Course: Network Programming (INT1433) · PTIT · Instructor: Dr. Hung N. Dang
One-sentence pitch: what your system does and for whom.
👋 New here? Read first: docs/student-guide.md · 📜 Brief & grading: INSTRUCTION.md · 🚀 Workflow: GETTING_STARTED.md
Template note: replace every (italic placeholder) below. Sections marked (mandatory) are required for grading.
| # | Full name | Student ID | Class | GitHub | Role |
|---|---|---|---|---|---|
| 1 | @ | ||||
| 2 | @ | ||||
| 3 | @ |
Topic: (e.g., Multi-room chat over TCP)
- Problem: (What problem are you solving? Who has it?)
- Our idea: (Your solution in 2–3 sentences.)
- What makes it non-trivial: (The networking challenge — e.g., resuming transfers, real-time sync, fairness under load.)
- Out of scope: (What you deliberately do not build.)
Full proposal: docs/proposal.md
- Connection & session: (handshake, authentication, ...)
- Custom protocol: specified in
docs/protocol-design.md - Concurrency: many clients at once, thread-safe shared state
- Robustness: abrupt disconnects, malformed messages, timeouts
- Feature A: (short description)
- Feature B: (short description)
graph LR
subgraph Clients
C1["Client 1 (CLI / GUI)"]
C2["Client 2 (CLI / GUI)"]
end
subgraph Server Container
S["Server (port 5000)<br/>concurrency model: ..."]
ST[("State / Storage")]
S --- ST
end
C1 -->|protocol messages over TCP/UDP| S
C2 -->|protocol messages over TCP/UDP| S
| Aspect | Our choice | Why (one line) |
|---|---|---|
| Language | ||
| Transport | TCP / UDP | |
| Framing | delimiter / length-prefix / fixed | |
| Concurrency model | thread-per-client / thread pool / event loop |
Details: docs/architecture.md
cp .env.example .env # or: make init
docker compose up --build -d # start the server (or: make up)
docker compose run --rm client # start a client — repeat in more terminals (or: make client)
docker compose down # stop everything (or: make down)(Add any extra steps, e.g., running a GUI client natively.)
(Screenshots or terminal output of: several clients at once, abrupt disconnect handling, your concurrency test.
Put images in docs/asset/. Full results: docs/testing-guide.md.)
| Document | Content |
|---|---|
docs/proposal.md |
M1 — problem, idea, scope, plan |
docs/protocol-design.md |
Application-layer protocol specification |
docs/architecture.md |
Concurrency model, components, connection lifecycle |
docs/testing-guide.md |
Test methods and our test evidence |
docs/ai-log.md |
AI usage log |
Policy:
INSTRUCTION.md§7. Disclosing AI use never lowers your score — hiding it does.
| Tool / model | Used by | Used for | Files / modules | Level |
|---|---|---|---|---|
| (e.g., Claude) | (member) | (e.g., draft protocol table, review locking code) | (paths) | (Assist / Co-write / Generated) |
Levels: Assist — explanations, suggestions, review; we wrote the code. Co-write — AI drafted parts, we substantially rewrote. Generated — AI wrote most of it; we reviewed, tested and can explain it.
(Key design decisions made by the team, possibly after comparing AI-suggested options. E.g., "Chose length-prefixed
framing over newline-delimited because file chunks may contain \n.")
(At least one concrete example: a bug, wrong assumption or bad design from AI, and how you detected and fixed it.)
See docs/ai-log.md.
| Member | Owns (modules / documents) | Key PRs / commits | AI-assisted parts | Contribution % |
|---|---|---|---|---|
(e.g., server/src/session/, docs/architecture.md) |
(e.g., #3, #7) | (e.g., protocol parser — Co-write) | ||
We confirm the table above is accurate and agreed by all members:
- Member 1
- Member 2
- Member 3
Based on the network-programming-starter template by Hung N. Dang.