1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466
use crate::*;
use alloc::format;
use core::convert::TryFrom;
mod f32;
mod f64;
/// ASN.1 `REAL` type
///
/// # Limitations
///
/// When encoding binary values, only base 2 is supported
#[derive(Debug, PartialEq)]
pub enum Real {
/// Non-special values
Binary {
mantissa: f64,
base: u32,
exponent: i32,
enc_base: u8,
},
/// Infinity (∞).
Infinity,
/// Negative infinity (−∞).
NegInfinity,
/// Zero
Zero,
}
impl Real {
/// Create a new `REAL` from the `f64` value.
pub fn new(f: f64) -> Self {
if f.is_infinite() {
if f.is_sign_positive() {
Self::Infinity
} else {
Self::NegInfinity
}
} else if f.abs() == 0.0 {
Self::Zero
} else {
let mut e = 0;
let mut f = f;
while f.fract() != 0.0 {
f *= 10.0_f64;
e -= 1;
}
Real::Binary {
mantissa: f,
base: 10,
exponent: e,
enc_base: 10,
}
.normalize_base10()
}
}
pub const fn with_enc_base(self, enc_base: u8) -> Self {
match self {
Real::Binary {
mantissa,
base,
exponent,
..
} => Real::Binary {
mantissa,
base,
exponent,
enc_base,
},
e => e,
}
}
fn normalize_base10(self) -> Self {
match self {
Real::Binary {
mantissa,
base: 10,
exponent,
enc_base: _enc_base,
} => {
let mut m = mantissa;
let mut e = exponent;
while m.abs() > f64::EPSILON && m.rem_euclid(10.0).abs() < f64::EPSILON {
m /= 10.0;
e += 1;
}
Real::Binary {
mantissa: m,
base: 10,
exponent: e,
enc_base: _enc_base,
}
}
_ => self,
}
}
/// Create a new binary `REAL`
#[inline]
pub const fn binary(mantissa: f64, base: u32, exponent: i32) -> Self {
Self::Binary {
mantissa,
base,
exponent,
enc_base: 2,
}
}
/// Returns `true` if this value is positive infinity or negative infinity, and
/// `false` otherwise.
#[inline]
pub fn is_infinite(&self) -> bool {
matches!(self, Real::Infinity | Real::NegInfinity)
}
/// Returns `true` if this number is not infinite.
#[inline]
pub fn is_finite(&self) -> bool {
matches!(self, Real::Zero | Real::Binary { .. })
}
/// Returns the 'f64' value of this `REAL`.
///
/// Returned value is a float, and may be infinite.
pub fn f64(&self) -> f64 {
match self {
Real::Binary {
mantissa,
base,
exponent,
..
} => {
let f = mantissa;
let exp = (*base as f64).powi(*exponent);
f * exp
}
Real::Zero => 0.0_f64,
Real::Infinity => f64::INFINITY,
Real::NegInfinity => f64::NEG_INFINITY,
}
}
/// Returns the 'f32' value of this `REAL`.
///
/// This functions casts the result of [`Real::f64`] to a `f32`, and loses precision.
pub fn f32(&self) -> f32 {
self.f64() as f32
}
}
impl<'a> TryFrom<Any<'a>> for Real {
type Error = Error;
fn try_from(any: Any<'a>) -> Result<Self> {
TryFrom::try_from(&any)
}
}
impl<'a, 'b> TryFrom<&'b Any<'a>> for Real {
type Error = Error;
fn try_from(any: &'b Any<'a>) -> Result<Self> {
any.tag().assert_eq(Self::TAG)?;
any.header.assert_primitive()?;
let data = &any.data;
if data.is_empty() {
return Ok(Real::Zero);
}
// code inspired from pyasn1
let first = data[0];
let rem = &data[1..];
if first & 0x80 != 0 {
// binary encoding (X.690 section 8.5.6)
// format of exponent
let (n, rem) = match first & 0x03 {
4 => {
let (b, rem) = rem
.split_first()
.ok_or_else(|| Error::Incomplete(Needed::new(1)))?;
(*b as usize, rem)
}
b => (b as usize + 1, rem),
};
if n >= rem.len() {
return Err(any.tag().invalid_value("Invalid float value(exponent)"));
}
// n cannot be 0 (see the +1 above)
let (eo, rem) = rem.split_at(n);
// so 'eo' cannot be empty
let mut e = if eo[0] & 0x80 != 0 { -1 } else { 0 };
// safety check: 'eo' length must be <= container type for 'e'
if eo.len() > 4 {
return Err(any.tag().invalid_value("Exponent too large (REAL)"));
}
for b in eo {
e = (e << 8) | (*b as i32);
}
// base bits
let b = (first >> 4) & 0x03;
let _enc_base = match b {
0 => 2,
1 => 8,
2 => 16,
_ => return Err(any.tag().invalid_value("Illegal REAL encoding base")),
};
let e = match b {
// base 2
0 => e,
// base 8
1 => e * 3,
// base 16
2 => e * 4,
_ => return Err(any.tag().invalid_value("Illegal REAL base")),
};
if rem.len() > 8 {
return Err(any.tag().invalid_value("Mantissa too large (REAL)"));
}
let mut p = 0;
for b in rem {
p = (p << 8) | (*b as i64);
}
// sign bit
let p = if first & 0x40 != 0 { -p } else { p };
// scale bits
let sf = (first >> 2) & 0x03;
let p = match sf {
0 => p as f64,
sf => {
// 2^sf: cannot overflow, sf is between 0 and 3
let scale = 2_f64.powi(sf as _);
(p as f64) * scale
}
};
Ok(Real::Binary {
mantissa: p,
base: 2,
exponent: e,
enc_base: _enc_base,
})
} else if first & 0x40 != 0 {
// special real value (X.690 section 8.5.8)
// there shall be only one contents octet,
if any.header.length != Length::Definite(1) {
return Err(Error::InvalidLength);
}
// with values as follows
match first {
0x40 => Ok(Real::Infinity),
0x41 => Ok(Real::NegInfinity),
_ => Err(any.tag().invalid_value("Invalid float special value")),
}
} else {
// decimal encoding (X.690 section 8.5.7)
let s = alloc::str::from_utf8(rem)?;
match first & 0x03 {
0x1 => {
// NR1
match s.parse::<u32>() {
Err(_) => Err(any.tag().invalid_value("Invalid float string encoding")),
Ok(v) => Ok(Real::new(v.into())),
}
}
0x2 /* NR2 */ | 0x3 /* NR3 */=> {
match s.parse::<f64>() {
Err(_) => Err(any.tag().invalid_value("Invalid float string encoding")),
Ok(v) => Ok(Real::new(v)),
}
}
c => {
Err(any.tag().invalid_value(&format!("Invalid NR ({})", c)))
}
}
}
}
}
impl CheckDerConstraints for Real {
fn check_constraints(any: &Any) -> Result<()> {
any.header.assert_primitive()?;
any.header.length.assert_definite()?;
// XXX more checks
Ok(())
}
}
impl DerAutoDerive for Real {}
impl Tagged for Real {
const TAG: Tag = Tag::RealType;
}
#[cfg(feature = "std")]
impl ToDer for Real {
fn to_der_len(&self) -> Result<usize> {
match self {
Real::Zero => Ok(0),
Real::Infinity | Real::NegInfinity => Ok(1),
Real::Binary { .. } => {
let mut sink = std::io::sink();
let n = self
.write_der_content(&mut sink)
.map_err(|_| Self::TAG.invalid_value("Serialization of REAL failed"))?;
Ok(n)
}
}
}
fn write_der_header(&self, writer: &mut dyn std::io::Write) -> SerializeResult<usize> {
let header = Header::new(
Class::Universal,
false,
Self::TAG,
Length::Definite(self.to_der_len()?),
);
header.write_der_header(writer).map_err(Into::into)
}
fn write_der_content(&self, writer: &mut dyn std::io::Write) -> SerializeResult<usize> {
match self {
Real::Zero => Ok(0),
Real::Infinity => writer.write(&[0x40]).map_err(Into::into),
Real::NegInfinity => writer.write(&[0x41]).map_err(Into::into),
Real::Binary {
mantissa,
base,
exponent,
enc_base: _enc_base,
} => {
if *base == 10 {
// using character form
let sign = if *exponent == 0 { "+" } else { "" };
let s = format!("\x03{}E{}{}", mantissa, sign, exponent);
return writer.write(s.as_bytes()).map_err(Into::into);
}
if *base != 2 {
return Err(Self::TAG.invalid_value("Invalid base for REAL").into());
}
let mut first: u8 = 0x80;
// choose encoding base
let enc_base = *_enc_base;
let (ms, mut m, enc_base, mut e) =
drop_floating_point(*mantissa, enc_base, *exponent);
assert!(m != 0);
if ms < 0 {
first |= 0x40
};
// exponent & mantissa normalization
match enc_base {
2 => {
while m & 0x1 == 0 {
m >>= 1;
e += 1;
}
}
8 => {
while m & 0x7 == 0 {
m >>= 3;
e += 1;
}
first |= 0x10;
}
_ /* 16 */ => {
while m & 0xf == 0 {
m >>= 4;
e += 1;
}
first |= 0x20;
}
}
// scale factor
// XXX in DER, sf is always 0 (11.3.1)
let mut sf = 0;
while m & 0x1 == 0 && sf < 4 {
m >>= 1;
sf += 1;
}
first |= sf << 2;
// exponent length and bytes
let len_e = match e.abs() {
0..=0xff => 1,
0x100..=0xffff => 2,
0x1_0000..=0xff_ffff => 3,
// e is an `i32` so it can't be longer than 4 bytes
// use 4, so `first` is ORed with 3
_ => 4,
};
first |= (len_e - 1) & 0x3;
// write first byte
let mut n = writer.write(&[first])?;
// write exponent
// special case: number of bytes from exponent is > 3 and cannot fit in 2 bits
#[allow(clippy::identity_op)]
if len_e == 4 {
let b = len_e & 0xff;
n += writer.write(&[b])?;
}
// we only need to write e.len() bytes
let bytes = e.to_be_bytes();
n += writer.write(&bytes[(4 - len_e) as usize..])?;
// write mantissa
let bytes = m.to_be_bytes();
let mut idx = 0;
for &b in bytes.iter() {
if b != 0 {
break;
}
idx += 1;
}
n += writer.write(&bytes[idx..])?;
Ok(n)
}
}
}
}
impl From<f32> for Real {
fn from(f: f32) -> Self {
Real::new(f.into())
}
}
impl From<f64> for Real {
fn from(f: f64) -> Self {
Real::new(f)
}
}
impl From<Real> for f32 {
fn from(r: Real) -> Self {
r.f32()
}
}
impl From<Real> for f64 {
fn from(r: Real) -> Self {
r.f64()
}
}
#[cfg(feature = "std")]
fn drop_floating_point(m: f64, b: u8, e: i32) -> (i8, u64, u8, i32) {
let ms = if m.is_sign_positive() { 1 } else { -1 };
let es = if e.is_positive() { 1 } else { -1 };
let mut m = m.abs();
let mut e = e;
//
if b == 8 {
m *= 2_f64.powi((e.abs() / 3) * es);
e = (e.abs() / 3) * es;
} else if b == 16 {
m *= 2_f64.powi((e.abs() / 4) * es);
e = (e.abs() / 4) * es;
}
//
while m.abs() > f64::EPSILON {
if m.fract() != 0.0 {
m *= b as f64;
e -= 1;
} else {
break;
}
}
(ms, m as u64, b, e)
}