🐸 rainfrog

a database tool for the terminal

[!WARNING] rainfrog is currently in beta

the goal for rainfrog is to provide a lightweight, terminal-based tool for interacting with databases.

features

• efficient navigation via vim-like keybindings and mouse controls
• query editor with keyword highlighting, session history, and favorites
• quickly copy data, filter tables, and switch between schemas
• shortcuts to view table metadata and properties
• cross-platform (macOS, linux, windows, android via termux)

why "rainfrog"?

frogs find refuge in elephant tracks

supported databases

the levels of support for different databases can be split into 4 tiers:

• tier 1: most frequently tested, bugs will be addressed as soon as possible.
• tier 2: often tested, fixed when necessary.
• tier 3: community supported; contributions are welcome, but otherwise development is minimal.
• tier 4: new or experimental; should be treated as unstable, and behavior is subject to change.

*the postgres driver can also be used to connect to other databases that support the postgres wire protocol, such as AWS Redshift. however, this functionality is not well tested. in theory, the mysql driver should be able to do the same for databases that support the mysql protocol. check each database's documentation for compatibility.

disclaimer

this software is currently under active development; expect breaking changes, and use at your own risk. it is not recommended to use this tool with write access on a production database.

table of contents

• installation
  • cargo
  • homebrew
  • arch linux
  • termux
  • nix
  • conda
  • pixi
  • install script
  • release page binaries
  • icons
• dependencies
  • oracle
• usage
  • with connection options
  • with connection url
  • with environment variables
  • docker run
• customization
  • settings
  • database connections
  • keybindings
    • n.b. for mac users
    • general
    • menu (list of schemas and tables)
    • query editor
    • query history
    • query favorites
    • results
• exports
• favorites
• roadmap
• known issues and limitations
• Contributing
• acknowledgements

installation

cargo

after installing rust (recommended to do so via rustup):

cargo install rainfrog --locked

by default, all features are included. some features may not be compatible with your platform; check Cargo.toml to see what feature combinations are available.

if you have cargo-binstall, you can run:

cargo binstall rainfrog

homebrew

brew install rainfrog

arch linux

arch linux users can install from the official repositories using pacman:

pacman -S rainfrog

termux

if you are using termux, you'll need to install rust via their package manager:

pkg install rust

and then make sure to install with termux features (and disable default features):

cargo install rainfrog --no-default-features

nix

nix-env -iA nixos.rainfrog

conda

conda install -c conda-forge rainfrog

pixi

The command below installs rainfrog from conda-forge, the same repository as conda, but unlike conda it can be installed user-globally using pixi.

pixi global install rainfrog

install script

there is a simple install script that assists in downloading and unpacking a binary from the release page to ~/.local/bin/, which you might want to add to your PATH variable if it isn't already there. you'll need to select which binary is appropriate for your system (if you're not sure, you can find out by installing rust and running rustc -vV to see the "host" target), and the script also needs jq and fzf installed to run.

curl -LSsf https://raw.githubusercontent.com/achristmascarl/rainfrog/main/install.sh | bash

release page binaries

1. manually download and unpack the appropriate binary for your os from the latest release (if you're not sure which binary to pick, you can find out by installing rust and running rustc -vV to see the "host" target)
2. move the binary to a folder in your PATH environment variable

icons

you will need to have a nerd font installed and used by your terminal for the icons to show up.

dependencies

for some of the databases supported by rainfrog, the driver has additional runtime dependencies. you can find instructions for installing those dependencies below.

oracle

the oracle driver uses ODPI-C, which requires Oracle Client libraries to be installed. instructions for installing the Oracle Instant Client can be found here: ODPI-C Installation.

after installation, you may need to manually copy the client files to a location where ODPI-C can find them, which is either in the same folder as the rainfrog binary, or in the library path of your os. more details can be found in the ODPI-C installation instructions: ODPI-C Client Library Loading.

usage

Usage: rainfrog [OPTIONS] [COMMAND]

Commands:
  edit  Edit the config file (create it first if missing)
  help  Print this message or the help of the given subcommand(s)

Options:
  -M, --mouse <MOUSE_MODE>   Whether to enable mouse event support. If enabled, the default mouse event handling for your terminal
                             will not work. [possible values: true, false]
  -u, --url <URL>            Full connection URL for the database, e.g. postgres://username:password@localhost:5432/dbname
      --username <USERNAME>  Username for database connection
      --password <PASSWORD>  Password for database connection
      --host <HOST>          Host for database connection (ex. localhost)
      --port <PORT>          Port for database connection (ex. 5432)
      --database <DATABASE>  Name of database for connection (ex. postgres)
      --driver <DRIVER>      Driver for database connection (ex. postgres)
  -h, --help                 Print help
  -V, --version              Print version

with connection options

if any options are not provided, you will be prompted to input them. if you do not provide an input, that option will default to what is in your environment variables.

rainfrog \
  --driver <db_driver> \
  --username <username> \
  --host <hostname> \
  --port <db_port> \
  --database <db_name>

with connection url

the connection_url must include all the necessary options for connecting to the database (ex. postgres://username:password@localhost:5432/postgres). it will take precedence over all connection options.

rainfrog --url $(connection_url)

with environment variables

if connection_url is not present, rainfrog will first check your DATABASE_URL environment variable, then it will check for DATABASE_URL in a .env file in the current directory or a parent directory.

docker run

for postgres and mysql, you can run it by specifying all of the options as environment variables:

docker run --platform linux/amd64 -it --rm --name rainfrog \
  --add-host host.docker.internal:host-gateway \
  -e db_driver="db_driver" \
  -e username="<username>" \
  -e password="<password>" \
  -e hostname="host.docker.internal" \
  -e db_port="<db_port>" \
  -e db_name="<db_name>" achristmascarl/rainfrog:latest

if you want to provide a custom combination of options and omit others, you can override the Dockerfile's CMD like so:

docker run --platform linux/amd64 -it --rm --name rainfrog \
  achristmascarl/rainfrog:latest \
  rainfrog # overrides CMD, additional options would come after

since sqlite is file-based, you may need to mount a path to the sqlite db as a volume in order to access it:

docker run --platform linux/amd64 -it --rm --name rainfrog \
  -v ~/code/rainfrog/dev/rainfrog.sqlite3:/rainfrog.sqlite3 \
  achristmascarl/rainfrog:latest \
  rainfrog --url sqlite:///rainfrog.sqlite3

running drivers which have runtime dependencies (like oracle) via docker is not currently supported.

customization

rainfrog can be customized by placing a rainfrog_config.toml file in one of the following locations depending on your os, as determined by the directories crate:

you can change the default config directory by exporting an environment variable. to make the change permanent, add it to your .zshrc/.bashrc/.*rc file:

export RAINFROG_CONFIG=~/.config/rainfrog

[!WARNING] this should be the path to the directory that the config file is in, NOT the path to the actual file

settings

database connections

database connections can be configured in the configuration file as shown below:

[db]
postgres-local = { host = "localhost", driver = "postgres", port = 5432, database = "postgres", username = "postgres",  default = true }
postgres-dev = { connection_string = "postgresql://postgres:test123@127.0.0.1:5432", driver = "postgres" }
mysql-local = { host = "localhost", driver = "mysql", port = 32768, database = "rainfrog", username = "root" }
sqlite-memory = { connection_string = "sqlite://:memory:", driver = "sqlite"}
sqlite-disk = { connection_string = "sqlite://./my_database.db", driver = "sqlite"}
oracle-local = { host = "localhost", driver = "oracle", port = 1521, database = "rainfrog", username = "rainfrog" }

the connection details can be provided in two formats: a raw connection string or specifying individual fields. connections input are prioritized in the following order:

• cli input
• DATABASE_URL env variable
• config file

if no database connection in the config is set as the default connection, a prompt will appear to select the desired database. The user will also be prompted for the password for the selected database and will have the option to store it in a platform specific keychain for future reuse. future plans for database connections include switching database without having to restart rainfrog.

keybindings

you can customize some of the default keybindings, but not all of them. to see a list of the ones you can customize, see the default config file at .config/rainfrog_config.toml. below are the default keybindings.

n.b. for mac users

on mac, for VS Code, terminal, kitty, and others, a setting for "use option as meta key" needs to be turned on for Alt/Opt keybindings to work. (In VS Code, it's "terminal.integrated.macOptionIsMeta": true in the settings; in kitty, it's macos_option_as_alt yes in the config.)

general