bend-collections

Releases (1)

• 1.0.0.0 indexed

Definitions (4,483)

src/crypto/sha512/core.bend

• def choose
• def majority
• def step
• def feedforward
• def nextW[t] from W[t-2], W[t-7], W[t-15], W[t-16].
• def window_roundsThe remaining q rounds (64 after the block's own sixteen): each derives its schedule word from the window, then drops the oldest word.
• def block_roundsThe block's own words first, then q window rounds (q = 64 for SHA-512); win is the block reversed.
• def compress16
• def blocksEvery whole block of sixteen words; a shorter tail is ignored (padded messages have none).
• def pack
• def lanesBig-endian 64-bit words, eight bytes each.
• def base256
• def bit_length8n as sixteen big-endian octets: n div 32 in the top fifteen, 8 (n mod 32) last.
• def zeros_if
• def zeros
• def suffix0x80, zeros up to 112 mod 128, then the 128-bit bit length.
• def octets
• def lane_octets
• def digest_bytesThe 64-byte digest H0 ..
• def initial
• def round_constants
• def hash_padded
• def sha512

src/crypto/sha512/sha512.bend

• def sha512SHA-512 (FIPS 180-4).
• def ascii
• def hex_digit_if
• def hex_digit
• def hexLowercase hex of a byte list.

src/crypto/sha512/types.bend

• type LaneA 64-bit word as its two 32-bit halves, most significant first.
• type StateThe working variables a..h (and the chaining value H0..H7).

src/crypto/sign.bend

• type Keypair
• def keypair_if
• def generate_keypair
• def word_bytes
• def random_bytes
• def generate_keypair_os
• def sign_if
• def sign
• def verify_if
• def verify

src/crypto/subtle.bend

• def diffOR of the XOR of every pair of positions, over the common prefix.
• def eqTrue exactly when a and b are the same list.

src/math/f64.bend

• type F64
• def word_lo
• def word_hi
• def of_bits
• def off
• def sgn
• def signbit
• def exp_field
• def frac
• def mag
• def nz
• def is_nan
• def is_inf
• def is_finite
• def is_zero
• def nan
• def inf
• def zero
• def one
• def neg
• def abs
• def copysign
• def nan_or
• def pack64
• def packsign, exponent field e and significand m; the hidden bit of m (bit 52) carries into the exponent field, as packToF64's addition does
• def zero_e
• def rp_fin3
• def rp_fin2
• def rp_finsig has its top bit at 62 and e is the plain exponent (0 <= e <= 0x7FD): add half an ulp of the kept 53 bits, shift, and clear bit 0 on a tie
• def rp_over
• def rp_neg
• def round_packe is the biased exponent minus one, offset by OFF; sig has its top bit at 62
• def rp62_picksig below 2^62 is shifted up once
• def rp62
• def nrp_exp
• def nrp_pick
• def nrp_sd
• def norm_round_packSoftFloat's normRoundPackToF64 (e offset)
• def norm_ethe offset exponent and 53-bit significand (hidden bit at 52) of a nonzero finite value, subnormals normalized (SoftFloat's normSubnormalF64Sig)
• def norm_f
• def am_small
• def am_big|x| + |y| with sign s, the larger exponent eL
• def am_eq_n
• def am_eq
• def am_case
• def add_mags
• def sm_small
• def sm_big|big| - |small| with sign s, the larger exponent eL
• def sm_exact3
• def sm_exactan exact difference of equal exponents, normalized without rounding
• def sm_eq2
• def cmp64
• def sm_eq
• def sm_case
• def sub_mags
• def add_pick
• def add
• def sub