How do I practice coding and algorithm questions?

Use PracHub's coding console to write, test, and debug your solutions in Python or JavaScript. View hints, test against sample inputs, and compare with official solutions.

What difficulty level is this coding question?

This is a hard difficulty Coding & Algorithms question, commonly asked during Technical Screen rounds at OpenAI.

What role is this question designed for?

This question is commonly asked for Software Engineer candidates at OpenAI during technical interviews.

Refactor a chat message processing function

Quick Overview

This question evaluates a candidate's ability to refactor and implement clear, maintainable message-processing logic—covering code organization, command parsing, string trimming, state mutation, and edge-case handling in an in-memory chat store, and it tests competencies in coding, algorithms, and software engineering within the Coding & Algorithms category. It is commonly asked because it exposes a developer's capacity to produce deterministic, testable, and linear-time behavior while improving readability and maintainability in real-world backend message workflows, and its level of abstraction is primarily practical application with some conceptual design considerations.

Refactor a chat message processing function

Company: OpenAI

Role: Software Engineer

Category: Coding & Algorithms

Difficulty: hard

Interview Round: Technical Screen

## Task You are given a poorly written JavaScript function that processes an **incoming chat message** and updates an in-memory store. The code is hard to read, mixes concerns, and barely passes a small test. Your job is to **refactor and improve** it while preserving behavior and making it easier to maintain. ## Requirements Implement (or refactor toward) a clean version of: - `processIncomingMessage(store, input)` Where: - `store` is an in-memory object: - `store.messages`: array of stored messages - `store.userStatus`: map `{ [userId]: 'online' | 'away' | 'meeting' }` - `store.userTacoCount`: map `{ [userId]: number }` - `input` is: - `{ userId: string, text: string, ts: number }` The function must: 1. Trim the message text and ignore it if empty. 2. If the message starts with a **slash command**, handle these commands: - `/away` → set `store.userStatus[userId] = 'away'` and store a system message (role=`'system'`) like: `"<userId> is now away"`. - `/meet <topic>` → set status to `'meeting'` and store a system message: `"<userId> is in a meeting: <topic>"`. If `<topic>` is missing, use `"(no topic)"`. - `/taco` → increment `store.userTacoCount[userId]` and store a system message: `"<userId> sent a taco"`. 3. Otherwise, store it as a normal chat message with role=`'user'`. 4. Return the updated `store` (mutating is allowed, but your refactor should be clear and consistent). ## Expectations - Refactor for readability: extract helpers, avoid deeply nested conditionals, use clear naming. - Make behavior deterministic and easy to test. - Keep time complexity linear in `text.length`. ## Example Input: `{ userId: 'u1', text: ' /meet design review ', ts: 10 }` - Status becomes `meeting` - A system message is appended. Input: `{ userId: 'u1', text: 'hello', ts: 11 }` - A user message is appended. (You may assume a small provided unit test suite will validate exact stored messages and status updates.)

Quick Answer: This question evaluates a candidate's ability to refactor and implement clear, maintainable message-processing logic—covering code organization, command parsing, string trimming, state mutation, and edge-case handling in an in-memory chat store, and it tests competencies in coding, algorithms, and software engineering within the Coding & Algorithms category. It is commonly asked because it exposes a developer's capacity to produce deterministic, testable, and linear-time behavior while improving readability and maintainability in real-world backend message workflows, and its level of abstraction is primarily practical application with some conceptual design considerations.

Implement `solution(store, input)`, a clean and testable version of a chat message processor. `store` is a dictionary with: - `store['messages']`: a list of stored message objects - `store['userStatus']`: a map from `userId` to `'online'`, `'away'`, or `'meeting'` - `store['userTacoCount']`: a map from `userId` to an integer taco count `input` is a dictionary with: - `userId`: string - `text`: string - `ts`: integer timestamp Process the incoming message as follows: 1. Trim the message text. If it becomes empty, ignore it and return the store unchanged. 2. If the trimmed text is one of the supported slash commands, handle it: - `/away`: - set `store['userStatus'][userId] = 'away'` - append a system message with text `"<userId> is now away"` - `/meet <topic>`: - set `store['userStatus'][userId] = 'meeting'` - append a system message with text `"<userId> is in a meeting: <topic>"` - if the topic is missing or blank, use `"(no topic)"` - `/taco`: - increment `store['userTacoCount'][userId]` by 1, starting from 0 if needed - append a system message with text `"<userId> sent a taco"` 3. Otherwise, append a normal user message. Every appended message must be a dictionary with exactly these keys: - `role`: `'system'` or `'user'` - `userId`: the sender's user id - `text`: the stored message text - `ts`: the original timestamp Only `/away`, `/meet`, and `/taco` are special. Any other text, including unknown slash-prefixed text like `/shrug`, must be stored as a normal user message after trimming. You may mutate `store` in place, but return the updated `store`.

Constraints

0 <= len(input['text']) <= 10^5
store always contains the keys 'messages', 'userStatus', and 'userTacoCount'
userId is a non-empty string
Only '/away', '/meet', and '/taco' are treated as commands
Your processing should run in O(n) time where n is the length of the message text

Examples

Input: ({'messages': [], 'userStatus': {}, 'userTacoCount': {}}, {'userId': 'u1', 'text': ' /meet design review ', 'ts': 10})

Expected Output: {'messages': [{'role': 'system', 'userId': 'u1', 'text': 'u1 is in a meeting: design review', 'ts': 10}], 'userStatus': {'u1': 'meeting'}, 'userTacoCount': {}}

Explanation: After trimming, this is a /meet command with topic 'design review'. Status becomes 'meeting' and a system message is appended.

Input: ({'messages': [], 'userStatus': {'u1': 'online'}, 'userTacoCount': {}}, {'userId': 'u1', 'text': ' hello team ', 'ts': 11})

Expected Output: {'messages': [{'role': 'user', 'userId': 'u1', 'text': 'hello team', 'ts': 11}], 'userStatus': {'u1': 'online'}, 'userTacoCount': {}}

Explanation: This is a normal chat message. The text is trimmed and stored with role 'user'.

Input: ({'messages': [{'role': 'user', 'userId': 'u2', 'text': 'hi', 'ts': 1}], 'userStatus': {'u1': 'online'}, 'userTacoCount': {'u1': 2}}, {'userId': 'u1', 'text': '/away', 'ts': 12})

Expected Output: {'messages': [{'role': 'user', 'userId': 'u2', 'text': 'hi', 'ts': 1}, {'role': 'system', 'userId': 'u1', 'text': 'u1 is now away', 'ts': 12}], 'userStatus': {'u1': 'away'}, 'userTacoCount': {'u1': 2}}

Explanation: The /away command updates the user's status and appends a system message, while preserving existing messages and taco counts.

Input: ({'messages': [], 'userStatus': {}, 'userTacoCount': {}}, {'userId': 'chef', 'text': '/taco', 'ts': 7})

Expected Output: {'messages': [{'role': 'system', 'userId': 'chef', 'text': 'chef sent a taco', 'ts': 7}], 'userStatus': {}, 'userTacoCount': {'chef': 1}}

Explanation: The /taco command increments the sender's taco count from 0 to 1 and stores a system message.

Input: ({'messages': [], 'userStatus': {}, 'userTacoCount': {}}, {'userId': 'u3', 'text': ' /meet ', 'ts': 20})

Expected Output: {'messages': [{'role': 'system', 'userId': 'u3', 'text': 'u3 is in a meeting: (no topic)', 'ts': 20}], 'userStatus': {'u3': 'meeting'}, 'userTacoCount': {}}

Explanation: The /meet command has no topic after trimming, so '(no topic)' is used.

Input: ({'messages': [], 'userStatus': {}, 'userTacoCount': {}}, {'userId': 'u9', 'text': ' /shrug ', 'ts': 33})

Expected Output: {'messages': [{'role': 'user', 'userId': 'u9', 'text': '/shrug', 'ts': 33}], 'userStatus': {}, 'userTacoCount': {}}

Explanation: Only /away, /meet, and /taco are special. Unknown slash-prefixed text is stored as a normal user message.

Input: ({'messages': [{'role': 'user', 'userId': 'u1', 'text': 'x', 'ts': 1}], 'userStatus': {'u1': 'online'}, 'userTacoCount': {'u1': 1}}, {'userId': 'u1', 'text': ' ', 'ts': 30})

Expected Output: {'messages': [{'role': 'user', 'userId': 'u1', 'text': 'x', 'ts': 1}], 'userStatus': {'u1': 'online'}, 'userTacoCount': {'u1': 1}}

Explanation: After trimming, the message is empty, so the store is returned unchanged.

Hints

Trim the text once at the start and return early if nothing remains.
A small helper for appending message objects can remove repeated code and make command handling clearer.

Last updated: Apr 25, 2026

Loading coding console...

Quick Overview

Refactor a chat message processing function

Company: OpenAI

Role: Software Engineer

Category: Coding & Algorithms

Difficulty: hard

Interview Round: Technical Screen

Constraints

0 <= len(input['text']) <= 10^5
store always contains the keys 'messages', 'userStatus', and 'userTacoCount'
userId is a non-empty string
Only '/away', '/meet', and '/taco' are treated as commands
Your processing should run in O(n) time where n is the length of the message text