leet2local 0.1.0

Solve LeetCode problems in your local IDE

Leet2Local

A CLI tool that brings LeetCode problems into your local IDE. Fetch problem descriptions, solve in your editor, run tests locally, and submit directly to LeetCode — all from the terminal.

lc fetch 1 --lang python
lc run 1
lc submit 1

---

Features

• Fetch any problem by number — creates a local directory with the description, code stub, and sample test cases
• Run locally — executes your solution against sample test cases without touching LeetCode servers
• Submit remotely — sends your code to LeetCode and polls for full results including hidden test cases, runtime, and memory percentile
• Multi-language — Python, JavaScript, C++, and Java
• VS Code integration — generates tasks.json for keyboard-shortcut access to all commands
• Secure auth — session cookies stored in OS native keychain (Windows Credential Manager / macOS Keychain)

---

Installation

Requirements: Python 3.11+

pip install leet2local

Or install from source:

git clone https://github.com/dhineshtheprogrammer/leet2local.git
cd Leet2Local
pip install -e .

For development (includes pytest and ruff):

pip install -e ".[dev]"

---

Quick Start

1. Log in

Get your session cookies from your browser after logging in to leetcode.com:

1. Open DevTools → Application → Cookies → leetcode.com
2. Copy LEETCODE_SESSION and csrftoken

lc login

2. Fetch a problem

lc fetch 1              # Two Sum, defaults to Python
lc fetch 42 --lang cpp  # Trapping Rain Water in C++

This creates:

./leetcode/1-two-sum/
├── question.md        # Problem description in Markdown
├── solution.py        # Code stub from LeetCode
└── test_cases.json    # Sample inputs and expected outputs

3. Write your solution

Open solution.py in your editor and implement the function.

4. Run locally

lc run 1

┌─ Test Results ──────────────────────────────────┐
│ #  │ Status │ Details                           │
│ 0  │  PASS  │                                   │
│ 1  │  PASS  │                                   │
│ 2  │  FAIL  │ got: [] expected: [0, 1]          │
└────────────────────────────────────────────────-┘
✗ 1/3 test(s) failed

5. Submit to LeetCode

lc submit 1

┌─ Accepted ──────────────────────────────────────┐
│ Runtime: 52 ms (beats 87.3%)                    │
│ Memory:  16.4 MB (beats 62.1%)                  │
│ Tests:   54/54                                  │
└─────────────────────────────────────────────────┘

---

Commands

Command	Description
lc login	Store LeetCode session cookies securely
lc fetch <number> [--lang]	Fetch problem and create local files
lc run <number>	Run solution against sample test cases
lc submit <number>	Submit to LeetCode and get full results
lc config	Show current configuration
lc config --set key=value	Update a config value
lc config --vscode-init	Generate .vscode/tasks.json

---

Language Support

Language	Flag	Runtime Required
Python 3	--lang python	python
JavaScript	--lang javascript	node
C++	--lang cpp	g++
Java	--lang java	javac, java

Set your default language so you don't need --lang on every fetch:

lc config --set settings.default_language=javascript

---

Configuration

On first run, create .leet2local.toml in your project root:

[settings]
default_language = "python"   # python | javascript | cpp | java
problems_dir = "./leetcode"   # where problem directories are created
run_mode = "local"            # local | remote

[api]
graphql_endpoint = "https://leetcode.com/graphql"
request_timeout = 30
max_retries = 5

[runtime]
python_cmd = "python"
node_cmd = "node"
java_cmd = "java"
javac_cmd = "javac"
cpp_compiler = "g++"
runner_timeout = 10           # seconds per test execution

[display]
show_difficulty = true
show_tags = false             # hide tags to avoid spoilers
syntax_highlight = true

Config is loaded from the nearest .leet2local.toml walking up from the current directory, falling back to ~/.config/leet2local/config.toml.

---

VS Code Integration

Generate keyboard-shortcut tasks:

lc config --vscode-init

This creates .vscode/tasks.json with three tasks. Add to your keybindings.json:

[
  { "key": "ctrl+shift+r", "command": "workbench.action.tasks.runTask", "args": "LC: Run Local" },
  { "key": "ctrl+shift+s", "command": "workbench.action.tasks.runTask", "args": "LC: Submit" }
]

---

How It Works

lc fetch 1
  └─ GraphQL: questionList (slug resolution)
  └─ GraphQL: questionData (description, stub, test cases)
  └─ HTML → Markdown (BeautifulSoup + markdownify)
  └─ Write question.md, solution.py, test_cases.json

lc run 1
  └─ Read test_cases.json
  └─ Render Jinja2 test harness template
  └─ subprocess → python harness.py
  └─ Parse LEET_PASS / LEET_FAIL / LEET_ERROR output lines

lc submit 1
  └─ GraphQL: submitCode mutation → submission_id
  └─ Poll: checkSubmission every 2s
  └─ Display Rich result table

---

Project Structure

Leet2Local/
├── leet2local/
│   ├── cli.py              # Typer CLI entry point
│   ├── auth.py             # Login, keyring credential storage
│   ├── fetcher.py          # GraphQL client, file creation
│   ├── runner.py           # Local runner dispatcher
│   ├── submitter.py        # Remote submission + polling
│   ├── html_parser.py      # HTML → Markdown conversion
│   ├── models.py           # Pydantic data models
│   ├── config.py           # TOML config management
│   ├── queries.py          # GraphQL query strings
│   ├── rate_limiter.py     # Tenacity retry decorator
│   └── runners/
│       ├── base.py         # Abstract Runner base class
│       ├── python_runner.py
│       ├── javascript_runner.py
│       ├── cpp_runner.py
│       └── java_runner.py
├── templates/              # Jinja2 test harness templates
├── scripts/
│   └── generate_slug_map.py  # Build slug fallback map
└── tests/

---

Limitations

• Sample test cases only for local runs — hidden test cases require lc submit
• LeetCode's API is unofficial and undocumented; field names may change
• C++ and Java require the respective compilers installed and on PATH
• lc submit requires a valid session cookie (obtained via lc login)

---

Development

# Run tests
pytest

# Lint
ruff check .

# Regenerate slug fallback map (run once, requires internet)
python scripts/generate_slug_map.py

---

License

MIT