Welcome to treelib’s documentation!¶

Redistributed under Apache License (2.0) since version 1.3.0.

Tree data structure is an important data structure in computer programming languages. It has important applications where hierarchical data connections are present such as computer folder structure and decision-tree algorithm in Machine Learning. Thus treelib is created to provide an efficient implementation of tree data structure in Python.

The main features of treelib includes:

Simple to use in both python 2 and 3.

Efficient operation of node indexing with the benefit of dictionary type.

Support various tree operations like traversing, insertion, deletion, node moving, shallow/deep copying, subtree cutting etc.

Support user-defined data payload to accelerate your model construction.

Has pretty tree showing and text/json dump for pretty show and offline analysis.

Contents:

Install¶

The rapidest way to install treelib is using the package management tools like easy_install or pip with command

$ sudo easy_install -U treelib

or the setup script

$ sudo python setup.py install

Note: With the package management tools, the hosted version may be falling behind current development branch on Github. If you encounter some problems, try the freshest version on Github or open issues to let me know your problem.

Useful APIs¶

This treelib is a simple module containing only two classes: Node and Tree. Tree is a self-contained structure with some nodes and connected by branches. One tree has and only has one root, while a node (except root) has several children and merely one parent.

Note: To solve the string compatibility between Python 2.x and 3.x, treelib follows the way of porting Python 3.x to 2/3. That means, all strings are manipulated as unicode and you do not need u’‘ prefix anymore. The impacted functions include str(), show() and save2file() routines. But if your data contains non-ascii characters and Python 2.x is used, you have to trigger the compatibility by declaring unicode_literals in the code:

>>> from __future__ import unicode_literals

`Node` Objects¶

class treelib.Node([tag[, identifier[, expanded]]])¶: A Node object contains basic properties such as node identifier, node tag, parent node, children nodes etc., and some operations for a node.

Class attributes are:

Node.ADD¶: Addition mode for method update_fpointer().

Node.DELETE¶: Deletion mode for method update_fpointer().

Node.INSERT¶: Behave in the same way with Node.ADD since version 1.1.

Instance attributes:

node.identifier¶: The unique ID of a node within the scope of a tree. This attribute can be accessed and modified with . and = operator respectively.

node.tag¶: The readable node name for human. This attribute can be accessed and modified with . and = operator respectively.

node.bpointer¶: The parent ID of a node. This attribute can be accessed and modified with . and = operator respectively.

node.fpointer¶: With a getting operator, a list of IDs of node’s children is obtained. With a setting operator, the value can be list, set, or dict. For list or set, it is converted to a list type by the package; for dict, the keys are treated as the node IDs.

Instance methods:

node.is_leaf()¶: Check if the node has children. Return False if the fpointer is empty or None.

node.is_root()¶: Check if the node is the root of present tree.

node.update_bpointer(nid)¶: Set the parent (indicated by the nid parameter) of a node.

node.update_fpointer(nid, mode=Node.ADD)¶: Update the children list with different modes: addition (Node.ADD or Node.INSERT) and deletion (Node.DELETE).

`Tree` Objects¶

class node.Tree(tree=None, deep=False)¶: The Tree object defines the tree-like structure based on Node objects. A new tree can be created from scratch without any parameter or a shallow/deep copy of another tree. When deep=True, a deepcopy operation is performed on feeding tree parameter and more memory is required to create the tree.

Class attributes are:

Tree.ROOT¶: Default value for the level parameter in tree’s methods.

Tree.DEPTH¶: The depth-first search mode for tree.

Tree.WIDTH¶: The width-first search mode for tree.

Tree.ZIGZAG¶: The ZIGZAG search mode for tree.

Instance attributes:

tree.root¶: Get or set the ID of the root. This attribute can be accessed and modified with . and = operator respectively.

Instance methods:

tree.size()¶: Get the number of nodes in this tree.

tree.contains(nid)¶: Check if the tree contains given node.

tree.parent(nid)¶: Obtain specific node’s parent (Node instance). Return None if the parent is None or does not exist in the tree.

tree.all_nodes()¶: Get the list of all the nodes randomly belonging to this tree.

tree.depth()¶: Get depth of the tree.

tree.leaves(nid)¶: Get leaves from given node.

tree.add_node(node[, parent])¶: Add a new node object to the tree and make the parent as the root by default.

tree.create_node(tag[, identifier[, parent]])¶: Create a new node and add it to this tree.

tree.expand_tree([nid[, mode[, filter[, key[, reverse]]]]]])¶: Traverse the tree nodes with different modes. nid refers to the expanding point to start; mode refers to the search mode (Tree.DEPTH, Tree.WIDTH); filter refers to the function of one variable to act on the Node object; key, reverse are present to sort :class:Node objects at the same level.

tree.get_node(nid)¶: Get the object of the node with ID of nid An alternative way is using ‘[]’ operation on the tree. But small difference exists between them: the get_node() will return None if nid is absent, whereas ‘[]’ will raise KeyError.

tree.is_branch(nid)¶: Get the children (only sons) list of the node with ID == nid.

tree.siblings(nid)¶: Get all the siblings of given nid.

tree.move_node(source, destination)¶: Move node (source) from its parent to another parent (destination).

tree.paste(nid, new_tree)¶: Paste a new tree to an existing tree, with nid becoming the parent of the root of this new tree.

tree.remove_node(nid)¶: Remove a node and free the memory along with its successors.

tree.link_past_node(nid)¶: Remove a node and link its children to its parent (root is not allowed).

tree.rsearch(nid[, filter])¶: Search the tree from nid to the root along links reservedly. Parameter filter refers to the function of one variable to act on the Node object.

tree.show([nid[, level[, idhidden[, filter[, key[, reverse[, line_type]]]]]]]])¶

Print the tree structure in hierarchy style. nid refers to the expanding point to start; level refers to the node level in the tree (root as level 0); idhidden refers to hiding the node ID when printing; filter refers to the function of one variable to act on the Node object; key, reverse are present to sort Node object in the same level.

You have three ways to output your tree data, i.e., stdout with show(), plain text file with save2file(), and json string with to_json(). The former two use the same backend to generate a string of tree structure in a text graph.

Version >= 1.2.7a: you can also spicify the line_type parameter (now: supporting ‘ascii’ [default], ‘ascii-ex’, ‘ascii-exr’, ‘ascii-em’, ‘ascii-emv’, ‘ascii-emh’) to the change graphical form.

tree.subtree(nid)¶: Return a soft copy of the subtree with nid being the root. The softness means all the nodes are shared between subtree and the original.

tree.remove_subtree(nid)¶: Return a subtree with nid being the root, and remove all nodes in the subtree from the original one.

tree.save2file(filename[, nid[, level[, idhidden[, filter[, key[, reverse]]]]]]])¶: Save the tree into file for offline analysis.

tree.to_json()¶: To format the tree in a JSON format.

Examples¶

Basic Usage¶

>>> from treelib import Node, Tree
>>> tree = Tree()
>>> tree.create_node("Harry", "harry")  # root node
>>> tree.create_node("Jane", "jane", parent="harry")
>>> tree.create_node("Bill", "bill", parent="harry")
>>> tree.create_node("Diane", "diane", parent="jane")
>>> tree.create_node("Mary", "mary", parent="diane")
>>> tree.create_node("Mark", "mark", parent="jane")
>>> tree.show()
Harry
├── Bill
└── Jane
    ├── Diane
    │   └── Mary
    └── Mark

API Examples¶

Example 1: Expand a tree with specific mode (Tree.DEPTH [default], Tree.WIDTH, Tree.ZIGZAG).

>>> print(','.join([tree[node].tag for node in \
            tree.expand_tree(mode=Tree.DEPTH)]))
Harry,Bill,Jane,Diane,Mary,Mark

Example 2: Expand tree with custom filter.

>>> print(','.join([tree[node].tag for node in \
            tree.expand_tree(filter = lambda x: \
            x.identifier != 'diane')]))
Harry,Bill,Jane,Mark

Example 3: Get a subtree with the root of ‘diane’.

>>> sub_t = tree.subtree('diane')
>>> sub_t.show()
Diane
└── Mary

Example 4: Paste a new tree to the original one.

>>> new_tree = Tree()
>>> new_tree.create_node("n1", 1)  # root node
>>> new_tree.create_node("n2", 2, parent=1)
>>> new_tree.create_node("n3", 3, parent=1)
>>> tree.paste('bill', new_tree)
>>> tree.show()
Harry
├── Bill
│   └── n1
│       ├── n2
│       └── n3
└── Jane
    ├── Diane
    │   └── Mary
    └── Mark

Example 5: Remove the existing node from the tree

>>> tree.remove_node(1)
>>> tree.show()
Harry
├── Bill
└── Jane
    ├── Diane
    │   └── Mary
    └── Mark

Example 6: Move a node to another parent.

>>> tree.move_node('mary', 'harry')
>>> tree.show()
Harry
├── Bill
├── Jane
│   ├── Diane
│   └── Mark
└── Mary

Example 7: Get the height of the tree.

>>> tree.depth()
2

Example 8: Get the level of a node.

>>> node = tree.get_node("bill")
>>> tree.depth(node)
1

Example 9: Print or dump tree structure. For example, the same tree in: basic example can be printed with ‘ascii-em’:

>>> tree.show(line_type="ascii-em")
Harry
╠══ Bill
╠══ Jane
║   ╠══ Diane
║   ╚══ Mark
╚══ Mary

In the JSON form, to_json() takes optional parameter with_data to trigger if the data field is appended into JSON string. For example,

>>> print(tree.to_json(with_data=True))
{"Harry": {"data": null, "children": [{"Bill": {"data": null}}, {"Jane": {"data": null, "children": [{"Diane": {"data": null}}, {"Mark": {"data": null}}]}}, {"Mary": {"data": null}}]}}

Advanced Usage¶

Sometimes, you need trees to store your own data. The newsest version of treelib supports .data variable to store whatever you want. For example, to define a flower tree with your own data:

>>> class Flower(object): \
        def __init__(self, color): \
            self.color = color

You can create a flower tree now:

>>> ftree = Tree()
>>> ftree.create_node("Root", "root")
>>> ftree.create_node("F1", "f1", parent='root', data=Flower("white"))
>>> ftree.create_node("F2", "f2", parent='root', data=Flower("red"))

Notes: Before version 1.2.5, you may need to inherit and modify the behaviors of tree. Both are supported since then. For flower example,

>>> class FlowerNode(treelib.Node): \
        def __init__(self, color): \
            self.color = color
>>> # create a new node
>>> fnode = FlowerNode("white")