How should you handle exceptions to maintain API/library reliability?

• A. Expose internal stack traces in error responses.
• B. Do not leak internals; convert to domain-specific errors; ensure cleanup and provide meaningful error codes/messages.
• C. Ignore exceptions and return success.
• D. Duplicate low-level errors to client for transparency.

Answer: (B)

Handling exceptions in APIs and libraries should preserve reliability and provide a stable, understandable surface for clients. The goal is to prevent leaking internal details while giving clients actionable feedback that matches the domain language and supports proper error handling.

Expose no internal stack traces or implementation specifics. Instead, catch errors at the boundary and map them to domain-specific failures that reflect what went wrong in business terms, such as missing resource, invalid input, or insufficient permissions. Return a clear, stable error code and a concise message that helps the caller understand what to do next. Alongside this, ensure proper cleanup: release resources, rollback transactions, and revert partial changes so the system remains consistent even when things go wrong. Keep detailed diagnostics and the original exception loggable on the server for debugging, but only surface a user-friendly surface to the client.

This approach helps clients handle errors deterministically, implement retries or alternatives when appropriate, and keeps the API surface clean and secure. Exposure of internal stack traces, or duplicating low-level errors to the client, can reveal sensitive details, confuse users, and undermine reliability, so they’re avoided. Ignoring exceptions and reporting success is misleading, as it hides failures and erodes trust in the API.