PyComp — Python Comprehension Converter

Automatically converts traditional Python for loops into list, set, and dictionary comprehensions — with a preview diff before applying any change.

Powered by a local FastAPI + PostgreSQL backend that acts as an intelligent cache: identical code is transformed once and served instantly from cache on subsequent requests.

Features

• Convert loops to comprehensions — detects convertible patterns and rewrites them in place
• Preview before applying — shows a native VS Code diff view so you can review the change before committing
• List, set, and dict comprehensions — handles all three types, with and without if conditions
• Smart cache — tracks every transformation in PostgreSQL; repeated code is served from cache instantly
• Transformation history — browse, insert, and clear past transformations from the Command Palette
• Context menu integration — right-click any selection in a Python file to convert it
• Keyboard shortcut — Ctrl+Shift+L / Cmd+Shift+L

Requirements

PyComp requires a local backend to be running before use.

1. Install Python dependencies

uv sync

2. Configure your .env

DATABASE_URL=postgresql://user:password@localhost:5432/pylens

3. Run database migrations

uv run alembic upgrade head

4. Start the backend

uv run uvicorn pylens.app.main:app --port 8000

The extension connects to http://localhost:8000 by default.

Usage

Convert a loop

1. Open a .py file
2. Select the code you want to convert (include the name = [] initialization line)
3. Press Ctrl+Shift+L (Cmd+Shift+L on Mac) — or right-click → PyComp: Convert to List Comprehension
4. Review the diff that appears on the right
5. Click Apply to replace the code, or Cancel to discard

Patterns supported

List comprehension

# Before
result = []
for x in items:
    result.append(x * 2)

# After
result = [x * 2 for x in items]

List comprehension with condition

# Before
evens = []
for n in range(20):
    if n % 2 == 0:
        evens.append(n)

# After
evens = [n for n in range(20) if n % 2 == 0]

Set comprehension

# Before
unique = set()
for word in words:
    unique.add(word.lower())

# After
unique = {word.lower() for word in words}

Dictionary comprehension

# Before
squared = {}
for n in numbers:
    squared[n] = n ** 2

# After
squared = {n: n ** 2 for n in numbers}

Patterns NOT converted (intentionally)

PyComp is conservative. It will not convert:

• for/else loops — else clauses have no equivalent in comprehensions
• Loops with break, continue, or return — semantics would change
• Loops with multiple statements in the body
• Loops where the accumulator variable is used inside the expression
• Pre-filled collections (result = [1, 2], d = {"a": 1})

Commands

Extension Settings

How the cache works

Every transformation is stored in PostgreSQL with a SHA-256 hash of the original code. When you submit the same code again, PyComp returns the cached result instantly and increments a hits counter — no AST parsing needed. You can inspect all cached transformations via PyComp: Show Transformation History.

License

MIT