mirror of
https://github.com/cheat/cheat.git
synced 2024-11-25 15:31:36 +01:00
80c91cbdee
Integrate `go-git` into the application, and use it to `git clone` cheatsheets when the installer runs. Previously, the installer required that `git` be installed on the system `PATH`, so this change has to big advantages: 1. It removes that system dependency on `git` 2. It paves the way for implementing the `--update` command Additionally, `cheat` now performs a `--depth=1` clone when installing cheatsheets, which should at least somewhat improve installation times (especially on slow network connections).
196 lines
4.3 KiB
Go
196 lines
4.3 KiB
Go
package ed25519
|
|
|
|
import fp "github.com/cloudflare/circl/math/fp25519"
|
|
|
|
type (
|
|
pointR1 struct{ x, y, z, ta, tb fp.Elt }
|
|
pointR2 struct {
|
|
pointR3
|
|
z2 fp.Elt
|
|
}
|
|
)
|
|
type pointR3 struct{ addYX, subYX, dt2 fp.Elt }
|
|
|
|
func (P *pointR1) neg() {
|
|
fp.Neg(&P.x, &P.x)
|
|
fp.Neg(&P.ta, &P.ta)
|
|
}
|
|
|
|
func (P *pointR1) SetIdentity() {
|
|
P.x = fp.Elt{}
|
|
fp.SetOne(&P.y)
|
|
fp.SetOne(&P.z)
|
|
P.ta = fp.Elt{}
|
|
P.tb = fp.Elt{}
|
|
}
|
|
|
|
func (P *pointR1) toAffine() {
|
|
fp.Inv(&P.z, &P.z)
|
|
fp.Mul(&P.x, &P.x, &P.z)
|
|
fp.Mul(&P.y, &P.y, &P.z)
|
|
fp.Modp(&P.x)
|
|
fp.Modp(&P.y)
|
|
fp.SetOne(&P.z)
|
|
P.ta = P.x
|
|
P.tb = P.y
|
|
}
|
|
|
|
func (P *pointR1) ToBytes(k []byte) error {
|
|
P.toAffine()
|
|
var x [fp.Size]byte
|
|
err := fp.ToBytes(k[:fp.Size], &P.y)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
err = fp.ToBytes(x[:], &P.x)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
b := x[0] & 1
|
|
k[paramB-1] = k[paramB-1] | (b << 7)
|
|
return nil
|
|
}
|
|
|
|
func (P *pointR1) FromBytes(k []byte) bool {
|
|
if len(k) != paramB {
|
|
panic("wrong size")
|
|
}
|
|
signX := k[paramB-1] >> 7
|
|
copy(P.y[:], k[:fp.Size])
|
|
P.y[fp.Size-1] &= 0x7F
|
|
p := fp.P()
|
|
if !isLessThan(P.y[:], p[:]) {
|
|
return false
|
|
}
|
|
|
|
one, u, v := &fp.Elt{}, &fp.Elt{}, &fp.Elt{}
|
|
fp.SetOne(one)
|
|
fp.Sqr(u, &P.y) // u = y^2
|
|
fp.Mul(v, u, ¶mD) // v = dy^2
|
|
fp.Sub(u, u, one) // u = y^2-1
|
|
fp.Add(v, v, one) // v = dy^2+1
|
|
isQR := fp.InvSqrt(&P.x, u, v) // x = sqrt(u/v)
|
|
if !isQR {
|
|
return false
|
|
}
|
|
fp.Modp(&P.x) // x = x mod p
|
|
if fp.IsZero(&P.x) && signX == 1 {
|
|
return false
|
|
}
|
|
if signX != (P.x[0] & 1) {
|
|
fp.Neg(&P.x, &P.x)
|
|
}
|
|
P.ta = P.x
|
|
P.tb = P.y
|
|
fp.SetOne(&P.z)
|
|
return true
|
|
}
|
|
|
|
// double calculates 2P for curves with A=-1.
|
|
func (P *pointR1) double() {
|
|
Px, Py, Pz, Pta, Ptb := &P.x, &P.y, &P.z, &P.ta, &P.tb
|
|
a, b, c, e, f, g, h := Px, Py, Pz, Pta, Px, Py, Ptb
|
|
fp.Add(e, Px, Py) // x+y
|
|
fp.Sqr(a, Px) // A = x^2
|
|
fp.Sqr(b, Py) // B = y^2
|
|
fp.Sqr(c, Pz) // z^2
|
|
fp.Add(c, c, c) // C = 2*z^2
|
|
fp.Add(h, a, b) // H = A+B
|
|
fp.Sqr(e, e) // (x+y)^2
|
|
fp.Sub(e, e, h) // E = (x+y)^2-A-B
|
|
fp.Sub(g, b, a) // G = B-A
|
|
fp.Sub(f, c, g) // F = C-G
|
|
fp.Mul(Pz, f, g) // Z = F * G
|
|
fp.Mul(Px, e, f) // X = E * F
|
|
fp.Mul(Py, g, h) // Y = G * H, T = E * H
|
|
}
|
|
|
|
func (P *pointR1) mixAdd(Q *pointR3) {
|
|
fp.Add(&P.z, &P.z, &P.z) // D = 2*z1
|
|
P.coreAddition(Q)
|
|
}
|
|
|
|
func (P *pointR1) add(Q *pointR2) {
|
|
fp.Mul(&P.z, &P.z, &Q.z2) // D = 2*z1*z2
|
|
P.coreAddition(&Q.pointR3)
|
|
}
|
|
|
|
// coreAddition calculates P=P+Q for curves with A=-1.
|
|
func (P *pointR1) coreAddition(Q *pointR3) {
|
|
Px, Py, Pz, Pta, Ptb := &P.x, &P.y, &P.z, &P.ta, &P.tb
|
|
addYX2, subYX2, dt2 := &Q.addYX, &Q.subYX, &Q.dt2
|
|
a, b, c, d, e, f, g, h := Px, Py, &fp.Elt{}, Pz, Pta, Px, Py, Ptb
|
|
fp.Mul(c, Pta, Ptb) // t1 = ta*tb
|
|
fp.Sub(h, Py, Px) // y1-x1
|
|
fp.Add(b, Py, Px) // y1+x1
|
|
fp.Mul(a, h, subYX2) // A = (y1-x1)*(y2-x2)
|
|
fp.Mul(b, b, addYX2) // B = (y1+x1)*(y2+x2)
|
|
fp.Mul(c, c, dt2) // C = 2*D*t1*t2
|
|
fp.Sub(e, b, a) // E = B-A
|
|
fp.Add(h, b, a) // H = B+A
|
|
fp.Sub(f, d, c) // F = D-C
|
|
fp.Add(g, d, c) // G = D+C
|
|
fp.Mul(Pz, f, g) // Z = F * G
|
|
fp.Mul(Px, e, f) // X = E * F
|
|
fp.Mul(Py, g, h) // Y = G * H, T = E * H
|
|
}
|
|
|
|
func (P *pointR1) oddMultiples(T []pointR2) {
|
|
var R pointR2
|
|
n := len(T)
|
|
T[0].fromR1(P)
|
|
_2P := *P
|
|
_2P.double()
|
|
R.fromR1(&_2P)
|
|
for i := 1; i < n; i++ {
|
|
P.add(&R)
|
|
T[i].fromR1(P)
|
|
}
|
|
}
|
|
|
|
func (P *pointR1) isEqual(Q *pointR1) bool {
|
|
l, r := &fp.Elt{}, &fp.Elt{}
|
|
fp.Mul(l, &P.x, &Q.z)
|
|
fp.Mul(r, &Q.x, &P.z)
|
|
fp.Sub(l, l, r)
|
|
b := fp.IsZero(l)
|
|
fp.Mul(l, &P.y, &Q.z)
|
|
fp.Mul(r, &Q.y, &P.z)
|
|
fp.Sub(l, l, r)
|
|
b = b && fp.IsZero(l)
|
|
fp.Mul(l, &P.ta, &P.tb)
|
|
fp.Mul(l, l, &Q.z)
|
|
fp.Mul(r, &Q.ta, &Q.tb)
|
|
fp.Mul(r, r, &P.z)
|
|
fp.Sub(l, l, r)
|
|
b = b && fp.IsZero(l)
|
|
return b
|
|
}
|
|
|
|
func (P *pointR3) neg() {
|
|
P.addYX, P.subYX = P.subYX, P.addYX
|
|
fp.Neg(&P.dt2, &P.dt2)
|
|
}
|
|
|
|
func (P *pointR2) fromR1(Q *pointR1) {
|
|
fp.Add(&P.addYX, &Q.y, &Q.x)
|
|
fp.Sub(&P.subYX, &Q.y, &Q.x)
|
|
fp.Mul(&P.dt2, &Q.ta, &Q.tb)
|
|
fp.Mul(&P.dt2, &P.dt2, ¶mD)
|
|
fp.Add(&P.dt2, &P.dt2, &P.dt2)
|
|
fp.Add(&P.z2, &Q.z, &Q.z)
|
|
}
|
|
|
|
func (P *pointR3) cneg(b int) {
|
|
t := &fp.Elt{}
|
|
fp.Cswap(&P.addYX, &P.subYX, uint(b))
|
|
fp.Neg(t, &P.dt2)
|
|
fp.Cmov(&P.dt2, t, uint(b))
|
|
}
|
|
|
|
func (P *pointR3) cmov(Q *pointR3, b int) {
|
|
fp.Cmov(&P.addYX, &Q.addYX, uint(b))
|
|
fp.Cmov(&P.subYX, &Q.subYX, uint(b))
|
|
fp.Cmov(&P.dt2, &Q.dt2, uint(b))
|
|
}
|