WordCollection Class

Word lookup based on a Patricia tree (a.k.a. Radix Tree, a.k.a. Trie data structure). This data structure is efficiently searchable by the prefix of words. Such a prefix search takes a string prefix, and returns all dictionary words that begin with that prefix.

This class ingests rank/word pair files in a given directory. The ranks are intended to be relative usage frequencies. The class manages these frequency ranks.

Public methods:

• add(word)

• contains(word)

• prefix_search(prefix)

• rank(word)

WordCollection can be subclassed.

TelPadEncodedWordCollection Subclass

Instances behave as the superclass WordCollection. However, all added words are encoded as if entered via a telephone pad. Each letter group of the telephone pad is represented by its first letter. Example: "and" --> "amd" (phone buttons 1,5,2).

The class can thus be used to search words by entering for each real word's letters 'c' the first letter of the telephone pad that contains 'c'. For word input, clients need not concern themselves with this encoding. That transformation occurs automatically.

However, calls to search_prefix() or contains() must encode the real words with the encoded version. Thus, instead of calling myColl.contains("and"), the client would call myColl.contains("amd"). Method encodeWord() takes a real word and returns the encoded version.

Method search_prefix() will usually contain a larger number of 'remaining possible words' than a regular #WordCollection. This is because the mapping from encoded to real words is one-to-many.