act/pkg/runner/action.go

541 lines
16 KiB
Go
Raw Normal View History

package runner
import (
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
"context"
"embed"
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
"fmt"
"io"
"io/fs"
"os"
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
"path"
"path/filepath"
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
"regexp"
"runtime"
"strings"
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
"github.com/kballard/go-shellquote"
"github.com/nektos/act/pkg/common"
"github.com/nektos/act/pkg/container"
"github.com/nektos/act/pkg/model"
log "github.com/sirupsen/logrus"
)
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
type actionStep interface {
step
getActionModel() *model.Action
}
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
type readAction func(step *model.Step, actionDir string, actionPath string, readFile actionYamlReader, writeFile fileWriter) (*model.Action, error)
type actionYamlReader func(filename string) (io.Reader, io.Closer, error)
type fileWriter func(filename string, data []byte, perm fs.FileMode) error
type runAction func(step actionStep, actionDir string, remoteAction *remoteAction) common.Executor
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
//go:embed res/trampoline.js
var trampoline embed.FS
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
func readActionImpl(step *model.Step, actionDir string, actionPath string, readFile actionYamlReader, writeFile fileWriter) (*model.Action, error) {
reader, closer, err := readFile("action.yml")
if os.IsNotExist(err) {
reader, closer, err = readFile("action.yaml")
if err != nil {
if _, closer, err2 := readFile("Dockerfile"); err2 == nil {
closer.Close()
action := &model.Action{
Name: "(Synthetic)",
Runs: model.ActionRuns{
Using: "docker",
Image: "Dockerfile",
},
}
log.Debugf("Using synthetic action %v for Dockerfile", action)
return action, nil
}
if step.With != nil {
if val, ok := step.With["args"]; ok {
var b []byte
if b, err = trampoline.ReadFile("res/trampoline.js"); err != nil {
return nil, err
}
err2 := writeFile(filepath.Join(actionDir, actionPath, "trampoline.js"), b, 0400)
if err2 != nil {
return nil, err2
}
action := &model.Action{
Name: "(Synthetic)",
Inputs: map[string]model.Input{
"cwd": {
Description: "(Actual working directory)",
Required: false,
Default: filepath.Join(actionDir, actionPath),
},
"command": {
Description: "(Actual program)",
Required: false,
Default: val,
},
},
Runs: model.ActionRuns{
Using: "node12",
Main: "trampoline.js",
},
}
log.Debugf("Using synthetic action %v", action)
return action, nil
}
}
return nil, err
}
} else if err != nil {
return nil, err
}
defer closer.Close()
action, err := model.ReadAction(reader)
log.Debugf("Read action %v from '%s'", action, "Unknown")
return action, err
}
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
func runActionImpl(step actionStep, actionDir string, remoteAction *remoteAction) common.Executor {
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
rc := step.getRunContext()
stepModel := step.getStepModel()
return func(ctx context.Context) error {
actionPath := ""
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
if remoteAction != nil && remoteAction.Path != "" {
actionPath = remoteAction.Path
}
action := step.getActionModel()
log.Debugf("About to run action %v", action)
if remoteAction != nil {
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
rc.ActionRepository = fmt.Sprintf("%s/%s", remoteAction.Org, remoteAction.Repo)
rc.ActionRef = remoteAction.Ref
} else {
rc.ActionRepository = ""
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
rc.ActionRef = ""
}
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
defer (func() {
// cleanup after the action is done, to avoid side-effects in
// the next step/action
rc.ActionRepository = ""
rc.ActionRef = ""
})()
// we need to merge with github-env again, since at the step setup
// time, we don't have all environment prepared
mergeIntoMap(step.getEnv(), rc.withGithubEnv(map[string]string{}))
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
populateEnvsFromInput(step.getEnv(), action, rc)
actionLocation := path.Join(actionDir, actionPath)
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
actionName, containerActionDir := getContainerActionPaths(stepModel, actionLocation, rc)
log.Debugf("type=%v actionDir=%s actionPath=%s workdir=%s actionCacheDir=%s actionName=%s containerActionDir=%s", stepModel.Type(), actionDir, actionPath, rc.Config.Workdir, rc.ActionCacheDir(), actionName, containerActionDir)
maybeCopyToActionDir := func() error {
if stepModel.Type() != model.StepTypeUsesActionRemote {
return nil
}
if err := removeGitIgnore(actionDir); err != nil {
return err
}
var containerActionDirCopy string
containerActionDirCopy = strings.TrimSuffix(containerActionDir, actionPath)
log.Debug(containerActionDirCopy)
if !strings.HasSuffix(containerActionDirCopy, `/`) {
containerActionDirCopy += `/`
}
return rc.JobContainer.CopyDir(containerActionDirCopy, actionDir+"/", rc.Config.UseGitIgnore)(ctx)
}
switch action.Runs.Using {
case model.ActionRunsUsingNode12, model.ActionRunsUsingNode16:
if err := maybeCopyToActionDir(); err != nil {
return err
}
containerArgs := []string{"node", path.Join(containerActionDir, action.Runs.Main)}
log.Debugf("executing remote job container: %s", containerArgs)
return rc.execJobContainer(containerArgs, *step.getEnv(), "", "")(ctx)
case model.ActionRunsUsingDocker:
location := actionLocation
if remoteAction == nil {
location = containerActionDir
}
return execAsDocker(ctx, step, actionName, location, remoteAction == nil)
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
case model.ActionRunsUsingComposite:
if err := maybeCopyToActionDir(); err != nil {
return err
}
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
return execAsComposite(step, containerActionDir)(ctx)
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
default:
return fmt.Errorf(fmt.Sprintf("The runs.using key must be one of: %v, got %s", []string{
model.ActionRunsUsingDocker,
model.ActionRunsUsingNode12,
model.ActionRunsUsingNode16,
model.ActionRunsUsingComposite,
}, action.Runs.Using))
}
}
}
// https://github.com/nektos/act/issues/228#issuecomment-629709055
// files in .gitignore are not copied in a Docker container
// this causes issues with actions that ignore other important resources
// such as `node_modules` for example
func removeGitIgnore(directory string) error {
gitIgnorePath := path.Join(directory, ".gitignore")
if _, err := os.Stat(gitIgnorePath); err == nil {
// .gitignore exists
log.Debugf("Removing %s before docker cp", gitIgnorePath)
err := os.Remove(gitIgnorePath)
if err != nil {
return err
}
}
return nil
}
// TODO: break out parts of function to reduce complexicity
// nolint:gocyclo
func execAsDocker(ctx context.Context, step actionStep, actionName string, basedir string, localAction bool) error {
rc := step.getRunContext()
action := step.getActionModel()
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
var prepImage common.Executor
var image string
if strings.HasPrefix(action.Runs.Image, "docker://") {
image = strings.TrimPrefix(action.Runs.Image, "docker://")
} else {
// "-dockeraction" enshures that "./", "./test " won't get converted to "act-:latest", "act-test-:latest" which are invalid docker image names
image = fmt.Sprintf("%s-dockeraction:%s", regexp.MustCompile("[^a-zA-Z0-9]").ReplaceAllString(actionName, "-"), "latest")
image = fmt.Sprintf("act-%s", strings.TrimLeft(image, "-"))
image = strings.ToLower(image)
contextDir := filepath.Join(basedir, action.Runs.Main)
anyArchExists, err := container.ImageExistsLocally(ctx, image, "any")
if err != nil {
return err
}
correctArchExists, err := container.ImageExistsLocally(ctx, image, rc.Config.ContainerArchitecture)
if err != nil {
return err
}
if anyArchExists && !correctArchExists {
wasRemoved, err := container.RemoveImage(ctx, image, true, true)
if err != nil {
return err
}
if !wasRemoved {
return fmt.Errorf("failed to remove image '%s'", image)
}
}
if !correctArchExists || rc.Config.ForceRebuild {
log.Debugf("image '%s' for architecture '%s' will be built from context '%s", image, rc.Config.ContainerArchitecture, contextDir)
var actionContainer container.Container
if localAction {
actionContainer = rc.JobContainer
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
}
prepImage = container.NewDockerBuildExecutor(container.NewDockerBuildExecutorInput{
ContextDir: contextDir,
ImageTag: image,
Container: actionContainer,
Platform: rc.Config.ContainerArchitecture,
})
} else {
log.Debugf("image '%s' for architecture '%s' already exists", image, rc.Config.ContainerArchitecture)
}
}
eval := rc.NewStepExpressionEvaluator(step)
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
cmd, err := shellquote.Split(eval.Interpolate(step.getStepModel().With["args"]))
if err != nil {
return err
}
if len(cmd) == 0 {
cmd = action.Runs.Args
evalDockerArgs(step, action, &cmd)
}
entrypoint := strings.Fields(eval.Interpolate(step.getStepModel().With["entrypoint"]))
if len(entrypoint) == 0 {
if action.Runs.Entrypoint != "" {
entrypoint, err = shellquote.Split(action.Runs.Entrypoint)
if err != nil {
return err
}
} else {
entrypoint = nil
}
}
stepContainer := newStepContainer(ctx, step, image, cmd, entrypoint)
return common.NewPipelineExecutor(
prepImage,
stepContainer.Pull(rc.Config.ForcePull),
stepContainer.Remove().IfBool(!rc.Config.ReuseContainers),
stepContainer.Create(rc.Config.ContainerCapAdd, rc.Config.ContainerCapDrop),
stepContainer.Start(true),
).Finally(
stepContainer.Remove().IfBool(!rc.Config.ReuseContainers),
).Finally(stepContainer.Close())(ctx)
}
func evalDockerArgs(step step, action *model.Action, cmd *[]string) {
rc := step.getRunContext()
stepModel := step.getStepModel()
oldInputs := rc.Inputs
defer func() {
rc.Inputs = oldInputs
}()
inputs := make(map[string]interface{})
eval := rc.NewExpressionEvaluator()
// Set Defaults
for k, input := range action.Inputs {
inputs[k] = eval.Interpolate(input.Default)
}
if stepModel.With != nil {
for k, v := range stepModel.With {
inputs[k] = eval.Interpolate(v)
}
}
rc.Inputs = inputs
stepEE := rc.NewStepExpressionEvaluator(step)
for i, v := range *cmd {
(*cmd)[i] = stepEE.Interpolate(v)
}
mergeIntoMap(step.getEnv(), action.Runs.Env)
ee := rc.NewStepExpressionEvaluator(step)
for k, v := range *step.getEnv() {
(*step.getEnv())[k] = ee.Interpolate(v)
}
}
func newStepContainer(ctx context.Context, step step, image string, cmd []string, entrypoint []string) container.Container {
rc := step.getRunContext()
stepModel := step.getStepModel()
rawLogger := common.Logger(ctx).WithField("raw_output", true)
logWriter := common.NewLineWriter(rc.commandHandler(ctx), func(s string) bool {
if rc.Config.LogOutput {
rawLogger.Infof("%s", s)
} else {
rawLogger.Debugf("%s", s)
}
return true
})
envList := make([]string, 0)
for k, v := range *step.getEnv() {
envList = append(envList, fmt.Sprintf("%s=%s", k, v))
}
envList = append(envList, fmt.Sprintf("%s=%s", "RUNNER_TOOL_CACHE", "/opt/hostedtoolcache"))
envList = append(envList, fmt.Sprintf("%s=%s", "RUNNER_OS", "Linux"))
envList = append(envList, fmt.Sprintf("%s=%s", "RUNNER_TEMP", "/tmp"))
binds, mounts := rc.GetBindsAndMounts()
stepContainer := container.NewContainer(&container.NewContainerInput{
Cmd: cmd,
Entrypoint: entrypoint,
WorkingDir: rc.Config.ContainerWorkdir(),
Image: image,
Username: rc.Config.Secrets["DOCKER_USERNAME"],
Password: rc.Config.Secrets["DOCKER_PASSWORD"],
Name: createContainerName(rc.jobContainerName(), stepModel.ID),
Env: envList,
Mounts: mounts,
NetworkMode: fmt.Sprintf("container:%s", rc.jobContainerName()),
Binds: binds,
Stdout: logWriter,
Stderr: logWriter,
Privileged: rc.Config.Privileged,
UsernsMode: rc.Config.UsernsMode,
Platform: rc.Config.ContainerArchitecture,
})
return stepContainer
}
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
func execAsComposite(step actionStep, containerActionDir string) common.Executor {
rc := step.getRunContext()
action := step.getActionModel()
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
return func(ctx context.Context) error {
// Disable some features of composite actions, only for feature parity with github
for _, compositeStep := range action.Runs.Steps {
if err := compositeStep.Validate(rc.Config.CompositeRestrictions); err != nil {
return err
}
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
}
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
eval := rc.NewExpressionEvaluator()
inputs := make(map[string]interface{})
for k, input := range action.Inputs {
inputs[k] = eval.Interpolate(input.Default)
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
}
if step.getStepModel().With != nil {
for k, v := range step.getStepModel().With {
inputs[k] = eval.Interpolate(v)
}
}
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
env := make(map[string]string)
for k, v := range rc.Env {
env[k] = eval.Interpolate(v)
}
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
for k, v := range step.getStepModel().Environment() {
env[k] = eval.Interpolate(v)
}
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
// run with the global config but without secrets
configCopy := *rc.Config
configCopy.Secrets = nil
// create a run context for the composite action to run in
compositerc := &RunContext{
Name: rc.Name,
JobName: rc.JobName,
Run: &model.Run{
JobID: "composite-job",
Workflow: &model.Workflow{
Name: rc.Run.Workflow.Name,
Jobs: map[string]*model.Job{
"composite-job": {},
},
},
},
Config: &configCopy,
StepResults: map[string]*model.StepResult{},
JobContainer: rc.JobContainer,
Inputs: inputs,
ActionPath: containerActionDir,
ActionRepository: rc.ActionRepository,
ActionRef: rc.ActionRef,
Env: env,
Masks: rc.Masks,
ExtraPath: rc.ExtraPath,
}
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
ctx = WithCompositeLogger(ctx, &compositerc.Masks)
// We need to inject a composite RunContext related command
// handler into the current running job container
// We need this, to support scoping commands to the composite action
// executing.
rawLogger := common.Logger(ctx).WithField("raw_output", true)
logWriter := common.NewLineWriter(compositerc.commandHandler(ctx), func(s string) bool {
if rc.Config.LogOutput {
rawLogger.Infof("%s", s)
} else {
rawLogger.Debugf("%s", s)
}
return true
})
oldout, olderr := compositerc.JobContainer.ReplaceLogWriter(logWriter, logWriter)
defer (func() {
rc.JobContainer.ReplaceLogWriter(oldout, olderr)
})()
err := compositerc.compositeExecutor(action)(ctx)
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
// Map outputs from composite RunContext to job RunContext
eval = compositerc.NewExpressionEvaluator()
for outputName, output := range action.Outputs {
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
rc.setOutput(ctx, map[string]string{
"name": outputName,
}, eval.Interpolate(output.Value))
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
}
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
rc.Masks = compositerc.Masks
rc.ExtraPath = compositerc.ExtraPath
return err
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
}
}
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
// Executor returns a pipeline executor for all the steps in the job
func (rc *RunContext) compositeExecutor(action *model.Action) common.Executor {
steps := make([]common.Executor, 0)
sf := &stepFactoryImpl{}
for i, step := range action.Runs.Steps {
if step.ID == "" {
step.ID = fmt.Sprintf("%d", i)
}
// create a copy of the step, since this composite action could
// run multiple times and we might modify the instance
stepcopy := step
step, err := sf.newStep(&stepcopy, rc)
if err != nil {
return common.NewErrorExecutor(err)
}
stepExec := common.NewPipelineExecutor(step.pre(), step.main(), step.post())
steps = append(steps, func(ctx context.Context) error {
err := stepExec(ctx)
if err != nil {
common.Logger(ctx).Errorf("%v", err)
common.SetJobError(ctx, err)
} else if ctx.Err() != nil {
common.Logger(ctx).Errorf("%v", ctx.Err())
common.SetJobError(ctx, ctx.Err())
}
return nil
})
}
steps = append(steps, common.JobError)
return func(ctx context.Context) error {
return common.NewPipelineExecutor(steps...)(common.WithJobErrorContainer(ctx))
}
}
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
func populateEnvsFromInput(env *map[string]string, action *model.Action, rc *RunContext) {
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
eval := rc.NewExpressionEvaluator()
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
for inputID, input := range action.Inputs {
envKey := regexp.MustCompile("[^A-Z0-9-]").ReplaceAllString(strings.ToUpper(inputID), "_")
envKey = fmt.Sprintf("INPUT_%s", envKey)
if _, ok := (*env)[envKey]; !ok {
refactor: remove composite action runcontext workaround (#1085) * refactor: remove composite action runcontext workaround The RunContext is cloned to execute a composite action with all its steps in a similar context. This required some workaround, since the command handler has kept a reference to the original RunContext. This is solved now, by replacing the docker LogWriter with a proper scoped LogWriter. This prepares for a simpler setup of composite actions to be able to create and re-create the composite RunContext for pre/main/post action steps. * test: check env-vars for local js and docker actions * test: test remote docker and js actions * fix: merge github context into env when read and setup * refacotr: simplify composite context setup * test: use a map matcher to test input setup * fix: restore composite log output Since we create a new line writer, we need to log the raw_output as well. Otherwise no output will be available from the log-writer * fix: add RunContext JobName to fill GITHUB_JOBNAME * test: use nektos/act-test-actions * fix: allow masking values in composite actions To allow masking of values from composite actions, we need to use a custom job logger with a reference to the masked values for the composite run context. * refactor: keep existing logger for composite actions To not introduce another new logger while still be able to use the masking from the composite action, we add the masks to the go context. To leverage that context, we also add the context to the log entries where the valueMasker then could get the actual mask values. With this way to 'inject' the masked values into the logger, we do - keep the logger - keep the coloring - stay away from inconsistencies due to parallel jobs * fix: re-add removed color increase This one should have never removed :-) * fix: add missing ExtraPath attribute * fix: merge run context env into composite run context env This adds a test and fix for the parent environment. It should be inherited by the composite environment. * test: add missing test case * fix: store github token next to secrets We must not expose the secrets to composite actions, but the `github.token` is available inside composite actions. To provide this we store the token in the config and create it in the GithubContext from there. The token can be used with `github.token` but is not available as `secrets.GITHUB_TOKEN`. This implements the same behavior as on GitHub. Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> * fixup! fix: allow masking values in composite actions * style: use tabs instead of spaces to fix linter errors Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <markus.noll@new-work.se> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2022-05-11 14:06:05 -05:00
(*env)[envKey] = eval.Interpolate(input.Default)
feat: split job steps into its own files/structs (#1004) * refactor: split step_context into separate files This commit moves functions from the step_context.go file into different files, but does otherwise not change anything. This is done to make it easier to review the changes made to these functions in the next commit, where we introduce a step factory to facilitate better unit testing of steps. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * refactor: introduce step factory and make steps testable With this commit we're introducing the `stepFactory` and interfaces and implementations for each different kind of step (run, docker, local and remote actions). Separating each step kind into its own interface and implementation makes it easier to reason about and to change behaviour of the step. By introducing interfaces we enable better unit testability as now each step implementation, the step factory and the job executor can be tested on their own by mocking out parts that are irrelevant. This commits prepares us for implementing pre/post actions in a later PR. Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> * fix: run post steps in reverse order * test: add missing asserts for mocks * refactor: use local reference instead of function This may make code more easy to follow. * refactor: correct typo in function name * test: use named structs * test: only expected valid calls There are mocks which are only called on certain conditions. * refactor: use step-model to get step name Using the step-logger we have to get the logger name from the step model. * test: only mock stopContainer if required Co-authored-by: Björn Brauer <bjoern.brauer@new-work.se> Co-authored-by: Marcus Noll <marcus.noll@new-work.se> Co-authored-by: Robert Kowalski <robert.kowalski@new-work.se> Co-authored-by: Philipp Hinrichsen <philipp.hinrichsen@new-work.se> Co-authored-by: Jonas Holland <jonas.holland@new-work.se> Co-authored-by: Casey Lee <cplee@nektos.com> Co-authored-by: Christopher Homberger <christopher.homberger@web.de>
2022-03-22 16:13:00 -05:00
}
}
}
func getContainerActionPaths(step *model.Step, actionDir string, rc *RunContext) (string, string) {
actionName := ""
containerActionDir := "."
if step.Type() != model.StepTypeUsesActionRemote {
actionName = getOsSafeRelativePath(actionDir, rc.Config.Workdir)
containerActionDir = rc.Config.ContainerWorkdir() + "/" + actionName
actionName = "./" + actionName
} else if step.Type() == model.StepTypeUsesActionRemote {
actionName = getOsSafeRelativePath(actionDir, rc.ActionCacheDir())
containerActionDir = ActPath + "/actions/" + actionName
}
if actionName == "" {
actionName = filepath.Base(actionDir)
if runtime.GOOS == "windows" {
actionName = strings.ReplaceAll(actionName, "\\", "/")
}
}
return actionName, containerActionDir
}
func getOsSafeRelativePath(s, prefix string) string {
actionName := strings.TrimPrefix(s, prefix)
if runtime.GOOS == "windows" {
actionName = strings.ReplaceAll(actionName, "\\", "/")
}
actionName = strings.TrimPrefix(actionName, "/")
return actionName
}