Trie object in python using hash tables for keeping track of children

optimized_trie.py

class Trie:
    """
    Trie object will try to optimize by keeping unseen suffixes in a single node,
    and then unravels those suffixes when the letters in the suffix are encountered
    """
    class Node:
        """Each trie node"""
        def __init__(self):
            self.children = {}
            self.count = 1
            self.suffix = None
            self.is_suffix = False

        def set_suffix(self, suffix):
            self.suffix = suffix
            self.is_suffix = True

        def unset_suffix(self):
            _t = self.suffix
            self.suffix = None
            self.is_suffix = False
            return _t

    def __init__(self):
        self.root = Trie.Node()

    def add(self, value):
        node = self.root
        for i, c in enumerate(value):
            # If the suffix is not in the trie already
            # just add entire suffix in a single node
            if c not in node.children:
                # Create a suffix type of node, containing multiple characters
                node.children[c] = Trie.Node()
                node.children[c].set_suffix(value[i+1:])
                return
            elif node.children[c].is_suffix:  # Reached a suffix node
                # Unravel the suffix
                _trie = Trie()
                _trie.root = node.children[c]
                _trie.add(node.children[c].unset_suffix())
            node = node.children[c]
            node.count += 1

    def get_count(self, value):
        node = self.root
        for c in value:
            # If suffix not in the trie
            if c not in node.children:
                return 0
            elif node.children[c].is_suffix:  # Reached a suffix node
                # Unravel the suffix
                _trie = Trie()
                _trie.root = node.children[c]
                _trie.add(node.children[c].unset_suffix())
            node = node.children[c]
        return node.count