Switch scalar to use get/set 32-byte arrays
This commit is contained in:
@@ -25,10 +25,10 @@ void static secp256k1_scalar_free(secp256k1_scalar_t *r);
|
||||
int static secp256k1_scalar_get_bits(const secp256k1_scalar_t *a, int offset, int count);
|
||||
|
||||
/** Set a scalar from a big endian byte array. */
|
||||
void static secp256k1_scalar_set_bin(secp256k1_scalar_t *r, const unsigned char *bin, int len, int *overflow);
|
||||
void static secp256k1_scalar_set_b32(secp256k1_scalar_t *r, const unsigned char *bin, int *overflow);
|
||||
|
||||
/** Convert a scalar to a byte array. */
|
||||
void static secp256k1_scalar_get_bin(unsigned char *bin, int len, const secp256k1_scalar_t* a);
|
||||
void static secp256k1_scalar_get_b32(unsigned char *bin, const secp256k1_scalar_t* a);
|
||||
|
||||
/** Add two scalars together (modulo the group order). */
|
||||
void static secp256k1_scalar_add(secp256k1_scalar_t *r, const secp256k1_scalar_t *a, const secp256k1_scalar_t *b);
|
||||
|
||||
Reference in New Issue
Block a user