bh-pixelfuse 0.0.5

Embed images into portable text files and extract them back

PixelFuse Backend

FastAPI service that embeds images into portable text files and extracts them back.

Docs

#	File	Description
001	Overview	What PixelFuse does and the text file format
002	API Reference	Endpoint contracts, fields, and error codes
003	Configuration	Env vars and CLI flags
004	Development	Local setup, tests, lint, Docker
005	Deployment	Render, CI, self-hosted Docker

Endpoints

Method	Path	Description
POST	/convert-embed/	Upload images → download base64 text file
POST	/extract-images/	Upload text file → download ZIP of images

Supports JPEG, PNG, and HEIC formats. Max 10 files per request (configurable).

No distortion. Embed → extract is lossless. Raw image bytes are base64-encoded and decoded back exactly — no re-encoding at any stage.

Quickstart

pip install -e ".[dev]"
pixelfuse serve --reload

Configuration

All settings are overridable via environment variables prefixed PIXELFUSE_:

Variable	Default	Description
PIXELFUSE_HOST	0.0.0.0	Bind host
PIXELFUSE_PORT	8000	Bind port
PIXELFUSE_LOG_LEVEL	info	Uvicorn log level
PIXELFUSE_MAX_UPLOAD_FILES	10	Max files per request
PIXELFUSE_RATE_LIMIT_PER_MINUTE	60	Max requests per minute per IP (0 = disabled)
PIXELFUSE_ALLOWED_ORIGINS	(required)	CORS origins (JSON array)

Copy .env.example to .env for local overrides.

Development

# Lint
flake8 src tests

# Type check
mypy src

# All environments
tox

Deployment

Deployed on Render via render.yaml. CI runs on every push via GitHub Actions.