Documentation
¶
Overview ¶
Package eval provides builtin evaluation suites.
Package eval provides evaluation framework for hawk-eco. It supports standardized test suites, agent evals, and scoring.
Index ¶
- func GetFixture(name string) (string, bool)
- func Materialize(test *Test) string
- func Now() int64
- func RunCommand(cmd string) (string, error)
- func RunShellCommand(cmd string) (string, error)
- type ExpectedResult
- type Report
- type Result
- type Suite
- type SuiteManager
- func (m *SuiteManager) FindSuite(name string) *Suite
- func (m *SuiteManager) List() []*Suite
- func (m *SuiteManager) LoadDefaultSuites() error
- func (m *SuiteManager) LoadFixture(name string) (string, bool)
- func (m *SuiteManager) LoadSuitesFromDir(dir string) error
- func (m *SuiteManager) Register(suite *Suite)
- func (m *SuiteManager) RegisterFixture(name, content string)
- func (m *SuiteManager) RunAllSuites() []*Report
- func (m *SuiteManager) RunSuite(name string) *Report
- func (m *SuiteManager) RunTest(test *Test) *Result
- func (m *SuiteManager) Summarize(results []*Result, totalDuration int64) *Summary
- type Summary
- type Test
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func GetFixture ¶
GetFixture retrieves a fixture by name.
func Materialize ¶
Materialize builds a command with fixtures substituted.
func RunCommand ¶
RunCommand runs a command and returns output.
func RunShellCommand ¶
RunShellCommand runs a shell command.
Types ¶
type ExpectedResult ¶
ExpectedResult represents expected test output.
type SuiteManager ¶
type SuiteManager struct {
// contains filtered or unexported fields
}
SuiteManager manages evaluation suites.
func NewSuiteManager ¶
func NewSuiteManager() *SuiteManager
NewSuiteManager creates a new suite manager.
func (*SuiteManager) FindSuite ¶
func (m *SuiteManager) FindSuite(name string) *Suite
FindSuite finds a suite by name.
func (*SuiteManager) List ¶
func (m *SuiteManager) List() []*Suite
List returns all registered suites.
func (*SuiteManager) LoadDefaultSuites ¶
func (m *SuiteManager) LoadDefaultSuites() error
LoadDefaultSuites loads suites from default locations.
func (*SuiteManager) LoadFixture ¶
func (m *SuiteManager) LoadFixture(name string) (string, bool)
LoadFixture loads a fixture file.
func (*SuiteManager) LoadSuitesFromDir ¶
func (m *SuiteManager) LoadSuitesFromDir(dir string) error
LoadSuitesFromDir loads suites from a directory.
func (*SuiteManager) Register ¶
func (m *SuiteManager) Register(suite *Suite)
Register adds a suite to the manager.
func (*SuiteManager) RegisterFixture ¶
func (m *SuiteManager) RegisterFixture(name, content string)
RegisterFixture registers a fixture.
func (*SuiteManager) RunAllSuites ¶
func (m *SuiteManager) RunAllSuites() []*Report
RunAllSuites runs all registered suites.
func (*SuiteManager) RunSuite ¶
func (m *SuiteManager) RunSuite(name string) *Report
RunSuite runs all tests in a suite.
func (*SuiteManager) RunTest ¶
func (m *SuiteManager) RunTest(test *Test) *Result
RunTest runs a single test.