dllistnode object in the list. None if list is empty.
dllist objects also support the following methods:
append(x)
Add x to the right side of the list and return inserted dllistnode.
appendleft(x)
Add x to the left side of the list and return inserted dllistnode.
appendright(x)
Add x to the right side of the list and return inserted dllistnode.
clear()
Remove all nodes from the list.
extend(iterable)
Append elements from iterable to the right side of the list.
extendleft(iterable)
Append elements from iterable to the left side of the list.
extendright(iterable)
Append elements from iterable to the right side of the list.
insert(x[, before])
Add x to the right side of the list if before is not specified, or insert x to the left side of dllistnode before. Return inserted dllistnode.
nodeat(index)
Return node (of type dllistnode) at index.
pop()
Remove and return an element’s value from the right side of the list.
popleft()
Remove and return an element’s value from the left side of the list.
popright()
Remove and return an element’s value from the right side of the list
remove(node)
Remove node from the list and return the element which was stored in it.
dllistnode objects
class llist.dllistnode([value])
Return a new doubly linked list node, initialized (optionally) with value.
dllistnode objects provide the following attributes:
next
Next node in the list. This attribute is read-only.
prev
Previous node in the list. This attribute is read-only.
value
Value stored in this node. Compiled from this reference
sllist
class llist.sllist([iterable])
Return a new singly linked list initialized with elements from iterable. If iterable is not specified, the new sllist is empty.
A similar set of attributes and operations are defined for this sllist object. See this reference for more information.
- 3,107
- 1
- 22
- 20
class Node(object):
def __init__(self, data=None, next=None):
self.data = data
self.next = next
def setData(self, data):
self.data = data
return self.data
def setNext(self, next):
self.next = next
def getNext(self):
return self.next
def hasNext(self):
return self.next != None
class singleLinkList(object):
def __init__(self):
self.head = None
def isEmpty(self):
return self.head == None
def insertAtBeginning(self, data):
newNode = Node()
newNode.setData(data)
if self.listLength() == 0:
self.head = newNode
else:
newNode.setNext(self.head)
self.head = newNode
def insertAtEnd(self, data):
newNode = Node()
newNode.setData(data)
current = self.head
while current.getNext() != None:
current = current.getNext()
current.setNext(newNode)
def listLength(self):
current = self.head
count = 0
while current != None:
count += 1
current = current.getNext()
return count
def print_llist(self):
current = self.head
print("List Start.")
while current != None:
print(current.getData())
current = current.getNext()
print("List End.")
if __name__ == '__main__':
ll = singleLinkList()
ll.insertAtBeginning(55)
ll.insertAtEnd(56)
ll.print_llist()
print(ll.listLength())
- 320
- 3
- 8
The following is what I came up with. It's similer to Riccardo C.'s, in this thread, except it prints the numbers in order instead of in reverse. I also made the LinkedList object a Python Iterator in order to print the list out like you would a normal Python list.
class Node:
def __init__(self, data=None):
self.data = data
self.next = None
def __str__(self):
return str(self.data)
class LinkedList:
def __init__(self):
self.head = None
self.curr = None
self.tail = None
def __iter__(self):
return self
def next(self):
if self.head and not self.curr:
self.curr = self.head
return self.curr
elif self.curr.next:
self.curr = self.curr.next
return self.curr
else:
raise StopIteration
def append(self, data):
n = Node(data)
if not self.head:
self.head = n
self.tail = n
else:
self.tail.next = n
self.tail = self.tail.next
# Add 5 nodes
ll = LinkedList()
for i in range(1, 6):
ll.append(i)
# print out the list
for n in ll:
print n
"""
Example output:
$ python linked_list.py
1
2
3
4
5
"""
- 1
- 1
- 5,352
- 6
- 22
- 27
-
It looks like there's a bug before raise StopIteration. If you're going to preserve the current node as an internal piece of state, you need to reset it before you stop iterating so that the next time the linked list is looped over, it will enter your first clause. – Tim Wilder Apr 30 '14 at 03:25
I just did this as a fun toy. It should be immutable as long as you don't touch the underscore-prefixed methods, and it implements a bunch of Python magic like indexing and len.
- 277
- 2
- 11
Here is my solution:
Implementation
class Node:
def __init__(self, initdata):
self.data = initdata
self.next = None
def get_data(self):
return self.data
def set_data(self, data):
self.data = data
def get_next(self):
return self.next
def set_next(self, node):
self.next = node
# ------------------------ Link List class ------------------------------- #
class LinkList:
def __init__(self):
self.head = None
def is_empty(self):
return self.head == None
def traversal(self, data=None):
node = self.head
index = 0
found = False
while node is not None and not found:
if node.get_data() == data:
found = True
else:
node = node.get_next()
index += 1
return (node, index)
def size(self):
_, count = self.traversal(None)
return count
def search(self, data):
node, _ = self.traversal(data)
return node
def add(self, data):
node = Node(data)
node.set_next(self.head)
self.head = node
def remove(self, data):
previous_node = None
current_node = self.head
found = False
while current_node is not None and not found:
if current_node.get_data() == data:
found = True
if previous_node:
previous_node.set_next(current_node.get_next())
else:
self.head = current_node
else:
previous_node = current_node
current_node = current_node.get_next()
return found
Usage
link_list = LinkList()
link_list.add(10)
link_list.add(20)
link_list.add(30)
link_list.add(40)
link_list.add(50)
link_list.size()
link_list.search(30)
link_list.remove(20)
Original Implementation Idea
- 31
- 2
I based this additional function on Nick Stinemates
def add_node_at_end(self, data):
new_node = Node()
node = self.curr_node
while node:
if node.next == None:
node.next = new_node
new_node.next = None
new_node.data = data
node = node.next
The method he has adds the new node at the beginning while I have seen a lot of implementations which usually add a new node at the end but whatever, it is fun to do.
- 1
- 1
class LL(object):
def __init__(self,val):
self.val = val
self.next = None
def pushNodeEnd(self,top,val):
if top is None:
top.val=val
top.next=None
else:
tmp=top
while (tmp.next != None):
tmp=tmp.next
newNode=LL(val)
newNode.next=None
tmp.next=newNode
def pushNodeFront(self,top,val):
if top is None:
top.val=val
top.next=None
else:
newNode=LL(val)
newNode.next=top
top=newNode
def popNodeFront(self,top):
if top is None:
return
else:
sav=top
top=top.next
return sav
def popNodeEnd(self,top):
if top is None:
return
else:
tmp=top
while (tmp.next != None):
prev=tmp
tmp=tmp.next
prev.next=None
return tmp
top=LL(10)
top.pushNodeEnd(top, 20)
top.pushNodeEnd(top, 30)
pop=top.popNodeEnd(top)
print (pop.val)
- 27
- 1
- 2
I've put a Python 2.x and 3.x singly-linked list class at https://pypi.python.org/pypi/linked_list_mod/
It's tested with CPython 2.7, CPython 3.4, Pypy 2.3.1, Pypy3 2.3.1, and Jython 2.7b2, and comes with a nice automated test suite.
It also includes LIFO and FIFO classes.
They aren't immutable though.
- 6,698
- 24
- 25
When using immutable linked lists, consider using Python's tuple directly.
ls = (1, 2, 3, 4, 5)
def first(ls): return ls[0]
def rest(ls): return ls[1:]
Its really that ease, and you get to keep the additional funcitons like len(ls), x in ls, etc.
- 37,567
- 15
- 65
- 82
-
Tuples don't have the performance characteristics he asked for. Your rest() is O(n) as opposed to O(1) for a linked list, as is consing a new head. – Brian Nov 11 '08 at 13:10
-
Right. My point is: Do not ask for linked lists to implement your algorithm, rather use the python features to optimally implement it. E.g. iterating over a linked list is O(n), as is iterating over a python tuple using "for x in t:" – Ber Nov 11 '08 at 19:29
-
i think the right way to use tuples to implement linked lists is the accepted answer here. your way uses immutable array-like-objects – Claudiu Nov 11 '08 at 21:56
class LinkedStack:
'''LIFO Stack implementation using a singly linked list for storage.'''
_ToList = []
#---------- nested _Node class -----------------------------
class _Node:
'''Lightweight, nonpublic class for storing a singly linked node.'''
__slots__ = '_element', '_next' #streamline memory usage
def __init__(self, element, next):
self._element = element
self._next = next
#--------------- stack methods ---------------------------------
def __init__(self):
'''Create an empty stack.'''
self._head = None
self._size = 0
def __len__(self):
'''Return the number of elements in the stack.'''
return self._size
def IsEmpty(self):
'''Return True if the stack is empty'''
return self._size == 0
def Push(self,e):
'''Add element e to the top of the Stack.'''
self._head = self._Node(e, self._head) #create and link a new node
self._size +=1
self._ToList.append(e)
def Top(self):
'''Return (but do not remove) the element at the top of the stack.
Raise exception if the stack is empty
'''
if self.IsEmpty():
raise Exception('Stack is empty')
return self._head._element #top of stack is at head of list
def Pop(self):
'''Remove and return the element from the top of the stack (i.e. LIFO).
Raise exception if the stack is empty
'''
if self.IsEmpty():
raise Exception('Stack is empty')
answer = self._head._element
self._head = self._head._next #bypass the former top node
self._size -=1
self._ToList.remove(answer)
return answer
def Count(self):
'''Return how many nodes the stack has'''
return self.__len__()
def Clear(self):
'''Delete all nodes'''
for i in range(self.Count()):
self.Pop()
def ToList(self):
return self._ToList
- 1,091
- 1
- 13
- 17
Linked List Class
class LinkedStack:
# Nested Node Class
class Node:
def __init__(self, element, next):
self.__element = element
self.__next = next
def get_next(self):
return self.__next
def get_element(self):
return self.__element
def __init__(self):
self.head = None
self.size = 0
self.data = []
def __len__(self):
return self.size
def __str__(self):
return str(self.data)
def is_empty(self):
return self.size == 0
def push(self, e):
newest = self.Node(e, self.head)
self.head = newest
self.size += 1
self.data.append(newest)
def top(self):
if self.is_empty():
raise Empty('Stack is empty')
return self.head.__element
def pop(self):
if self.is_empty():
raise Empty('Stack is empty')
answer = self.head.element
self.head = self.head.next
self.size -= 1
return answer
Usage
from LinkedStack import LinkedStack
x = LinkedStack()
x.push(10)
x.push(25)
x.push(55)
for i in range(x.size - 1, -1, -1):
print '|', x.data[i].get_element(), '|' ,
#next object
if x.data[i].get_next() == None:
print '--> None'
else:
print x.data[i].get_next().get_element(), '-|----> ',
Output
| 55 | 25 -|----> | 25 | 10 -|----> | 10 | --> None
- 19,881
- 24
- 93
- 121
Here is my simple implementation:
class Node:
def __init__(self):
self.data = None
self.next = None
def __str__(self):
return "Data %s: Next -> %s"%(self.data, self.next)
class LinkedList:
def __init__(self):
self.head = Node()
self.curNode = self.head
def insertNode(self, data):
node = Node()
node.data = data
node.next = None
if self.head.data == None:
self.head = node
self.curNode = node
else:
self.curNode.next = node
self.curNode = node
def printList(self):
print self.head
l = LinkedList()
l.insertNode(1)
l.insertNode(2)
l.insertNode(34)
Output:
Data 1: Next -> Data 2: Next -> Data 34: Next -> Data 4: Next -> None
- 3,277
- 5
- 19
- 23
class LinkedList:
def __init__(self, value):
self.value = value
self.next = None
def insert(self, node):
if not self.next:
self.next = node
else:
self.next.insert(node)
def __str__(self):
if self.next:
return '%s -> %s' % (self.value, str(self.next))
else:
return ' %s ' % self.value
if __name__ == "__main__":
items = ['a', 'b', 'c', 'd', 'e']
ll = None
for item in items:
if ll:
next_ll = LinkedList(item)
ll.insert(next_ll)
else:
ll = LinkedList(item)
print('[ %s ]' % ll)
- 71
- 1
- 3
First of all, I assume you want linked lists. In practice, you can use collections.deque, whose current CPython implementation is a doubly linked list of blocks (each block contains an array of 62 cargo objects). It subsumes linked list's functionality. You can also search for a C extension called llist on pypi. If you want a pure-Python and easy-to-follow implementation of the linked list ADT, you can take a look at my following minimal implementation.
class Node (object):
""" Node for a linked list. """
def __init__ (self, value, next=None):
self.value = value
self.next = next
class LinkedList (object):
""" Linked list ADT implementation using class.
A linked list is a wrapper of a head pointer
that references either None, or a node that contains
a reference to a linked list.
"""
def __init__ (self, iterable=()):
self.head = None
for x in iterable:
self.head = Node(x, self.head)
def __iter__ (self):
p = self.head
while p is not None:
yield p.value
p = p.next
def prepend (self, x): # 'appendleft'
self.head = Node(x, self.head)
def reverse (self):
""" In-place reversal. """
p = self.head
self.head = None
while p is not None:
p0, p = p, p.next
p0.next = self.head
self.head = p0
if __name__ == '__main__':
ll = LinkedList([6,5,4])
ll.prepend(3); ll.prepend(2)
print list(ll)
ll.reverse()
print list(ll)
- 3
- 1
Sample of a doubly linked list (save as linkedlist.py):
class node:
def __init__(self, before=None, cargo=None, next=None):
self._previous = before
self._cargo = cargo
self._next = next
def __str__(self):
return str(self._cargo) or None
class linkedList:
def __init__(self):
self._head = None
self._length = 0
def add(self, cargo):
n = node(None, cargo, self._head)
if self._head:
self._head._previous = n
self._head = n
self._length += 1
def search(self,cargo):
node = self._head
while (node and node._cargo != cargo):
node = node._next
return node
def delete(self,cargo):
node = self.search(cargo)
if node:
prev = node._previous
nx = node._next
if prev:
prev._next = node._next
else:
self._head = nx
nx._previous = None
if nx:
nx._previous = prev
else:
prev._next = None
self._length -= 1
def __str__(self):
print 'Size of linked list: ',self._length
node = self._head
while node:
print node
node = node._next
Testing (save as test.py):
from linkedlist import node, linkedList
def test():
print 'Testing Linked List'
l = linkedList()
l.add(10)
l.add(20)
l.add(30)
l.add(40)
l.add(50)
l.add(60)
print 'Linked List after insert nodes:'
l.__str__()
print 'Search some value, 30:'
node = l.search(30)
print node
print 'Delete some value, 30:'
node = l.delete(30)
l.__str__()
print 'Delete first element, 60:'
node = l.delete(60)
l.__str__()
print 'Delete last element, 10:'
node = l.delete(10)
l.__str__()
if __name__ == "__main__":
test()
Output:
Testing Linked List
Linked List after insert nodes:
Size of linked list: 6
60
50
40
30
20
10
Search some value, 30:
30
Delete some value, 30:
Size of linked list: 5
60
50
40
20
10
Delete first element, 60:
Size of linked list: 4
50
40
20
10
Delete last element, 10:
Size of linked list: 3
50
40
20
- 1,989
- 2
- 24
- 36
I did also write a Single Linked List based on some tutorial, which has the basic two Node and Linked List classes, and some additional methods for insertion, delete, reverse, sorting, and such.
It's not the best or easiest, works OK though.
"""
Single Linked List (SLL):
A simple object-oriented implementation of Single Linked List (SLL)
with some associated methods, such as create list, count nodes, delete nodes, and such.
"""
class Node:
"""
Instantiates a node
"""
def __init__(self, value):
"""
Node class constructor which sets the value and link of the node
"""
self.info = value
self.link = None
class SingleLinkedList:
"""
Instantiates the SLL class
"""
def __init__(self):
"""
SLL class constructor which sets the value and link of the node
"""
self.start = None
def create_single_linked_list(self):
"""
Reads values from stdin and appends them to this list and creates a SLL with integer nodes
"""
try:
number_of_nodes = int(input(" Enter a positive integer between 1-50 for the number of nodes you wish to have in the list: "))
if number_of_nodes <= 0 or number_of_nodes > 51:
print(" The number of nodes though must be an integer between 1 to 50!")
self.create_single_linked_list()
except Exception as e:
print(" Error: ", e)
self.create_single_linked_list()
try:
for _ in range(number_of_nodes):
try:
data = int(input(" Enter an integer for the node to be inserted: "))
self.insert_node_at_end(data)
except Exception as e:
print(" Error: ", e)
except Exception as e:
print(" Error: ", e)
def count_sll_nodes(self):
"""
Counts the nodes of the linked list
"""
try:
p = self.start
n = 0
while p is not None:
n += 1
p = p.link
if n >= 1:
print(f" The number of nodes in the linked list is {n}")
else:
print(f" The SLL does not have a node!")
except Exception as e:
print(" Error: ", e)
def search_sll_nodes(self, x):
"""
Searches the x integer in the linked list
"""
try:
position = 1
p = self.start
while p is not None:
if p.info == x:
print(f" YAAAY! We found {x} at position {position}")
return True
#Increment the position
position += 1
#Assign the next node to the current node
p = p.link
else:
print(f" Sorry! We couldn't find {x} at any position. Maybe, you might want to use option 9 and try again later!")
return False
except Exception as e:
print(" Error: ", e)
def display_sll(self):
"""
Displays the list
"""
try:
if self.start is None:
print(" Single linked list is empty!")
return
display_sll = " Single linked list nodes are: "
p = self.start
while p is not None:
display_sll += str(p.info) + "\t"
p = p.link
print(display_sll)
except Exception as e:
print(" Error: ", e)
def insert_node_in_beginning(self, data):
"""
Inserts an integer in the beginning of the linked list
"""
try:
temp = Node(data)
temp.link = self.start
self.start = temp
except Exception as e:
print(" Error: ", e)
def insert_node_at_end(self, data):
"""
Inserts an integer at the end of the linked list
"""
try:
temp = Node(data)
if self.start is None:
self.start = temp
return
p = self.start
while p.link is not None:
p = p.link
p.link = temp
except Exception as e:
print(" Error: ", e)
def insert_node_after_another(self, data, x):
"""
Inserts an integer after the x node
"""
try:
p = self.start
while p is not None:
if p.info == x:
break
p = p.link
if p is None:
print(f" Sorry! {x} is not in the list.")
else:
temp = Node(data)
temp.link = p.link
p.link = temp
except Exception as e:
print(" Error: ", e)
def insert_node_before_another(self, data, x):
"""
Inserts an integer before the x node
"""
try:
# If list is empty
if self.start is None:
print(" Sorry! The list is empty.")
return
# If x is the first node, and new node should be inserted before the first node
if x == self.start.info:
temp = Node(data)
temp.link = p.link
p.link = temp
# Finding the reference to the prior node containing x
p = self.start
while p.link is not None:
if p.link.info == x:
break
p = p.link
if p.link is not None:
print(f" Sorry! {x} is not in the list.")
else:
temp = Node(data)
temp.link = p.link
p.link = temp
except Exception as e:
print(" Error: ", e)
def insert_node_at_position(self, data, k):
"""
Inserts an integer in k position of the linked list
"""
try:
# if we wish to insert at the first node
if k == 1:
temp = Node(data)
temp.link = self.start
self.start = temp
return
p = self.start
i = 1
while i < k-1 and p is not None:
p = p.link
i += 1
if p is None:
print(f" The max position is {i}")
else:
temp = Node(data)
temp.link = self.start
self.start = temp
except Exception as e:
print(" Error: ", e)
def delete_a_node(self, x):
"""
Deletes a node of a linked list
"""
try:
# If list is empty
if self.start is None:
print(" Sorry! The list is empty.")
return
# If there is only one node
if self.start.info == x:
self.start = self.start.link
# If more than one node exists
p = self.start
while p.link is not None:
if p.link.info == x:
break
p = p.link
if p.link is None:
print(f" Sorry! {x} is not in the list.")
else:
p.link = p.link.link
except Exception as e:
print(" Error: ", e)
def delete_sll_first_node(self):
"""
Deletes the first node of a linked list
"""
try:
if self.start is None:
return
self.start = self.start.link
except Exception as e:
print(" Error: ", e)
def delete_sll_last_node(self):
"""
Deletes the last node of a linked list
"""
try:
# If the list is empty
if self.start is None:
return
# If there is only one node
if self.start.link is None:
self.start = None
return
# If there is more than one node
p = self.start
# Increment until we find the node prior to the last node
while p.link.link is not None:
p = p.link
p.link = None
except Exception as e:
print(" Error: ", e)
def reverse_sll(self):
"""
Reverses the linked list
"""
try:
prev = None
p = self.start
while p is not None:
next = p.link
p.link = prev
prev = p
p = next
self.start = prev
except Exception as e:
print(" Error: ", e)
def bubble_sort_sll_nodes_data(self):
"""
Bubble sorts the linked list on integer values
"""
try:
# If the list is empty or there is only one node
if self.start is None or self.start.link is None:
print(" The list has no or only one node and sorting is not required.")
end = None
while end != self.start.link:
p = self.start
while p.link != end:
q = p.link
if p.info > q.info:
p.info, q.info = q.info, p.info
p = p.link
end = p
except Exception as e:
print(" Error: ", e)
def bubble_sort_sll(self):
"""
Bubble sorts the linked list
"""
try:
# If the list is empty or there is only one node
if self.start is None or self.start.link is None:
print(" The list has no or only one node and sorting is not required.")
end = None
while end != self.start.link:
r = p = self.start
while p.link != end:
q = p.link
if p.info > q.info:
p.link = q.link
q.link = p
if p != self.start:
r.link = q.link
else:
self.start = q
p, q = q, p
r = p
p = p.link
end = p
except Exception as e:
print(" Error: ", e)
def sll_has_cycle(self):
"""
Tests the list for cycles using Tortoise and Hare Algorithm (Floyd's cycle detection algorithm)
"""
try:
if self.find_sll_cycle() is None:
return False
else:
return True
except Exception as e:
print(" Error: ", e)
def find_sll_cycle(self):
"""
Finds cycles in the list, if any
"""
try:
# If there is one node or none, there is no cycle
if self.start is None or self.start.link is None:
return None
# Otherwise,
slowR = self.start
fastR = self.start
while slowR is not None and fastR is not None:
slowR = slowR.link
fastR = fastR.link.link
if slowR == fastR:
return slowR
return None
except Exception as e:
print(" Error: ", e)
def remove_cycle_from_sll(self):
"""
Removes the cycles
"""
try:
c = self.find_sll_cycle()
# If there is no cycle
if c is None:
return
print(f" There is a cycle at node: ", c.info)
p = c
q = c
len_cycles = 0
while True:
len_cycles += 1
q = q.link
if p == q:
break
print(f" The cycle length is {len_cycles}")
len_rem_list = 0
p = self.start
while p != q:
len_rem_list += 1
p = p.link
q = q.link
print(f" The number of nodes not included in the cycle is {len_rem_list}")
length_list = len_rem_list + len_cycles
print(f" The SLL length is {length_list}")
# This for loop goes to the end of the SLL, and set the last node to None and the cycle is removed.
p = self.start
for _ in range(length_list-1):
p = p.link
p.link = None
except Exception as e:
print(" Error: ", e)
def insert_cycle_in_sll(self, x):
"""
Inserts a cycle at a node that contains x
"""
try:
if self.start is None:
return False
p = self.start
px = None
prev = None
while p is not None:
if p.info == x:
px = p
prev = p
p = p.link
if px is not None:
prev.link = px
else:
print(f" Sorry! {x} is not in the list.")
except Exception as e:
print(" Error: ", e)
def merge_using_new_list(self, list2):
"""
Merges two already sorted SLLs by creating new lists
"""
merge_list = SingleLinkedList()
merge_list.start = self._merge_using_new_list(self.start, list2.start)
return merge_list
def _merge_using_new_list(self, p1, p2):
"""
Private method of merge_using_new_list
"""
if p1.info <= p2.info:
Start_merge = Node(p1.info)
p1 = p1.link
else:
Start_merge = Node(p2.info)
p2 = p2.link
pM = Start_merge
while p1 is not None and p2 is not None:
if p1.info <= p2.info:
pM.link = Node(p1.info)
p1 = p1.link
else:
pM.link = Node(p2.info)
p2 = p2.link
pM = pM.link
#If the second list is finished, yet the first list has some nodes
while p1 is not None:
pM.link = Node(p1.info)
p1 = p1.link
pM = pM.link
#If the second list is finished, yet the first list has some nodes
while p2 is not None:
pM.link = Node(p2.info)
p2 = p2.link
pM = pM.link
return Start_merge
def merge_inplace(self, list2):
"""
Merges two already sorted SLLs in place in O(1) of space
"""
merge_list = SingleLinkedList()
merge_list.start = self._merge_inplace(self.start, list2.start)
return merge_list
def _merge_inplace(self, p1, p2):
"""
Merges two already sorted SLLs in place in O(1) of space
"""
if p1.info <= p2.info:
Start_merge = p1
p1 = p1.link
else:
Start_merge = p2
p2 = p2.link
pM = Start_merge
while p1 is not None and p2 is not None:
if p1.info <= p2.info:
pM.link = p1
pM = pM.link
p1 = p1.link
else:
pM.link = p2
pM = pM.link
p2 = p2.link
if p1 is None:
pM.link = p2
else:
pM.link = p1
return Start_merge
def merge_sort_sll(self):
"""
Sorts the linked list using merge sort algorithm
"""
self.start = self._merge_sort_recursive(self.start)
def _merge_sort_recursive(self, list_start):
"""
Recursively calls the merge sort algorithm for two divided lists
"""
# If the list is empty or has only one node
if list_start is None or list_start.link is None:
return list_start
# If the list has two nodes or more
start_one = list_start
start_two = self._divide_list(self_start)
start_one = self._merge_sort_recursive(start_one)
start_two = self._merge_sort_recursive(start_two)
start_merge = self._merge_inplace(start_one, start_two)
return start_merge
def _divide_list(self, p):
"""
Divides the linked list into two almost equally sized lists
"""
# Refers to the third nodes of the list
q = p.link.link
while q is not None and p is not None:
# Increments p one node at the time
p = p.link
# Increments q two nodes at the time
q = q.link.link
start_two = p.link
p.link = None
return start_two
def concat_second_list_to_sll(self, list2):
"""
Concatenates another SLL to an existing SLL
"""
# If the second SLL has no node
if list2.start is None:
return
# If the original SLL has no node
if self.start is None:
self.start = list2.start
return
# Otherwise traverse the original SLL
p = self.start
while p.link is not None:
p = p.link
# Link the last node to the first node of the second SLL
p.link = list2.start
def test_merge_using_new_list_and_inplace(self):
"""
"""
LIST_ONE = SingleLinkedList()
LIST_TWO = SingleLinkedList()
LIST_ONE.create_single_linked_list()
LIST_TWO.create_single_linked_list()
print("1️⃣ The unsorted first list is: ")
LIST_ONE.display_sll()
print("2️⃣ The unsorted second list is: ")
LIST_TWO.display_sll()
LIST_ONE.bubble_sort_sll_nodes_data()
LIST_TWO.bubble_sort_sll_nodes_data()
print("1️⃣ The sorted first list is: ")
LIST_ONE.display_sll()
print("2️⃣ The sorted second list is: ")
LIST_TWO.display_sll()
LIST_THREE = LIST_ONE.merge_using_new_list(LIST_TWO)
print("The merged list by creating a new list is: ")
LIST_THREE.display_sll()
LIST_FOUR = LIST_ONE.merge_inplace(LIST_TWO)
print("The in-place merged list is: ")
LIST_FOUR.display_sll()
def test_all_methods(self):
"""
Tests all methods of the SLL class
"""
OPTIONS_HELP = """
Select a method from 1-19:
ℹ️ (1) Create a single liked list (SLL).
ℹ️ (2) Display the SLL.
ℹ️ (3) Count the nodes of SLL.
ℹ️ (4) Search the SLL.
ℹ️ (5) Insert a node at the beginning of the SLL.
ℹ️ (6) Insert a node at the end of the SLL.
ℹ️ (7) Insert a node after a specified node of the SLL.
ℹ️ (8) Insert a node before a specified node of the SLL.
ℹ️ (9) Delete the first node of SLL.
ℹ️ (10) Delete the last node of the SLL.
ℹ️ (11) Delete a node you wish to remove.
ℹ️ (12) Reverse the SLL.
ℹ️ (13) Bubble sort the SLL by only exchanging the integer values.
ℹ️ (14) Bubble sort the SLL by exchanging links.
ℹ️ (15) Merge sort the SLL.
ℹ️ (16) Insert a cycle in the SLL.
ℹ️ (17) Detect if the SLL has a cycle.
ℹ️ (18) Remove cycle in the SLL.
ℹ️ (19) Test merging two bubble-sorted SLLs.
ℹ️ (20) Concatenate a second list to the SLL.
ℹ️ (21) Exit.
"""
self.create_single_linked_list()
while True:
print(OPTIONS_HELP)
UI_OPTION = int(input(" Enter an integer for the method you wish to run using the above help: "))
if UI_OPTION == 1:
data = int(input(" Enter an integer to be inserted at the end of the list: "))
x = int(input(" Enter an integer to be inserted after that: "))
self.insert_node_after_another(data, x)
elif UI_OPTION == 2:
self.display_sll()
elif UI_OPTION == 3:
self.count_sll_nodes()
elif UI_OPTION == 4:
data = int(input(" Enter an integer to be searched: "))
self.search_sll_nodes(data)
elif UI_OPTION == 5:
data = int(input(" Enter an integer to be inserted at the beginning: "))
self.insert_node_in_beginning(data)
elif UI_OPTION == 6:
data = int(input(" Enter an integer to be inserted at the end: "))
self.insert_node_at_end(data)
elif UI_OPTION == 7:
data = int(input(" Enter an integer to be inserted: "))
x = int(input(" Enter an integer to be inserted before that: "))
self.insert_node_before_another(data, x)
elif UI_OPTION == 8:
data = int(input(" Enter an integer for the node to be inserted: "))
k = int(input(" Enter an integer for the position at which you wish to insert the node: "))
self.insert_node_before_another(data, k)
elif UI_OPTION == 9:
self.delete_sll_first_node()
elif UI_OPTION == 10:
self.delete_sll_last_node()
elif UI_OPTION == 11:
data = int(input(" Enter an integer for the node you wish to remove: "))
self.delete_a_node(data)
elif UI_OPTION == 12:
self.reverse_sll()
elif UI_OPTION == 13:
self.bubble_sort_sll_nodes_data()
elif UI_OPTION == 14:
self.bubble_sort_sll()
elif UI_OPTION == 15:
self.merge_sort_sll()
elif UI_OPTION == 16:
data = int(input(" Enter an integer at which a cycle has to be formed: "))
self.insert_cycle_in_sll(data)
elif UI_OPTION == 17:
if self.sll_has_cycle():
print(" The linked list has a cycle. ")
else:
print(" YAAAY! The linked list does not have a cycle. ")
elif UI_OPTION == 18:
self.remove_cycle_from_sll()
elif UI_OPTION == 19:
self.test_merge_using_new_list_and_inplace()
elif UI_OPTION == 20:
list2 = self.create_single_linked_list()
self.concat_second_list_to_sll(list2)
elif UI_OPTION == 21:
break
else:
print(" Option must be an integer, between 1 to 21.")
print()
if __name__ == '__main__':
# Instantiates a new SLL object
SLL_OBJECT = SingleLinkedList()
SLL_OBJECT.test_all_methods()
- 26,487
- 10
- 35
- 65
Expanding Nick Stinemates's answer
class Node(object):
def __init__(self):
self.data = None
self.next = None
class LinkedList:
def __init__(self):
self.head = None
def prepend_node(self, data):
new_node = Node()
new_node.data = data
new_node.next = self.head
self.head = new_node
def append_node(self, data):
new_node = Node()
new_node.data = data
current = self.head
while current.next:
current = current.next
current.next = new_node
def reverse(self):
""" In-place reversal, modifies exiting list"""
previous = None
current_node = self.head
while current_node:
temp = current_node.next
current_node.next = previous
previous = current_node
current_node = temp
self.head = previous
def search(self, data):
current_node = self.head
try:
while current_node.data != data:
current_node = current_node.next
return True
except:
return False
def display(self):
if self.head is None:
print("Linked list is empty")
else:
current_node = self.head
while current_node:
print(current_node.data)
current_node = current_node.next
def list_length(self):
list_length = 0
current_node = self.head
while current_node:
list_length += 1
current_node = current_node.next
return list_length
def main():
linked_list = LinkedList()
linked_list.prepend_node(1)
linked_list.prepend_node(2)
linked_list.prepend_node(3)
linked_list.append_node(24)
linked_list.append_node(25)
linked_list.display()
linked_list.reverse()
linked_list.display()
print(linked_list.search(1))
linked_list.reverse()
linked_list.display()
print("Lenght of singly linked list is: " + str(linked_list.list_length()))
if __name__ == "__main__":
main()
I think the implementation below fill the bill quite gracefully.
'''singly linked lists, by Yingjie Lan, December 1st, 2011'''
class linkst:
'''Singly linked list, with pythonic features.
The list has pointers to both the first and the last node.'''
__slots__ = ['data', 'next'] #memory efficient
def __init__(self, iterable=(), data=None, next=None):
'''Provide an iterable to make a singly linked list.
Set iterable to None to make a data node for internal use.'''
if iterable is not None:
self.data, self.next = self, None
self.extend(iterable)
else: #a common node
self.data, self.next = data, next
def empty(self):
'''test if the list is empty'''
return self.next is None
def append(self, data):
'''append to the end of list.'''
last = self.data
self.data = last.next = linkst(None, data)
#self.data = last.next
def insert(self, data, index=0):
'''insert data before index.
Raise IndexError if index is out of range'''
curr, cat = self, 0
while cat < index and curr:
curr, cat = curr.next, cat+1
if index<0 or not curr:
raise IndexError(index)
new = linkst(None, data, curr.next)
if curr.next is None: self.data = new
curr.next = new
def reverse(self):
'''reverse the order of list in place'''
current, prev = self.next, None
while current: #what if list is empty?
next = current.next
current.next = prev
prev, current = current, next
if self.next: self.data = self.next
self.next = prev
def delete(self, index=0):
'''remvoe the item at index from the list'''
curr, cat = self, 0
while cat < index and curr.next:
curr, cat = curr.next, cat+1
if index<0 or not curr.next:
raise IndexError(index)
curr.next = curr.next.next
if curr.next is None: #tail
self.data = curr #current == self?
def remove(self, data):
'''remove first occurrence of data.
Raises ValueError if the data is not present.'''
current = self
while current.next: #node to be examined
if data == current.next.data: break
current = current.next #move on
else: raise ValueError(data)
current.next = current.next.next
if current.next is None: #tail
self.data = current #current == self?
def __contains__(self, data):
'''membership test using keyword 'in'.'''
current = self.next
while current:
if data == current.data:
return True
current = current.next
return False
def __iter__(self):
'''iterate through list by for-statements.
return an iterator that must define the __next__ method.'''
itr = linkst()
itr.next = self.next
return itr #invariance: itr.data == itr
def __next__(self):
'''the for-statement depends on this method
to provide items one by one in the list.
return the next data, and move on.'''
#the invariance is checked so that a linked list
#will not be mistakenly iterated over
if self.data is not self or self.next is None:
raise StopIteration()
next = self.next
self.next = next.next
return next.data
def __repr__(self):
'''string representation of the list'''
return 'linkst(%r)'%list(self)
def __str__(self):
'''converting the list to a string'''
return '->'.join(str(i) for i in self)
#note: this is NOT the class lab! see file linked.py.
def extend(self, iterable):
'''takes an iterable, and append all items in the iterable
to the end of the list self.'''
last = self.data
for i in iterable:
last.next = linkst(None, i)
last = last.next
self.data = last
def index(self, data):
'''TODO: return first index of data in the list self.
Raises ValueError if the value is not present.'''
#must not convert self to a tuple or any other containers
current, idx = self.next, 0
while current:
if current.data == data: return idx
current, idx = current.next, idx+1
raise ValueError(data)
- 21
- 1
My 2 cents
class Node:
def __init__(self, value=None, next=None):
self.value = value
self.next = next
def __str__(self):
return str(self.value)
class LinkedList:
def __init__(self):
self.first = None
self.last = None
def add(self, x):
current = Node(x, None)
try:
self.last.next = current
except AttributeError:
self.first = current
self.last = current
else:
self.last = current
def print_list(self):
node = self.first
while node:
print node.value
node = node.next
ll = LinkedList()
ll.add("1st")
ll.add("2nd")
ll.add("3rd")
ll.add("4th")
ll.add("5th")
ll.print_list()
# Result:
# 1st
# 2nd
# 3rd
# 4th
# 5th
- 751
- 6
- 22
enter code here
enter code here
class node:
def __init__(self):
self.data = None
self.next = None
class linked_list:
def __init__(self):
self.cur_node = None
self.head = None
def add_node(self,data):
new_node = node()
if self.head == None:
self.head = new_node
self.cur_node = new_node
new_node.data = data
new_node.next = None
self.cur_node.next = new_node
self.cur_node = new_node
def list_print(self):
node = self.head
while node:
print (node.data)
node = node.next
def delete(self):
node = self.head
next_node = node.next
del(node)
self.head = next_node
a = linked_list()
a.add_node(1)
a.add_node(2)
a.add_node(3)
a.add_node(4)
a.delete()
a.list_print()
-
You answer an old question which has already has several well received answers and you don't give any explanation. What is the reason of posting your version? Does it have any benefit over the already presented solutions? Or any other added value? Please edit your answer and add some explanation to make your answer more complete. – honk Oct 30 '14 at 17:14
my double Linked List might be understandable to noobies. If you are familiar with DS in C, this is quite readable.
# LinkedList..
class node:
def __init__(self): ##Cluster of Nodes' properties
self.data=None
self.next=None
self.prev=None
class linkedList():
def __init__(self):
self.t = node() // for future use
self.cur_node = node() // current node
self.start=node()
def add(self,data): // appending the LL
self.new_node = node()
self.new_node.data=data
if self.cur_node.data is None:
self.start=self.new_node //For the 1st node only
self.cur_node.next=self.new_node
self.new_node.prev=self.cur_node
self.cur_node=self.new_node
def backward_display(self): //Displays LL backwards
self.t=self.cur_node
while self.t.data is not None:
print(self.t.data)
self.t=self.t.prev
def forward_display(self): //Displays LL Forward
self.t=self.start
while self.t.data is not None:
print(self.t.data)
self.t=self.t.next
if self.t.next is None:
print(self.t.data)
break
def main(self): //This is kind of the main
function in C
ch=0
while ch is not 4: //Switch-case in C
ch=int(input("Enter your choice:"))
if ch is 1:
data=int(input("Enter data to be added:"))
ll.add(data)
ll.main()
elif ch is 2:
ll.forward_display()
ll.main()
elif ch is 3:
ll.backward_display()
ll.main()
else:
print("Program ends!!")
return
ll=linkedList()
ll.main()
Though many more simplifications can be added to this code, I thought a raw implementation would me more grabbable.
- 620
- 4
- 21
- 21
- 2
Current Implementation of Linked List in Python requires for creation of a separate class, called Node, so that they can be connected using a main Linked List class. In the provided implementation, the Linked List is created without defining a separate class for a node. Using the proposed implementation, Linked Lists are easier to understand and can be simply visualized using the print function.
class Linkedlist:
def __init__(self):
self.outer = None
def add_outermost(self, dt):
self.outer = [dt, self.outer]
def add_innermost(self, dt):
p = self.outer
if not p:
self.outer = [dt, None]
return
while p[1]:
p = p[1]
p[1] = [dt, None]
def visualize(self):
p = self.outer
l = 'Linkedlist: '
while p:
l += (str(p[0])+'->')
p = p[1]
print(l + 'None')
ll = Linkedlist()
ll.add_innermost(8)
ll.add_outermost(3)
ll.add_outermost(5)
ll.add_outermost(2)
ll.add_innermost(7)
print(ll.outer)
ll.visualize()
If you want to just create a simple liked list then refer this code
l=[1,[2,[3,[4,[5,[6,[7,[8,[9,[10]]]]]]]]]]
for visualize execution for this cod Visit http://www.pythontutor.com/visualize.html#mode=edit