This is a repository with implementation and comparison of non-clustered database indexes on Python programming language for the assignment in Advanced Databases course at Innopolis University.
Authors: Ilya Borovik, Artur Khayaliev, Boris Makaev.
The task was to implement three indexes:
Source code of indexes is provided in the package indexes.
Search capabilities of each index are compared with Naive index (simple for-loop search).
As a unit (row) of data in the table we use objects of class Item. Each item has:
- a single value
key - a field for data called
valuewhich does not affect index structures, but used only in index tests.
As a test data for the indexes any lists of items can be passed.
We provide several methods in the file list_generators.py for generating lists of items with integer or string keys. To make tests more determined we also provide a pregenerated file with items in the directory data.
To test the indexes run the file index_test.py using a Python (>=3.6) interpreter.
Default run includes construction of indexes for a table in the defined above file,
search, insertion and deletion tests.
To change the configuration of tests, please, modify the code in the file index_test.py. Comments in the file display other test options.