Skip to main content

Linked Deep Dictionaries in Python.

Project description

Binder CircleCI Documentation Status License PyPI

LinkedDeepDict

A Lightweight Python library to manage nested dictionaries with parent-child relationships. On top of being a compatible drop-in replcement of the build in dict class, the self replicating default factory makes the creation of complex nested layouts effortless.

Documentation

Click here to read the documentation.

Installation

This is optional, but we suggest you to create a dedicated virtual enviroment at all times to avoid conflicts with your other projects. Create a folder, open a command shell in that folder and use the following command

>>> python -m venv venv_name

Once the enviroment is created, activate it via typing

>>> .\venv_name\Scripts\activate

The library can be installed (either in a virtual enviroment or globally) from PyPI using pip on Python >= 3.7:

>>> pip install linkeddeepdict

Usage

In every case where you'd want to use a dict, you can use a LinkedDeepDict as a drop-in replacement, but on top of what a simple dictionary provides, a LinkedDeepDict is more capable, as it provides a machinery to handle nested layouts. It is basically an ordered defaultdict with a self replicating default factory.

>>> from linkeddeepdict import LinkedDeepDict
>>> data = LinkedDeepDict()

A LinkedDeepDict is essentially a nested default dictionary. Being nested refers to the fact that you can do this:

>>> data['a']['b']['c']['e'] = 1
>>> data['a']['b']['d'] = 2

Notice that the object carves a way up until the last key, without needing to create each level explicitly. What happens is that every time a key is missing in a data, the object creates a new instance, which then is also ready to handle missing keys or data. Accessing nested subdictionaries works in a similar fashion:

>>> data['a']['b']['c']['e']
1

To allow for a more Pythonic feel, it also supports array-like indexing, so that the following operations are valid:

>>> data['a', 'b', 'c', 'e'] = 3
>>> data['a', 'b', 'c', 'e']
3

Of course, this is something that we can easily replicate using pure Python in one line, without the need for fancy stuff:

>>> data = {'a' : {'b' : {'c' : {'e' : 3}, 'd' : 2}}}    

The key point is that we loop over a pure dict instance, we get

>>> [k for k in data.keys()]
['a']    

But if we use a LinkedDeepDict class and the option deep=True when accessing keys, values or items of dictionaries, the following happens:

>>> [k for k in LinkedDeepDict(data).keys(deep=True)]
['e', 'd']    

We can see, that in this case, iteration goes over keys, that actually hold on to some data, and does not return the containers themselves. If we do the same experiment with the values, it shows that the LinkedDeepDict only returns the leafs of the data-tree and the behaviour is fundamentally different:

>>> [k for k in data.values()]
[{'b': {'c': {'e': 3}, 'd': 2}}]    
>>> [k for k in LinkedDeepDict(data).values(deep=True)]
[3, 2]    

It is important, that the call obj.values(deep=True) still returns a generator object, which makes it memory efficient when looping over large datasets.

>>> LinkedDeepDict(data).values(deep=True)
<generator object OrderedDefaultDict.values at 0x0000028F209D54A0>    

Testing

To run all tests, open up a console in the root directory of the project and type the following

>>> python -m unittest

Dependencies

The only dependency is six, to provide basic continuity between major Python versions 2 and 3.

License

This package is licensed under the MIT license.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

linkeddeepdict-1.1.0.tar.gz (11.6 kB view details)

Uploaded Source

Built Distribution

linkeddeepdict-1.1.0-py3-none-any.whl (9.2 kB view details)

Uploaded Python 3

File details

Details for the file linkeddeepdict-1.1.0.tar.gz.

File metadata

  • Download URL: linkeddeepdict-1.1.0.tar.gz
  • Upload date:
  • Size: 11.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.10.11

File hashes

Hashes for linkeddeepdict-1.1.0.tar.gz
Algorithm Hash digest
SHA256 9c65bc34c110ea1a2071a411f6481485d4a5da9da029996ead8aca2c39288609
MD5 e0d31af45ebc5d2c96713b0509615bff
BLAKE2b-256 348f21a617a363e2a9011612f64b2117856affcbc57d7e41e6d1411c084a6d7b

See more details on using hashes here.

File details

Details for the file linkeddeepdict-1.1.0-py3-none-any.whl.

File metadata

File hashes

Hashes for linkeddeepdict-1.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 7c076e7feedeea7356276f1dabc07b94f014349b3aba7ee225a56eb8d9011288
MD5 62f6cc49b00ebd8e207463731298446c
BLAKE2b-256 8f3c202f6aa0d0566d647280ceff9b18b30f49098be7a754c7d24cc3b4231dc8

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page