docker

package
v1.15.1-beta.5 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Aug 17, 2026 License: MPL-2.0 Imports: 26 Imported by: 0

Documentation

Index

Constants

This section is empty.

Variables

View Source 
var (
	ErrContainerTerminal = errors.New("container in terminal state")
	ErrContainerExited   = errors.New("container exited after start")
	ErrImageNotFound     = errors.New("image not found")
)

Functions

func LoadDockerConfig 

func LoadDockerConfig() (*configfile.ConfigFile, error)

func PrepareForGroupCancellation 

func PrepareForGroupCancellation(cmd *exec.Cmd)

PrepareForGroupCancellation sets the Cancel function of the given exec.Cmd to kill the entire process group, allowing for cleanup of child processes. This is necessary because exec.Cmd does not automatically kill child processes when the context is canceled.

Types

type Client 

type Client struct {
	dockerclient.CommonAPIClient
}

Client is a client for docker.

func NewClient 

func NewClient(ctx context.Context) (*Client, error)

NewClient creates a new docker client.

type ContainerRuntime added in v1.1.0 

type ContainerRuntime interface {
	Name() RuntimeName
	SupportsInternalBuildKit() bool
	SupportsSignalProxy() bool
	SupportsMountConsistency() bool
	NeedsUserNamespaceArgs() bool
	GPUAvailable(ctx context.Context, helper *DockerHelper) (bool, error)
}

ContainerRuntime describes the capabilities of a container runtime (Docker, Podman, nerdctl). Rather than checking IsPodman()/IsNerdctl() at every call site, callers query capability methods that express *why* the code branches.

func DetectRuntime added in v1.1.0 

func DetectRuntime(dockerCommand string) ContainerRuntime

DetectRuntime probes the binary at dockerCommand to determine which runtime it is. The result is cached per binary path.

func RuntimeFromName added in v1.1.0 

func RuntimeFromName(name string) (ContainerRuntime, error)

RuntimeFromName returns a ContainerRuntime for the given explicit name. Empty string defaults to Docker. Unknown non-empty names return an error to catch typos in provider configuration.

type DockerBuilder 

type DockerBuilder int

DockerBuilder represents the Docker builder types. 

const (
	DockerBuilderDefault DockerBuilder = iota
	DockerBuilderBuildX
	DockerBuilderBuildKit
)

Enum values for DockerBuilder.

func DockerBuilderFromString 

func DockerBuilderFromString(s string) (DockerBuilder, error)

func (DockerBuilder) String 

func (db DockerBuilder) String() string

type DockerHelper 

type DockerHelper struct {
	DockerCommand string
	// for a running container, we cannot pass down the container ID to the driver without introducing
	// changes in the driver interface (which we do not want to do). So, to get around this, we pass
	// it down to the driver during docker helper initialization.
	ContainerID string
	// allow command to have a custom environment
	Environment []string
	Builder     DockerBuilder
	Runtime     ContainerRuntime
	// Elevator, when set, runs docker commands through a privilege-elevation
	// helper (pkexec/sudo/doas). Nil disables elevation.
	Elevator *Elevator
}

func (*DockerHelper) ClientVersion added in v1.3.1 

func (r *DockerHelper) ClientVersion(ctx context.Context) string

ClientVersion returns the docker CLI version (e.g. "29.5.3"), or "".

func (*DockerHelper) DeleteVolume 

func (r *DockerHelper) DeleteVolume(ctx context.Context, volume string) error

func (*DockerHelper) EnsureElevated added in v1.10.0 

func (r *DockerHelper) EnsureElevated() error

EnsureElevated authenticates the configured elevator once; concurrent callers block on the single prompt. No-op and safe when no elevator is set.

func (*DockerHelper) FindContainer 

func (r *DockerHelper) FindContainer(ctx context.Context, labels []string) ([]string, error)

FindContainer will try to find a container based on the input labels. If no container is found, it will search for the labels manually inspecting containers.

func (*DockerHelper) FindContainerByID 

func (r *DockerHelper) FindContainerByID(
	ctx context.Context,
	containerIds []string,
) (*config.ContainerDetails, error)

func (*DockerHelper) FindContainerJSON 

func (r *DockerHelper) FindContainerJSON(ctx context.Context, labels []string) ([]string, error)

FindContainerJSON will manually search for containers with matching labels. This is useful in case the `--filter` doesn't work.

func (*DockerHelper) FindDevContainer 

func (r *DockerHelper) FindDevContainer(
	ctx context.Context,
	labels []string,
) (*config.ContainerDetails, error)

func (*DockerHelper) GPUSupportEnabled 

func (r *DockerHelper) GPUSupportEnabled() (bool, error)

func (*DockerHelper) GetContainerLogs 

func (r *DockerHelper) GetContainerLogs(
	ctx context.Context,
	id string,
	stdout io.Writer,
	stderr io.Writer,
) error

func (*DockerHelper) GetImageTag 

func (r *DockerHelper) GetImageTag(ctx context.Context, imageID string) (string, error)

func (*DockerHelper) GetRuntime added in v1.1.0 

func (r *DockerHelper) GetRuntime() ContainerRuntime

GetRuntime returns the container runtime for this helper. If no runtime was explicitly set, it auto-detects from the docker command.

func (*DockerHelper) Inspect 

func (r *DockerHelper) Inspect(
	ctx context.Context,
	ids []string,
	inspectType string,
	obj any,
) error

func (*DockerHelper) InspectContainers 

func (r *DockerHelper) InspectContainers(
	ctx context.Context,
	ids []string,
) ([]config.ContainerDetails, error)

func (*DockerHelper) InspectImage 

func (r *DockerHelper) InspectImage(
	ctx context.Context,
	imageName string,
	tryRemote bool,
) (*config.ImageDetails, error)

func (*DockerHelper) IsNerdctl 

func (r *DockerHelper) IsNerdctl() bool

func (*DockerHelper) IsPodman 

func (r *DockerHelper) IsPodman() bool

func (*DockerHelper) LingerWarning added in v1.5.1 

func (r *DockerHelper) LingerWarning(ctx context.Context) string

func (*DockerHelper) Ping added in v1.11.0 

func (r *DockerHelper) Ping(ctx context.Context) error

Ping reports whether the runtime daemon is reachable, returning its own message (e.g. "Cannot connect to Podman") on failure. It runs a bare `info` and judges reachability by exit status: `--format` field names differ between docker (.ServerVersion) and podman/nerdctl, so a shared template would falsely fail non-docker runtimes.

func (*DockerHelper) PodmanMachineExists 

func (r *DockerHelper) PodmanMachineExists(ctx context.Context) (bool, error)

PodmanMachineExists reports whether a Podman machine exists.

func (*DockerHelper) Pull 

func (r *DockerHelper) Pull(ctx context.Context, opts PullOptions) error

func (*DockerHelper) Remove 

func (r *DockerHelper) Remove(ctx context.Context, id string) error

func (*DockerHelper) Run 

func (r *DockerHelper) Run(ctx context.Context, args []string, streams Streams) error

func (*DockerHelper) RunWithDir 

func (r *DockerHelper) RunWithDir(
	ctx context.Context,
	dir string,
	args []string,
	streams Streams,
) error

func (*DockerHelper) RunWithEnv added in v1.10.0 

func (r *DockerHelper) RunWithEnv(
	ctx context.Context,
	extraEnv []string,
	args []string,
	streams Streams,
) error

RunWithEnv runs a command with extra environment variables for this invocation only (used for env-backed buildx secrets).

func (*DockerHelper) StartContainer 

func (r *DockerHelper) StartContainer(ctx context.Context, containerId string) error

func (*DockerHelper) StartPodmanMachine added in v1.11.0 

func (r *DockerHelper) StartPodmanMachine(ctx context.Context) error

StartPodmanMachine starts the default Podman machine, which must already exist.

func (*DockerHelper) StartRootlessPodmanSocket 

func (r *DockerHelper) StartRootlessPodmanSocket(ctx context.Context) error

StartRootlessPodmanSocket starts the podman.socket systemd unit for rootless Podman.

func (*DockerHelper) Stop 

func (r *DockerHelper) Stop(ctx context.Context, id string) error

func (*DockerHelper) WaitContainerRunning 

func (r *DockerHelper) WaitContainerRunning(ctx context.Context, containerID string) error

WaitContainerRunning waits for the given container to be running, returning an error if it is in a terminal state or does not become running within a timeout.

type Elevator added in v1.10.0 

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

Elevator runs docker commands through a privilege-elevation helper (pkexec, sudo, doas). It authenticates once, up front, so an operation's many commands share a single prompt via the warmed OS credential cache.

func ElevatorFromName added in v1.10.0 

func ElevatorFromName(name string) (*Elevator, error)

ElevatorFromName maps a helper name to an Elevator; "" and "none" return (nil, nil), unknown names error.

type PullOptions added in v1.3.1 

type PullOptions struct {
	Image    string
	Platform string
	Stdin    io.Reader
	Stdout   io.Writer
	Stderr   io.Writer
}

PullOptions configures an image pull.

type RuntimeName added in v1.1.0 

type RuntimeName string

RuntimeName identifies a container runtime. 

const (
	RuntimeDocker  RuntimeName = "docker"
	RuntimePodman  RuntimeName = "podman"
	RuntimeNerdctl RuntimeName = "nerdctl"
)

type Streams added in v1.10.0 

type Streams struct {
	Stdin  io.Reader
	Stdout io.Writer
	Stderr io.Writer
}

Source Files

View all Source files

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.