| 1 | // SPDX-License-Identifier: Apache-2.0 |
| 2 | |
| 3 | //! Low level CBOR parsing tools |
| 4 | //! |
| 5 | //! This crate contains low-level types for encoding and decoding items in |
| 6 | //! CBOR. This crate is usable in both `no_std` and `no_alloc` environments. |
| 7 | //! To understand how this crate works, first we will look at the structure |
| 8 | //! of a CBOR item on the wire. |
| 9 | //! |
| 10 | //! # Anatomy of a CBOR Item |
| 11 | //! |
| 12 | //! This is a brief anatomy of a CBOR item on the wire. |
| 13 | //! |
| 14 | //! ```text |
| 15 | //! +------------+-----------+ |
| 16 | //! | | | |
| 17 | //! | Major | Minor | |
| 18 | //! | (3bits) | (5bits) | |
| 19 | //! | | | |
| 20 | //! +------------+-----------+ |
| 21 | //! ^ ^ |
| 22 | //! | | |
| 23 | //! +-----+ +-----+ |
| 24 | //! | | |
| 25 | //! | | |
| 26 | //! +----------------------------+--------------+ |
| 27 | //! | | | | |
| 28 | //! | Prefix | Affix | Suffix | |
| 29 | //! | (1 byte) | (0-8 bytes) | (0+ bytes) | |
| 30 | //! | | | | |
| 31 | //! +------------+---------------+--------------+ |
| 32 | //! |
| 33 | //! | | | |
| 34 | //! +------------+---------------+--------------+ |
| 35 | //! | | |
| 36 | //! v v |
| 37 | //! |
| 38 | //! Header Body |
| 39 | //! ``` |
| 40 | //! |
| 41 | //! The `ciborium` crate works by providing the `Decoder` and `Encoder` types |
| 42 | //! which provide input and output for a CBOR header (see: `Header`). From |
| 43 | //! there, you can either handle the body yourself or use the provided utility |
| 44 | //! functions. |
| 45 | //! |
| 46 | //! For more information on the CBOR format, see |
| 47 | //! [RFC 7049](https://tools.ietf.org/html/rfc7049). |
| 48 | //! |
| 49 | //! # Decoding |
| 50 | //! |
| 51 | //! In order to decode CBOR, you will create a `Decoder` from a reader. The |
| 52 | //! decoder instance will allow you to `Decoder::pull()` `Header` instances |
| 53 | //! from the input. |
| 54 | //! |
| 55 | //! Most CBOR items are fully contained in their headers and therefore have no |
| 56 | //! body. These items can be evaluated directly from the `Header` instance. |
| 57 | //! |
| 58 | //! Bytes and text items have a body but do not contain child items. Since |
| 59 | //! both bytes and text values may be segmented, parsing them can be a bit |
| 60 | //! tricky. Therefore, we provide helper functions to parse these types. See |
| 61 | //! `Decoder::bytes()` and `Decoder::text()` for more details. |
| 62 | //! |
| 63 | //! Array and map items have a body which contains child items. These can be |
| 64 | //! parsed by simply doing `Decoder::pull()` to parse the child items. |
| 65 | //! |
| 66 | //! ## Example |
| 67 | //! |
| 68 | //! ```rust |
| 69 | //! use ciborium_ll::{Decoder, Header}; |
| 70 | //! use ciborium_io::Read as _; |
| 71 | //! |
| 72 | //! let input = b" \x6dHello, World!" ; |
| 73 | //! let mut decoder = Decoder::from(&input[..]); |
| 74 | //! let mut chunks = 0; |
| 75 | //! |
| 76 | //! match decoder.pull().unwrap() { |
| 77 | //! Header::Text(len) => { |
| 78 | //! let mut segments = decoder.text(len); |
| 79 | //! while let Some(mut segment) = segments.pull().unwrap() { |
| 80 | //! let mut buffer = [0u8; 7]; |
| 81 | //! while let Some(chunk) = segment.pull(&mut buffer[..]).unwrap() { |
| 82 | //! match chunk { |
| 83 | //! "Hello, " if chunks == 0 => chunks = 1, |
| 84 | //! "World!" if chunks == 1 => chunks = 2, |
| 85 | //! _ => panic!("received unexpected chunk" ), |
| 86 | //! } |
| 87 | //! } |
| 88 | //! } |
| 89 | //! } |
| 90 | //! |
| 91 | //! _ => panic!("received unexpected value" ), |
| 92 | //! } |
| 93 | //! |
| 94 | //! assert_eq!(chunks, 2); |
| 95 | //! ``` |
| 96 | //! |
| 97 | //! # Encoding |
| 98 | //! |
| 99 | //! To encode values to CBOR, create an `Encoder` from a writer. The encoder |
| 100 | //! instance provides the `Encoder::push()` method to write a `Header` value |
| 101 | //! to the wire. CBOR item bodies can be written directly. |
| 102 | //! |
| 103 | //! For bytes and text, there are the `Encoder::bytes()` and `Encoder::text()` |
| 104 | //! utility functions, respectively, which will properly segment the output |
| 105 | //! on the wire for you. |
| 106 | //! |
| 107 | //! ## Example |
| 108 | //! |
| 109 | //! ```rust |
| 110 | //! use ciborium_ll::{Encoder, Header}; |
| 111 | //! use ciborium_io::Write as _; |
| 112 | //! |
| 113 | //! let mut buffer = [0u8; 19]; |
| 114 | //! let mut encoder = Encoder::from(&mut buffer[..]); |
| 115 | //! |
| 116 | //! // Write the structure |
| 117 | //! encoder.push(Header::Map(Some(1))).unwrap(); |
| 118 | //! encoder.push(Header::Positive(7)).unwrap(); |
| 119 | //! encoder.text("Hello, World!" , 7).unwrap(); |
| 120 | //! |
| 121 | //! // Validate our output |
| 122 | //! encoder.flush().unwrap(); |
| 123 | //! assert_eq!(b" \xa1\x07\x7f\x67Hello, \x66World! \xff" , &buffer[..]); |
| 124 | //! ``` |
| 125 | |
| 126 | #![cfg_attr (not(feature = "std" ), no_std)] |
| 127 | #![deny (missing_docs)] |
| 128 | #![deny (clippy::all)] |
| 129 | #![deny (clippy::cargo)] |
| 130 | |
| 131 | #[cfg (feature = "alloc" )] |
| 132 | extern crate alloc; |
| 133 | |
| 134 | mod dec; |
| 135 | mod enc; |
| 136 | mod hdr; |
| 137 | mod seg; |
| 138 | |
| 139 | pub use dec::*; |
| 140 | pub use enc::*; |
| 141 | pub use hdr::*; |
| 142 | pub use seg::{Segment, Segments}; |
| 143 | |
| 144 | /// Simple value constants |
| 145 | pub mod simple { |
| 146 | #![allow (missing_docs)] |
| 147 | |
| 148 | pub const FALSE: u8 = 20; |
| 149 | pub const TRUE: u8 = 21; |
| 150 | pub const NULL: u8 = 22; |
| 151 | pub const UNDEFINED: u8 = 23; |
| 152 | } |
| 153 | |
| 154 | /// Tag constants |
| 155 | pub mod tag { |
| 156 | #![allow (missing_docs)] |
| 157 | |
| 158 | pub const BIGPOS: u64 = 2; |
| 159 | pub const BIGNEG: u64 = 3; |
| 160 | } |
| 161 | |
| 162 | #[derive(Debug)] |
| 163 | struct InvalidError(()); |
| 164 | |
| 165 | #[derive(Copy, Clone, Debug, PartialEq, Eq)] |
| 166 | enum Major { |
| 167 | Positive, |
| 168 | Negative, |
| 169 | Bytes, |
| 170 | Text, |
| 171 | Array, |
| 172 | Map, |
| 173 | Tag, |
| 174 | Other, |
| 175 | } |
| 176 | |
| 177 | #[derive(Copy, Clone, Debug, PartialEq, Eq)] |
| 178 | enum Minor { |
| 179 | This(u8), |
| 180 | Next1([u8; 1]), |
| 181 | Next2([u8; 2]), |
| 182 | Next4([u8; 4]), |
| 183 | Next8([u8; 8]), |
| 184 | More, |
| 185 | } |
| 186 | |
| 187 | impl AsRef<[u8]> for Minor { |
| 188 | #[inline ] |
| 189 | fn as_ref(&self) -> &[u8] { |
| 190 | match self { |
| 191 | Self::More => &[], |
| 192 | Self::This(..) => &[], |
| 193 | Self::Next1(x) => x.as_ref(), |
| 194 | Self::Next2(x) => x.as_ref(), |
| 195 | Self::Next4(x) => x.as_ref(), |
| 196 | Self::Next8(x) => x.as_ref(), |
| 197 | } |
| 198 | } |
| 199 | } |
| 200 | |
| 201 | impl AsMut<[u8]> for Minor { |
| 202 | #[inline ] |
| 203 | fn as_mut(&mut self) -> &mut [u8] { |
| 204 | match self { |
| 205 | Self::More => &mut [], |
| 206 | Self::This(..) => &mut [], |
| 207 | Self::Next1(x) => x.as_mut(), |
| 208 | Self::Next2(x) => x.as_mut(), |
| 209 | Self::Next4(x) => x.as_mut(), |
| 210 | Self::Next8(x) => x.as_mut(), |
| 211 | } |
| 212 | } |
| 213 | } |
| 214 | |
| 215 | #[derive(Copy, Clone, Debug, PartialEq, Eq)] |
| 216 | struct Title(pub Major, pub Minor); |
| 217 | |
| 218 | #[cfg (test)] |
| 219 | mod tests { |
| 220 | use super::*; |
| 221 | |
| 222 | macro_rules! neg { |
| 223 | ($i:expr) => { |
| 224 | Header::Negative((($i as i128) ^ !0) as u64) |
| 225 | }; |
| 226 | } |
| 227 | |
| 228 | #[allow (clippy::excessive_precision)] |
| 229 | #[test] |
| 230 | fn leaf() { |
| 231 | use core::f64::{INFINITY, NAN}; |
| 232 | |
| 233 | let data = &[ |
| 234 | (Header::Positive(0), "00" , true), |
| 235 | (Header::Positive(1), "01" , true), |
| 236 | (Header::Positive(10), "0a" , true), |
| 237 | (Header::Positive(23), "17" , true), |
| 238 | (Header::Positive(24), "1818" , true), |
| 239 | (Header::Positive(25), "1819" , true), |
| 240 | (Header::Positive(100), "1864" , true), |
| 241 | (Header::Positive(1000), "1903e8" , true), |
| 242 | (Header::Positive(1000000), "1a000f4240" , true), |
| 243 | (Header::Positive(1000000000000), "1b000000e8d4a51000" , true), |
| 244 | ( |
| 245 | Header::Positive(18446744073709551615), |
| 246 | "1bffffffffffffffff" , |
| 247 | true, |
| 248 | ), |
| 249 | (neg!(-18446744073709551616), "3bffffffffffffffff" , true), |
| 250 | (neg!(-1), "20" , true), |
| 251 | (neg!(-10), "29" , true), |
| 252 | (neg!(-100), "3863" , true), |
| 253 | (neg!(-1000), "3903e7" , true), |
| 254 | (Header::Float(0.0), "f90000" , true), |
| 255 | (Header::Float(-0.0), "f98000" , true), |
| 256 | (Header::Float(1.0), "f93c00" , true), |
| 257 | (Header::Float(1.1), "fb3ff199999999999a" , true), |
| 258 | (Header::Float(1.5), "f93e00" , true), |
| 259 | (Header::Float(65504.0), "f97bff" , true), |
| 260 | (Header::Float(100000.0), "fa47c35000" , true), |
| 261 | (Header::Float(3.4028234663852886e+38), "fa7f7fffff" , true), |
| 262 | (Header::Float(1.0e+300), "fb7e37e43c8800759c" , true), |
| 263 | (Header::Float(5.960464477539063e-8), "f90001" , true), |
| 264 | (Header::Float(0.00006103515625), "f90400" , true), |
| 265 | (Header::Float(-4.0), "f9c400" , true), |
| 266 | (Header::Float(-4.1), "fbc010666666666666" , true), |
| 267 | (Header::Float(INFINITY), "f97c00" , true), |
| 268 | (Header::Float(NAN), "f97e00" , true), |
| 269 | (Header::Float(-INFINITY), "f9fc00" , true), |
| 270 | (Header::Float(INFINITY), "fa7f800000" , false), |
| 271 | (Header::Float(NAN), "fa7fc00000" , false), |
| 272 | (Header::Float(-INFINITY), "faff800000" , false), |
| 273 | (Header::Float(INFINITY), "fb7ff0000000000000" , false), |
| 274 | (Header::Float(NAN), "fb7ff8000000000000" , false), |
| 275 | (Header::Float(-INFINITY), "fbfff0000000000000" , false), |
| 276 | (Header::Simple(simple::FALSE), "f4" , true), |
| 277 | (Header::Simple(simple::TRUE), "f5" , true), |
| 278 | (Header::Simple(simple::NULL), "f6" , true), |
| 279 | (Header::Simple(simple::UNDEFINED), "f7" , true), |
| 280 | (Header::Simple(16), "f0" , true), |
| 281 | (Header::Simple(24), "f818" , true), |
| 282 | (Header::Simple(255), "f8ff" , true), |
| 283 | (Header::Tag(0), "c0" , true), |
| 284 | (Header::Tag(1), "c1" , true), |
| 285 | (Header::Tag(23), "d7" , true), |
| 286 | (Header::Tag(24), "d818" , true), |
| 287 | (Header::Tag(32), "d820" , true), |
| 288 | (Header::Bytes(Some(0)), "40" , true), |
| 289 | (Header::Bytes(Some(4)), "44" , true), |
| 290 | (Header::Text(Some(0)), "60" , true), |
| 291 | (Header::Text(Some(4)), "64" , true), |
| 292 | ]; |
| 293 | |
| 294 | for (header, bytes, encode) in data.iter().cloned() { |
| 295 | let bytes = hex::decode(bytes).unwrap(); |
| 296 | |
| 297 | let mut decoder = Decoder::from(&bytes[..]); |
| 298 | match (header, decoder.pull().unwrap()) { |
| 299 | // NaN equality... |
| 300 | (Header::Float(l), Header::Float(r)) if l.is_nan() && r.is_nan() => (), |
| 301 | |
| 302 | // Everything else... |
| 303 | (l, r) => assert_eq!(l, r), |
| 304 | } |
| 305 | |
| 306 | if encode { |
| 307 | let mut buffer = [0u8; 1024]; |
| 308 | let mut writer = &mut buffer[..]; |
| 309 | let mut encoder = Encoder::from(&mut writer); |
| 310 | encoder.push(header).unwrap(); |
| 311 | |
| 312 | let len = writer.len(); |
| 313 | assert_eq!(&bytes[..], &buffer[..1024 - len]); |
| 314 | } |
| 315 | } |
| 316 | } |
| 317 | |
| 318 | #[test] |
| 319 | fn node() { |
| 320 | let data: &[(&str, &[Header])] = &[ |
| 321 | ("80" , &[Header::Array(Some(0))]), |
| 322 | ( |
| 323 | "83010203" , |
| 324 | &[ |
| 325 | Header::Array(Some(3)), |
| 326 | Header::Positive(1), |
| 327 | Header::Positive(2), |
| 328 | Header::Positive(3), |
| 329 | ], |
| 330 | ), |
| 331 | ( |
| 332 | "98190102030405060708090a0b0c0d0e0f101112131415161718181819" , |
| 333 | &[ |
| 334 | Header::Array(Some(25)), |
| 335 | Header::Positive(1), |
| 336 | Header::Positive(2), |
| 337 | Header::Positive(3), |
| 338 | Header::Positive(4), |
| 339 | Header::Positive(5), |
| 340 | Header::Positive(6), |
| 341 | Header::Positive(7), |
| 342 | Header::Positive(8), |
| 343 | Header::Positive(9), |
| 344 | Header::Positive(10), |
| 345 | Header::Positive(11), |
| 346 | Header::Positive(12), |
| 347 | Header::Positive(13), |
| 348 | Header::Positive(14), |
| 349 | Header::Positive(15), |
| 350 | Header::Positive(16), |
| 351 | Header::Positive(17), |
| 352 | Header::Positive(18), |
| 353 | Header::Positive(19), |
| 354 | Header::Positive(20), |
| 355 | Header::Positive(21), |
| 356 | Header::Positive(22), |
| 357 | Header::Positive(23), |
| 358 | Header::Positive(24), |
| 359 | Header::Positive(25), |
| 360 | ], |
| 361 | ), |
| 362 | ("a0" , &[Header::Map(Some(0))]), |
| 363 | ( |
| 364 | "a201020304" , |
| 365 | &[ |
| 366 | Header::Map(Some(2)), |
| 367 | Header::Positive(1), |
| 368 | Header::Positive(2), |
| 369 | Header::Positive(3), |
| 370 | Header::Positive(4), |
| 371 | ], |
| 372 | ), |
| 373 | ("9fff" , &[Header::Array(None), Header::Break]), |
| 374 | ( |
| 375 | "9f018202039f0405ffff" , |
| 376 | &[ |
| 377 | Header::Array(None), |
| 378 | Header::Positive(1), |
| 379 | Header::Array(Some(2)), |
| 380 | Header::Positive(2), |
| 381 | Header::Positive(3), |
| 382 | Header::Array(None), |
| 383 | Header::Positive(4), |
| 384 | Header::Positive(5), |
| 385 | Header::Break, |
| 386 | Header::Break, |
| 387 | ], |
| 388 | ), |
| 389 | ( |
| 390 | "9f01820203820405ff" , |
| 391 | &[ |
| 392 | Header::Array(None), |
| 393 | Header::Positive(1), |
| 394 | Header::Array(Some(2)), |
| 395 | Header::Positive(2), |
| 396 | Header::Positive(3), |
| 397 | Header::Array(Some(2)), |
| 398 | Header::Positive(4), |
| 399 | Header::Positive(5), |
| 400 | Header::Break, |
| 401 | ], |
| 402 | ), |
| 403 | ( |
| 404 | "83018202039f0405ff" , |
| 405 | &[ |
| 406 | Header::Array(Some(3)), |
| 407 | Header::Positive(1), |
| 408 | Header::Array(Some(2)), |
| 409 | Header::Positive(2), |
| 410 | Header::Positive(3), |
| 411 | Header::Array(None), |
| 412 | Header::Positive(4), |
| 413 | Header::Positive(5), |
| 414 | Header::Break, |
| 415 | ], |
| 416 | ), |
| 417 | ( |
| 418 | "83019f0203ff820405" , |
| 419 | &[ |
| 420 | Header::Array(Some(3)), |
| 421 | Header::Positive(1), |
| 422 | Header::Array(None), |
| 423 | Header::Positive(2), |
| 424 | Header::Positive(3), |
| 425 | Header::Break, |
| 426 | Header::Array(Some(2)), |
| 427 | Header::Positive(4), |
| 428 | Header::Positive(5), |
| 429 | ], |
| 430 | ), |
| 431 | ( |
| 432 | "9f0102030405060708090a0b0c0d0e0f101112131415161718181819ff" , |
| 433 | &[ |
| 434 | Header::Array(None), |
| 435 | Header::Positive(1), |
| 436 | Header::Positive(2), |
| 437 | Header::Positive(3), |
| 438 | Header::Positive(4), |
| 439 | Header::Positive(5), |
| 440 | Header::Positive(6), |
| 441 | Header::Positive(7), |
| 442 | Header::Positive(8), |
| 443 | Header::Positive(9), |
| 444 | Header::Positive(10), |
| 445 | Header::Positive(11), |
| 446 | Header::Positive(12), |
| 447 | Header::Positive(13), |
| 448 | Header::Positive(14), |
| 449 | Header::Positive(15), |
| 450 | Header::Positive(16), |
| 451 | Header::Positive(17), |
| 452 | Header::Positive(18), |
| 453 | Header::Positive(19), |
| 454 | Header::Positive(20), |
| 455 | Header::Positive(21), |
| 456 | Header::Positive(22), |
| 457 | Header::Positive(23), |
| 458 | Header::Positive(24), |
| 459 | Header::Positive(25), |
| 460 | Header::Break, |
| 461 | ], |
| 462 | ), |
| 463 | ]; |
| 464 | |
| 465 | for (bytes, headers) in data { |
| 466 | let bytes = hex::decode(bytes).unwrap(); |
| 467 | |
| 468 | // Test decoding |
| 469 | let mut decoder = Decoder::from(&bytes[..]); |
| 470 | for header in headers.iter().cloned() { |
| 471 | assert_eq!(header, decoder.pull().unwrap()); |
| 472 | } |
| 473 | |
| 474 | // Test encoding |
| 475 | let mut buffer = [0u8; 1024]; |
| 476 | let mut writer = &mut buffer[..]; |
| 477 | let mut encoder = Encoder::from(&mut writer); |
| 478 | |
| 479 | for header in headers.iter().cloned() { |
| 480 | encoder.push(header).unwrap(); |
| 481 | } |
| 482 | |
| 483 | let len = writer.len(); |
| 484 | assert_eq!(&bytes[..], &buffer[..1024 - len]); |
| 485 | } |
| 486 | } |
| 487 | } |
| 488 | |