Documentation
¶
Overview ¶
Package encryptiontest provides support for testing encryption and associated key management.
Index ¶
- Constants
- Variables
- func ExpectedRuns(n1, n2 int) float64
- func IsRandom(buf []byte, significance Significance) bool
- func IsTestStatisticRandom(t float64, significance Significance) bool
- func Median(data []byte) byte
- func MedianRuns(data []byte) (above, below, runs int)
- func RunAtSignificanceLevel(significance Significance, fn func(s Significance) bool) (failed int, expectedFailures int, iterations int, success bool)
- func RunsTestStatistic(runs, n1, n2 int) float64
- func Variance(n1, n2 int) float64
- type FakeAESRegistry
- func (fr *FakeAESRegistry) BlockSize() int
- func (fr *FakeAESRegistry) GenerateKey() (ID []byte, err error)
- func (fr *FakeAESRegistry) HMACSize() int
- func (fr *FakeAESRegistry) NewBlock(ID []byte, opts ...interface{}) (hmacHash hash.Hash, block cipher.Block, err error)
- func (fr *FakeAESRegistry) NewGCM(block cipher.Block, opts ...interface{}) (aead cipher.AEAD, err error)
- type Significance
Constants ¶
const MinDataSize = 20000
MinDataSize is a suggested minimum data size for the runstest to be meaingfully and reliably used.
Variables ¶
var ( // TestKey is the key used in tests. TestKey = []byte("0123456789abcdef") // TestID is the ID used in tests. TestID = []byte("0123456789abcdef") // BadID will generate an error if passed to NewBlock. BadID = []byte(strings.Repeat("0", 16)) // FailGenKey will generate an error when GenerateKey is called. FailGenKey = []byte(strings.Repeat("1", 16)) // FailNewGCMKey will generate an error when NewGCM is called. FailNewGCMKey = []byte(strings.Repeat("2", 16)) )
Functions ¶
func ExpectedRuns ¶
ExpectedRuns returns the approximation for the number of expected runs when n1 and n2 are the number of positive/negative values in a sample. expeted runs = 2 * n1 * n2 / n1 + n2 n1 and n2 must be >= 10.
func IsRandom ¶
func IsRandom(buf []byte, significance Significance) bool
IsRandom uses the runstest to determine if the supplied data is random at the requested significance level. Note, that, the test is expected to fail (at the rate implied by the sigfnificance level) even if the data is random. Hence any user of this test should take that into account; the RunAtSignificanceLevel function in this package is provided as a convenience for doing so.
func IsTestStatisticRandom ¶
func IsTestStatisticRandom(t float64, significance Significance) bool
IsTestStatisticRandom returns true if the suipplied test statistic value meets the criteria defined in http://www.itl.nist.gov/div898/handbook/eda/section3/eda35d.htm for being random at the requested significance level. Also see: http://influentialpoints.com/Training/runs_tests-principles-properties-assumptions.htm and note the assumptions:
- In its usual form, the test is not sensitive to departures from randomness for run lengths of two.
- The runs test has the wrong type error rate if used to evaluate the independence of errors in time-series regression models.
func MedianRuns ¶
MedianRuns returns the number of runs in the supplied byte slice where a run is defined as a sequence of consecutive positive (> median) or negative (< median) values. Median values are ignored.
func RunAtSignificanceLevel ¶
func RunAtSignificanceLevel(significance Significance, fn func(s Significance) bool) (failed int, expectedFailures int, iterations int, success bool)
RunAtSignificanceLevel runs the supplied function sufficient times to measure the false positive rate for the requested significance level. The supplied function should return the success of the test being measured. returns true if the function meets the requested significance level.
func RunsTestStatistic ¶
RunsTestStatistic calculates the 'runs test' stastistic as per: http://www.itl.nist.gov/div898/handbook/eda/section3/eda35d.htm and other sources.
func Variance ¶
Variance returns the approximation to the standard deviation of the number of runs where n1 and n2 are the number of positive/negative values in a sample. variance = 2 * n1 * n2 * (2 * n1 * n2 − n1 − n2)
-------------------------------------
(n1 + n2)^2 * (n1 + n2 − 1)
n1 and n2 must be >= 10
Types ¶
type FakeAESRegistry ¶
type FakeAESRegistry struct {
Key []byte
}
FakeAESRegistry represents a fake aes-128 (16-byte key) registry with hmac-sha512 for use with tests.
func NewFakeAESRegistry ¶
func NewFakeAESRegistry() *FakeAESRegistry
NewFakeAESRegistry returns a new FakeAESRegistry
func (*FakeAESRegistry) BlockSize ¶
func (fr *FakeAESRegistry) BlockSize() int
BlockSize implements encryted.KeyRegistry.
func (*FakeAESRegistry) GenerateKey ¶
func (fr *FakeAESRegistry) GenerateKey() (ID []byte, err error)
GenerateKey implements encrypted.KeyRegistry.
func (*FakeAESRegistry) HMACSize ¶
func (fr *FakeAESRegistry) HMACSize() int
HMACSize implements encryted.KeyRegistry.
type Significance ¶
type Significance float64
Significance represents the statistical significance level acceptable for a given test. Note that the significance level represents the chance of a false positive, so choosing the .2% level implies a .2% change of a false positive.
const ( // PointTwoPercent represents a 0.2% significance level. PointTwoPercent Significance = 0.999 // OnePercent represents a 1% significance level. OnePercent Significance = 0.995 // FivePercent represents a 5% significance level. FivePercent Significance = 0.975 )
func (Significance) String ¶
func (s Significance) String() string
Source Files