Python dictionary and set implemented using prefix trees
This package provides PrefixDict, a dictionary like object, and PrefixSet, set like object, that are implemented using using prefix trees, or tries. Using tries provides the following unique features when compared to the builtin dict and set objects.
- Keys are returned in sorted order.
- Slice operations for getting, setting and deleting values.
Trie based collections are useful when ordered access to key and values is a requirement.
PrefixDict and PrefixSet behave like the builtin dict and set objects. They are implementations of the MutableMapping and MutableSet abstract base classes. They also support the same constructors as the builtins.
>>> from prefixtree import PrefixDict >>> pd = PrefixDict(a=0, b=1) >>> pd['c'] = 2 >>> 'a' in pd True >>> 'd' in pd False
The only incompatible API difference between prefixtree collections and the builtins is that PrefixDict and PrefixSet only support strings as keys. Unicode strings will be encoded to byte strings before and after use.
Unlike the bultins, it’s possible to use slices when getting, setting and deleting values from prefixtree collecionts.
>>> list(pd['a':'c']) [0, 1, 2] >>> pd[:'b'] = [4, 3] >>> list(pd['a':'c':-1]) [2, 3, 4]
PrefixDict also has additional methods not present on builtin dicts.
- commonprefix(key), to find the longest comment prefix with current keys.
- startswith(prefix), iterates over current keys with matching prefix.
Refer to the full prefixtree documentation on Read The Docs for details.
prefixtree is implemented to be compatible with Python 2.x and Python 3.x. It has been tested against the following Python implementations:
- CPython 2.6
- CPython 2.7
- CPython 3.2
- PyPy 1.9.0
Continuous integration testing is provided by Travis CI.
Full documentation for prefixtree is hosted by Read The Docs.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.