Explain Existing Codebase and Extend It | Dataford Interview Questions

Given an existing codebase, explain how it works and its architecture/flow, then implement an additional feature from a problem statement and discuss extensions.

Asked in the live coding / codebase walkthrough stage. For the implementation portion, treat the codebase as a directed module dependency graph and return every module affected by one or more changed modules.

Input and Output

graph maps each module to the modules it directly imports. changed is a list of changed module names. Return all changed modules and their direct or indirect dependents, excluding unknown modules, sorted lexicographically.

Example: graph = {"app": ["auth"], "auth": ["config"], "config": []}, changed = ["config"] returns ["app", "auth", "config"].

Discuss how the design could change for streaming updates, cycles, or very large codebases.

Constraints

1 <= number of modules <= 5000
Each module name is a unique non-empty string
Every dependency list may contain zero or more module names
Unknown dependencies and changed module names are ignored
The returned list must be sorted lexicographically

Function Signature

def affected_modules(graph, changed):

Your question is Explain Existing Codebase and Extend It. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem