Google Autocomplete Project

Dini Deutsch & Rivki Levy

Description

The goal of the project was to provide an autocomplete service for a huge data source. This is done by a system that can run win two different modes:

• ingestion mode - the program will read the data and store it in a file, formed as a trie of the words in the resource with their filepath and some more relevant information so runtime will be short and retrieving the data will be efficient.
• query mode - the program will read the trie file into an actual trie and then waits for user input. When it gets an input it search until it gets five matches. if less then five returned it will try to search for completions with one letter missed\replaced\added.

Usage

within a project

first edit the mode to ingestion/query by sending it through the argv.

then use a code like this:

import Main
import sys
main = Main(sys.argv[1])
main.run("some_file.txt", "some_resource.txt")

by cmd

python project_path (mode = query/ingestion)