What for dinner - lab 2 - variant 3

This is an example project which demonstrates project structure and necessary CI checks. It is not the best structure for real-world projects, but good enough for educational purposes.

Project structure

• BinaryTree.py -- implementation of BinaryTreeNode class.

• BinaryTree_test.py -- unit and PBT tests for BinaryTreeNode

Features

• PBT: test_add_commutative

Contribution

• Lu Bin ([email protected]) -- All work.
• Wang Yining ([email protected]) -- All work.

Changelog

• 26.05.2024 - 7
  • Update README.
  • Fix bugs in code.
• 24.05.2024 - 6
  • Fix bugs in code.
• 17.05.2024 - 5
  • Fix bugs in code.
• 10.05.2024 - 4
  • Implement PBT test.
• 05.05.2024 - 3
  • Implement unit test.
• 29.03.2022 - 2
  • Add test coverage.
• 29.03.2022 - 1
  • Update README.
• 29.05.2024 - 0
  • Initial.

Design notes

• Compare mutable and immutable implementation:
  • Variable programming allows the value of variables to change while the code is running,making programming more intuitive and achieving higher performance.
  • However, in a multi-threaded or concurrent environment, variable states can easily lead to race conditions and data inconsistency issues.
  • In immutable programming, once a variable is assigned a value, it cannot be changed. Each modification requires a new data structure, which is more suitable for multi-threaded or concurrent environments, and there is no need to worry about data race conditions.
  • However, every time the data is modified, a new copy will be created, which may cause more memory consumption and garbage collection pressure.
  • At the same time, it is necessary to change the programming idea when programming, which is more cumbersome.
• Note implementation restriction:
  • In immutable programming, all objects cannot be changed after they are created, and each change requires the generation of a new data structure.
  • Therefore, the copy() function is used repeatedly in the code to build a new data structure based on the original data structure to ensure that the original structure will not change.
• Find possible implementation errors, which can pass property-based tests:
  • When implementing a Binary Tree based set, special conditions are sometimes overlooked. For example, if the implemented function fails to handle empty inputs. Due to the nature of property-based testing (PBT), automatically generated inputs may not include empty inputs. Consequently, the test may pass, giving the impression that the function is problem-free. However, when we manually introduce marginal conditions such as empty lists or empty trees, the test will fail.