picker

type Model ¶

type Model struct {
	// contains filtered or unexported fields
}

Model is the bubbletea picker model.

func New(enabled map[string]string) Model

func (m Model) Chosen() *Option

func (m Model) Init() tea.Cmd

func (m Model) Language() string

func (m Model) Update(msg tea.Msg) (tea.Model, tea.Cmd)

func (m Model) View() string

type Option ¶

type Option struct {
	Label       string
	Build       string
	Run         string
	Description string
}

Option represents a predefined build+run pair for a language.

Why two fields instead of one shell string? Some testers (git, http-server) change CWD before invoking your_program.sh. The server runs Build inside the submission dir in a subshell, then exec's Run from the tester's CWD. Splitting them lets compiled languages still find their sources, and lets the run step stay writable to the tester's working directory (needed for things like `git init` writing .git there).

Conventions used in the presets below:

• Compiled outputs go to /tmp/app (absolute path, always writable in our Alpine runner, cleared between submissions by Docker --rm).
• Interpreted runs reference "$_dir/<file>" so they don't break when the tester chdirs. $_dir is defined in the your_program.sh header on the server.
• `make` presets assume the user's Makefile emits ./app at CWD — this is documented in the Description. If the user needs something else, they need a custom command (not yet supported).

func (o Option) IsMakePreset() bool