depush 0.1.0

Deploy a versioned codebase to S3/MinIO, SSH, or a local directory

depush

Deploy a versioned codebase directory to S3/MinIO, SSH, or a local directory.

depush reads a version file from your codebase directory and deploys all files to {prefix}/{version}/ at the destination. Stale files at the destination that no longer exist in the source are automatically deleted, keeping deployments in sync.

Configuration is flexible: use CLI flags, environment variables, or a YAML config file. Precedence (highest to lowest): CLI flag > environment variable > YAML config > default.

---

Installation

pip install depush

Or with uv:

uv add depush

---

Quick start

Your codebase directory must contain a version file:

my-project/
├── version        # contains e.g. "1.2.3"
├── main.py
└── ...

Deploy to S3 / MinIO

depush --target s3 --prefix mylib \
  --s3-bucket deployments \
  --s3-endpoint http://localhost:9000 \
  --s3-access-key admin \
  --s3-secret-key secret

Files are uploaded to s3://deployments/mylib/1.2.3/.

Deploy over SSH

depush --target ssh --prefix mylib \
  --ssh-host myserver.example.com \
  --ssh-user deploy \
  --ssh-key-file ~/.ssh/id_rsa

Files are copied to /deployments/mylib/1.2.3/ on the remote host.

Deploy to a local directory

depush --target local --prefix mylib --local-dest /srv/releases

Files are copied to /srv/releases/mylib/1.2.3/.

Dry run

Add --dry-run to any command to preview what would be deployed without making any changes:

depush --target s3 --prefix mylib --s3-bucket deployments --dry-run

---

Configuration file

Place a depush.yaml in the current directory (or pass --config path/to/file.yaml) to avoid repeating flags:

target: s3
prefix: mylib
codebase_dir: ./codebase
dry_run: false

s3:
  bucket: deployments
  endpoint: http://localhost:9000   # omit for AWS
  region: us-east-1
  access_key: admin
  secret_key: secret

# local:
#   dest: ./dist

# ssh:
#   host: myserver.example.com
#   port: 22
#   user: deploy
#   key_file: ~/.ssh/id_rsa
#   deploy_root: /deployments

---

Environment variables

All options can also be set via environment variables:

Variable	Description	Default
DEPLOY_TARGET	Deployment target: s3, ssh, or local	(required)
DEPLOY_PREFIX	Path prefix, e.g. mylib → mylib/{version}/	(required)
DEPLOY_CODEBASE_DIR	Path to codebase directory	.
DEPLOY_DRY_RUN	Set to 1, true, or yes to preview	false
DEPLOY_CONFIG	Path to a YAML config file	depush.yaml if present
DEPLOY_LOCAL_DEST	Root destination for local deployments	./dist
S3_BUCKET	S3/MinIO bucket name	(required for s3)
S3_ENDPOINT	Custom endpoint URL for MinIO	(AWS default)
S3_REGION	AWS region	us-east-1
S3_PROFILE	AWS credentials profile name (from ~/.aws/credentials); falls back to AWS_PROFILE
S3_ACCESS_KEY	Access key / MinIO username
S3_SECRET_KEY	Secret key / MinIO password
SSH_HOST	SSH server hostname or IP	(required for ssh)
SSH_PORT	SSH port	22
SSH_USER	SSH username	admin
SSH_PASSWORD	SSH password (omit to use key auth)
SSH_KEY_FILE	Path to private key file
SSH_DEPLOY_ROOT	Absolute path on the remote server	/deployments

---

Ignoring files

Create a .depushignore file in your codebase directory using the same syntax as .gitignore to exclude files from deployment:

*.log
__pycache__/
.env

---

Programmatic use

from depush import depush

# Parse args and deploy
depush.main()

# Or use individual functions
files = depush.collect_files(Path("./my-codebase"))