go-task/task · error

invalid method

Error message

invalid method

What it means

ErrInvalidMethod is a sentinel error thrown by NewSourcesChecker when the fingerprint 'method' name on a task is not one of the supported values ('timestamp' or 'checksum' or 'none'). It lets callers that only need a fingerprint value distinguish a bad method name from a checker that failed on the sources themselves. It is compared with errors.Is / require.ErrorIs in tests and wrapped with the offending method name in the message.

Source

Thrown at internal/fingerprint/sources.go:11

package fingerprint

import (
	"fmt"

	"github.com/go-task/task/v3/errors"
)

// ErrInvalidMethod lets callers that only need a fingerprint value tell a bad
// method name apart from a checker failing on the sources themselves.
var ErrInvalidMethod = errors.New("invalid method")

func NewSourcesChecker(method, tempDir string, dry bool) (SourcesCheckable, error) {
	switch method {
	case "timestamp":
		return NewTimestampChecker(tempDir, dry), nil
	case "checksum":
		return NewChecksumChecker(tempDir, dry), nil
	case "none":
		return NoneChecker{}, nil
	default:
		return nil, fmt.Errorf(`task: %w "%s"`, ErrInvalidMethod, method)
	}
}

View on GitHub (pinned to 385e5ad92a)

Solutions

1. Set the task's `method:` to one of: none, checksum, timestamp (check spelling)
2. If calling NewSourcesChecker directly, validate the method string against the supported set before calling
3. Wrap the call and surface the wrapped method name (message contains the bad value) to find which task is at fault
4. Omit `method:` entirely to use the default (checksum)

Example fix

# before (Taskfile.yml)
method: sha256

# after
method: checksum

var validMethods = map[string]bool{"": true, "none": true, "checksum": true, "timestamp": true}
if !validMethods[method] {
	return fmt.Errorf("unsupported fingerprint method %q: use none, checksum, or timestamp", method)
}
checker, err := fingerprint.NewSourcesChecker(method, tempDir, dry)

func isValidMethod(m string) bool {
	switch m {
	case "", "none", "checksum", "timestamp":
		return true
	}
	return false
}

checker, err := fingerprint.NewSourcesChecker(method, tempDir, dry)
if errors.Is(err, fingerprint.ErrInvalidMethod) {
	return fmt.Errorf("task %q has an invalid method %q; use none, checksum, or timestamp", task.Name(), method)
}
if err != nil {
	return err
}

When it happens

Trigger: Calling NewSourcesChecker with a method string other than "timestamp", "checksum", or "none" (it hits the switch's default case). Also triggered indirectly via compiledTask building a checker for a task whose `method:` field is misspelled, and by f.SourceValue / f.UpToDate in tests using an invalid method.

Common situations: Typo in a Taskfile's `method:` attribute (e.g. `methdo: checksum`, `method: sha256`), copying a Taskfile using a method name from another tool, or programmatically passing an unvalidated user-supplied method string to NewSourcesChecker.

Understand the failure class

Background: Invalid enum value errors: "Unknown type", "Invalid scope", "must be one of" — when a string is not on the library's allowed list — this error's family across 23 libraries.

Related errors

• task: %w "%s"
• task: Included Taskfiles can't have dotenv declarations. Ple
• loop var must be a delimiter-separated string, list or a map
• task: invalid run "%s"
• task: output style %q not recognized

AI-assisted analysis of go-task/task@385e5ad92a (2026-09-05). Data as JSON: /api/errors/ccf96b9c8c44cdc4. Report an issue: GitHub.