Generate a C/C++ project using a modern development environment
- Run the jumpstart image once to pull the latest image and print the initial
help text:
docker run --rm davidzemon/jumpstart
- The following help text will be printed:
As the instructions explain, create a script on your
Welcome to jumpstart! The recommended way to execute jumpstart is by saving these lines to a script named 'jumpstart'. For detailed instructions, please visit https://github.com/DavidZemon/jumpstart #!/bin/bash set -e set -x docker pull davidzemon/jumpstart:latest docker run -it --rm \ -u "$(id -u):$(id -g)" \ -v "$(pwd):$(pwd)" \ -w "$(pwd)" \ davidzemon/jumpstart:latest "$@"
$PATH
with this content. Be sure to make the script executable withchmod +x ~/bin/jumpstart
, substituting~/bin/jumpstart
with whatever path you chose for the script. - Create an empty directory for your new project and run the newly created
jumpstart
scriptmkdir ~/sample cd ~/sample jumpstart
- Answer each interactive prompt to configure your new project. Each prompt has a default value in square brackets which will be used if no input is provided.
Jumpstart exposes each option as a command-line argument, allowing
non-interactive use of the generator. View the available arguments by invoking
the Docker wrapper script with the -h
/--help
argument:
jumpstart -h
To make non-interactive use even easier, a special --defaults
command-line
argument is exposed. Use this (probably with one or two other command-line
arguments) to quickly generate a project with only minor changes from the
default template. This can be useful for any number of reasons, including
comparing exactly what changes when a specific option is modified.
Though the working directory is expected to be empty, there is an exception
built into jumpstart which allows a file with license text to exist in the
working directory. Use this as an easy way to provide your own license text via
the --license
argument.