AttribDict

Introduction

AttribDict is an easy to use and easy to read dict, it is more flexible and human readable.

Examples

>>> from attribdict import AttribDict as Dict
>>> _d = {"attr"+str(i): i for i in range(4)}
>>> d = Dict(_d) # create a AttribDict instance from a dict
>>> print(d)
attr0: 0
attr1: 1
attr2: 2
attr3: 3

You can also recursively create an attribute from other data type including dict:

# continue to previous code
>>> d.attr4.subattr1.subsubattr1 = {1, 2, 4} # recursively create attribute
>>> d.attr5.subattr1 = {"subsubattr"+str(i): i for i in range(3)} # recursively create from a dict
>>> print(d)
attr0: 0
attr1: 1
attr2: 2
attr3: 3
attr4:
    - subattr1:
        - subsubattr1: {1, 2, 4}
attr5:
    - subattr1:
        - subsubattr0: 0
        - subsubattr1: 1
        - subsubattr2: 2

Additionally, AttribDict also support create attributes from a recursively dict:

# continue to previous code
>>> _d = {"_attr1": {"_subattr1": {"_subsubattr1": "hello attribdict"}}} # create attributes from a recursively dict
>>> d._attr = _d
>>> print(d)
attr0: 0
attr1: 1
attr2: 2
attr3: 3
attr4:
    - subattr1:
        - subsubattr1: {1, 2, 4}
attr5:
    - subattr1:
        - subsubattr0: 0
        - subsubattr1: 1
        - subsubattr2: 2
_attr:
    - _attr1:
        - _subattr1:
            - _subsubattr1: hello attribdict

If you want to convert an AttribDict instance into a python dict, just use .as_dict method:

# continue to previous code
>>> _dict = d.as_dict()
>>> print(_dict)
{'attr0': 0, 'attr1': 1, 'attr2': 2, 'attr3': 3, 'attr4': {'subattr1': {'subsubattr1': {1, 2, 4}}}, 'attr5': {'subattr1': {'subsubattr0': 0, 'subsubattr1': 1, 'subsubattr2': 2}}, '_attr': {'_attr1': {'_subattr1': {'_subsubattr1': 'hello attribdict'}}}}

You can also access value in traditional dict way, for example:

>>> print(d["_attr"].as_dict())
{'_attr1': {'_subattr1': {'_subsubattr1': 'hello attribdict'}}}

Please feel free if you want to access value in following style:

>>> print(d._attr["_attr1"]._subattr1["_subsubattr1"])
hello attribdict

AttribDict instance is iterable, thus you can use for statement and so on to iteratively visit the attributes and corresponding values. Note that it will return (attribute, value) pair, sub-AttribDict will be converted to a python dict for more convenient use:

# continue to previous code
>>> for key, value in d:
...     print("key: ", key)
...     print("value: ", value)
...
key:  attr0
value:  0
key:  attr1
value:  1
key:  attr2
value:  2
key:  attr3
value:  3
key:  attr4
value:  {'subattr1': {'subsubattr1': {1, 2, 4}}}
key:  attr5
value:  {'subattr1': {'subsubattr0': 0, 'subsubattr1': 1, 'subsubattr2': 2}}
key:  _attr
value:  {'_attr1': {'_subattr1': {'_subsubattr1': 'hello attribdict'}}}

Here is another example about iterable, the rest code is omitted:

>>> it = iter(d)
>>> next(it)
('attr0', 0)
>>> next(it)
('attr1', 1)
>>> next(it)
('attr2', 2) 

AttribDict offers two ways to copy and deepcopy an instance:

>>> import copy
>>> d_copy = d.copy()
>>> print(d._attr is d_copy._attr)
True
>>> copy_d = copy.copy(d)
>>> print(d._attr is copy_d._attr)
>>> True
>>> d_deepcopy = d.deepcopy()
>>> print(d._attr is d_deepcopy._attr)
False
>>> print(d.attr4.subattr1.subsubattr1 is d_deepcopy.attr4.subattr1.subsubattr1)
False
>>> deepcopy_d = copy.deepcopy(d)
>>> print(d._attr is deepcopy_d._attr)
>>> False
>>> print(d.attr4.subattr1.subsubattr1 is deepcopy_d.attr4.subattr1.subsubattr1)
False

If you want to check whether an object contains attribute attrX, please use obj.hasattr("attrX") but hasattr(obj, "attrX") as the latter one will create a new AttribDict instance named attrX.

>>> d.hasattr("attrX")
False
>>> hasattr(d, "attrX")
True

Pickling AttribDict instance.

>>> import pickle
>>> with open("path2file", "wb") as fp:
>>> ... pickle.dump(d, fp)
>>> ...
>>> with open("path2file", "rb") as fp:
>>> ... loaded_d = pickle.load(fp)
>>> ...
>>> loaded_d == d
True
>>> loaded_d is d
False

Installation

You can install AttribDict by pip.

pip install attribdict