London | 26-Jul-SDC | Roman Sanaye | Sprint 2 | Implement LRU Cache

[RomanSanaye]

Learners, PR Template

Self checklist

• I have titled my PR with Region | Cohort | FirstName LastName | Sprint | Assignment Title
• My changes meet the requirements of the task
• I have tested my changes
• My changes follow the style guide

Task code

CYF-1181

Changelist

• Implemented an O(1) worst-case time complexity LRU Cache using a Python dictionary for instant lookups and a Doubly Linked List for tracking usage order.
• Added strict limit validation in the constructor to raise a ValueError for zero or negative limits.
• Handled edge cases including single-node lists, middle-node re-linking, and tail eviction.
• Verified implementation against the provided test suite (lru_cache_test.py) with all tests passing successfully.

Questions

No questions—all unit tests passed successfully!

[cjyuan]

To better adhere to the Single-Responsibility Principle (SRP) from SOLID design principles, it's preferable to implement the "doubly linked list" and the "LRU Cache" as separate classes, with the linked list used inside LruCache to manage ordering.

You could copy your linked_list.py to this folder.

Alternatively, use a suitable Python built-in class to maintain order.

Could you update your code using one of these approaches?

[RomanSanaye]

Hi @cjyuan ,
Great catch and thanks for pointing out the SOLID principles! I've refactored the code to completely separate the DoublyLinkedList and LruCache into distinct classes. Now, LruCache handles only cache operations and delegates all node ordering/pointer manipulation to the DoublyLinkedList to properly adhere to the Single-Responsibility Principle.

[cjyuan]

Looks good!