Insert function for bst not working - python-3.x

Question

If anyone can help me with the recursion part?
I omitted some part thnking that it was not necessary but after that function didn't work.
class node(object):
def __init__(self,value):
self.data=value
self.left=None
self.right=None
def insert(Node,value):
if Node is None:
Node=node(value)
else:
if value<Node.data:
## if Node.left is None: Node.left=node(value)
## else: insert(Node.left,value)
insert(Node.left,value)
else:
## if Node.left is None: Node.left=node(value)
## else: insert(Node.left,value)
insert(Node.right,value)

Answer 1

Because you'll need to create new node instances for your empty branches.
Let's see what happens with both definitions, printing out the values of the Node parameter:
def insert(Node, value):
print(Node)
if Node is None:
Node=node(value)
else:
if value<Node.data:
insert(Node.left,value)
else:
insert(Node.right,value)
On the REPL:
In [17]: badTree = node(10)
In [18]: badTree.data
Out[18]: 10
In [20]: badTree.left.data
----------------------------------------------------------------------
AttributeError Traceback (most recent call last)
<ipython-input-20-3cb52def2967> in <module>()
----> 1 badTree.left.data
AttributeError: 'NoneType' object has no attribute 'data'
In [21]: badTree.right.data
----------------------------------------------------------------------
AttributeError Traceback (most recent call last)
<ipython-input-21-b6f1267c9d29> in <module>()
----> 1 badTree.right.data
AttributeError: 'NoneType' object has no attribute 'data'
Let's see what happens if we insert an element to badTree:
In [22]: insert(badTree, 2)
<__main__.node object at 0x7f706bf34d30>
None
In [23]: badTree.left.data
----------------------------------------------------------------------
AttributeError Traceback (most recent call last)
<ipython-input-23-3cb52def2967> in <module>()
----> 1 badTree.left.data
AttributeError: 'NoneType' object has no attribute 'data'
It fails to insert the element, but why? In badTree, both left and right branches are None (empty), what means that we need to create new trees there as we won't be able to recursively add new values, as we lose the reference to the main Node. This is, if we call insert(Node.left, value) before assigning a new object to Node.left, it will be equivalent to call insert(None, value) (this the value None, you can see it when we called insert on badTree), which will call recursively to insert and execute None = node(value) which will do nothing.
What happens if we use this definition instead:
def insert(Node, value):
print(Node)
if Node is None:
Node = node(value)
else:
if value < Node.data:
if Node.left is None:
Node.left = node(value)
else:
insert(Node.left, value)
else:
if Node.right is None:
Node.right = node(value)
else:
insert(Node.right, value)
On the REPL:
In [25]: newTree = node(4)
In [26]: insert(newTree, 10)
<__main__.node object at 0x7f706bf30668>
In [27]: insert(newTree, 12)
<__main__.node object at 0x7f706bf30668>
<__main__.node object at 0x7f706bf30a58>
Now see that it doesn't fail anymore, as we are keeping track of the memory addresses in the tree nodes and thus we can reference those objects (trees) which will let us do the insertions in-place.