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

//! CBOR and serialization.
//!
//! # What is CBOR?
//! [CBOR](http://cbor.io) is a way to encode data in a space-efficient and fast binary format.
//! CBORs data-model is a superset of the JSONs.
//!
//! A simple object describing a person in diagnostic notation (it is actually JSON plus some
//! annotations) looks like
//!
//! ```json
//! {
//!     "FirstName": "John",
//!     "LastName": "Doe",
//!     "Age": 43,
//!     "Address": {
//!         "Street": "Downing Street 10",
//!         "City": "London",
//!         "Country": "Great Britain"
//!     },
//!     "PhoneNumbers": [
//!         "+44 1234567",
//!         "+44 2345678"
//!     ]
//! }
//! ```
//!
//! The CBOR encoded object with comments in hexadecimal notation looks like
//!
//! ```cbor
//! a5                                      # map(5)
//!    69                                   # text(9)
//!       46697273744e616d65                # "FirstName"
//!    64                                   # text(4)
//!       4a6f686e                          # "John"
//!    68                                   # text(8)
//!       4c6173744e616d65                  # "LastName"
//!    63                                   # text(3)
//!       446f65                            # "Doe"
//!    63                                   # text(3)
//!       416765                            # "Age"
//!    18 2b                                # unsigned(43)
//!    67                                   # text(7)
//!       41646472657373                    # "Address"
//!    a3                                   # map(3)
//!       66                                # text(6)
//!          537472656574                   # "Street"
//!       71                                # text(17)
//!          446f776e696e6720537472656574203130 # "Downing Street 10"
//!       64                                # text(4)
//!          43697479                       # "City"
//!       66                                # text(6)
//!          4c6f6e646f6e                   # "London"
//!       67                                # text(7)
//!          436f756e747279                 # "Country"
//!       6d                                # text(13)
//!          4772656174204272697461696e     # "Great Britain"
//!    6c                                   # text(12)
//!       50686f6e654e756d62657273          # "PhoneNumbers"
//!    82                                   # array(2)
//!       6b                                # text(11)
//!          2b34342031323334353637         # "+44 1234567"
//!       6b                                # text(11)
//!          2b34342032333435363738         # "+44 2345678"
//! ```
//! While the JSON encoding is 174 bytes long the CBOR representation is only 141 bytes long.
//! This is 19% shorter! Sometimes compression will even better, but never CBOR will be longer
//! than the corresponding JSON. More importantly CBOR supports binary data, custom data tyes,
//! annotations for dates, times and expected encoding and is faster to serialize and deserialize.
//! It can even be used on embedded devices.
//!
//! # Type-based Serialization and Deserialization
//! Serde provides a mechanism for low boilerplate serialization & deserialization of values to and
//! from CBOR via the serialization API. To be able to serialize a piece of data, it must implement
//! the `serde::Serialize` trait. To be able to deserialize a piece of data, it must implement the
//! `serde::Deserialize` trait. Serde provides an annotation to automatically generate the
//! code for these traits: `#[derive(Serialize, Deserialize)]`.
//!
//! The CBOR API also provides an enum `serde_cbor::Value`.
//!
//! # Packed Encoding
//! When serializing structs or enums in CBOR the keys or enum variant names will be serialized
//! as string keys to a map. Esspecially in embedded environments this can increase the file
//! size too much. In packed encoding the keys and variants will be serialized as variable sized
//! integers. The first 24 entries in any struct consume only a single byte!
//! To serialize a document in packed encoding use `ser::to_(vec|writer)_packed`, deserialization
//! works without any changes.
//!
//! # Self describing documents
//! In some contexts different formats are used but there is no way to declare the format used
//! out of band. For this reason CBOR has a magic number that may be added before any document.
//! The *`_sd` (for *s*elf*d*escribe) append the magic number before documents.
//!
//! # Examples
//! Read a CBOR value that is known to be a map of string keys to string values and print it.
//!
//! ```rust
//! use std::collections::BTreeMap;
//! use serde_cbor::from_slice;
//!
//! let slice = b"\xa5aaaAabaBacaCadaDaeaE";
//! let value: BTreeMap<String, String> = from_slice(slice).unwrap();
//! println!("{:?}", value); // {"e": "E", "d": "D", "a": "A", "c": "C", "b": "B"}
//! ```
//!
//! Read a general CBOR value with an unknown content.
//!
//! ```rust
//! use serde_cbor::{from_slice, Value};
//!
//! let slice = b"\x82\x01\xa1aaab";
//! let value: Value = from_slice(slice).unwrap();
//! println!("{:?}", value); // Array([U64(1), Object({String("a"): String("b")})])
//! ```
//!
//! Serialize an object.
//!
//! ```rust
//! use std::collections::BTreeMap;
//! use serde_cbor::to_vec;
//!
//! let mut programming_languages = BTreeMap::new();
//! programming_languages.insert("rust", vec!["safe", "concurrent", "fast"]);
//! programming_languages.insert("python", vec!["powerful", "friendly", "open"]);
//! programming_languages.insert("js", vec!["lightweight", "interpreted", "object-oriented"]);
//! let encoded = to_vec(&programming_languages);
//! assert_eq!(encoded.unwrap().len(), 103);
//! ```

#![deny(missing_docs)]

extern crate byteorder;

#[macro_use]
extern crate serde;

mod read;
pub mod de;
pub mod error;
pub mod ser;
pub mod value;

#[doc(inline)]
pub use de::{from_slice, from_reader, Deserializer};
#[doc(inline)]
pub use ser::{to_writer, to_vec, Serializer};
#[doc(inline)]
pub use value::{Value, ObjectKey, to_value};