pipeline

Documentation

Index

• func CheckNames(p Pipeline) error
• type Edge
  • func (e Edge) From() graph.Node
  • func (e Edge) ID() int64
  • func (e Edge) To() graph.Node
• type NameError
  • func (ne *NameError) Error() string
• type Node
  • func (n Node) DOTID() string
  • func (n Node) ID() int64
• type Parallelism
• type Pipeline
  • func FindAndReplaceVariables(p Pipeline, file []byte) (Pipeline, error)
  • func ParseConfig(filename string) (*Pipeline, error)
  • func ReadPipelineDescription(file []byte, filename string) (Pipeline, error)
  • func (p Pipeline) FixDependencies()
  • func (p Pipeline) String() string
  • func (p *Pipeline) WriteDOT(filename string) error
  • func (p Pipeline) WritePipelineDescription(filename string) error
• type Stage
  • func (stage Stage) String() string
• type Variable
  • func (v Variable) String() string

Functions

func CheckNames

func CheckNames(p Pipeline) error

Verify that the pipeline name and pipeline stage names are valid.

Types

type Edge

type Edge struct {
	F Node
	T Node
	// contains filtered or unexported fields
}

func (Edge) From

func (e Edge) From() graph.Node

func (Edge) ID

func (e Edge) ID() int64

func (Edge) To

func (e Edge) To() graph.Node

type NameError

type NameError struct {
	OffendingName string
	Explanation   string
}

func (*NameError) Error

func (ne *NameError) Error() string

type Node

type Node struct {
	Name string
}

func (Node) DOTID

func (n Node) DOTID() string

func (Node) ID

func (n Node) ID() int64

type Parallelism

type Parallelism struct {
	Strategy string
	Constant int
}

type Pipeline

type Pipeline struct {
	Name      string
	Stages    []*Stage
	Comment   string
	Variables []Variable
	Commit    bool
	Runtime   time.Duration
	Version   string
}

func FindAndReplaceVariables

func FindAndReplaceVariables(p Pipeline, file []byte) (Pipeline, error)

Finds and replaces all variable names with their respective single values. On success it returns the file contents of the pipeline description file. For multi-value variables it will create one stage per variable value. We assume that these can run concurrently.

func ParseConfig

func ParseConfig(filename string) (*Pipeline, error)

Parses the pipeline configuration and returns the pipeline. It will verify that names are valid, find and replace variable names and create parallel pipeline stages.

func ReadPipelineDescription

func ReadPipelineDescription(file []byte, filename string) (Pipeline, error)

Read a pipeline description.

func (Pipeline) FixDependencies

func (p Pipeline) FixDependencies()

Since we support parallelizing stages, a stage has to wait for automatically generated stages. E.g. if we have parallelized stage input_A and input_B, both with following stages process_A and process_B. process_A will have to wait for input_A, but the pipeline definition will only list `input` as a dependency. This function fixes and cleans up the "dependency graph". NOTE TO SELF: refactor this one jesus christ it's bad.

func (Pipeline) String

func (p Pipeline) String() string

Stringify a pipeline description.

func (*Pipeline) WriteDOT

func (p *Pipeline) WriteDOT(filename string) error

func (Pipeline) WritePipelineDescription

func (p Pipeline) WritePipelineDescription(filename string) error

Write a pipeline description to a file.

type Stage

type Stage struct {
	Name             string
	Image            string
	Entrypoint       []string
	Cmd              []string
	Env              []string
	Inputs           []string
	Volumes          []string
	Parallelism      Parallelism
	Cache            bool
	Comment          string
	MountPropagation string
	Version          string

	Runtime time.Duration
	// contains filtered or unexported fields
}

func (Stage) String

func (stage Stage) String() string

Stringify a pipeline stage.

type Variable

type Variable struct {
	Name   string
	Values []string
}

func (Variable) String

func (v Variable) String() string