Compare commits

..

6 commits

15 changed files with 716 additions and 1228 deletions

1
.gitignore vendored
View file

@ -1,2 +1 @@
/target /target
/Cargo.lock

460
Cargo.lock generated Normal file
View file

@ -0,0 +1,460 @@
# This file is automatically @generated by Cargo.
# It is not intended for manual editing.
version = 3
[[package]]
name = "base64"
version = "0.13.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9e1b586273c5702936fe7b7d6896644d8be71e6314cfe09d3167c95f712589e8"
[[package]]
name = "bitflags"
version = "2.4.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "327762f6e5a765692301e5bb513e0d9fef63be86bbc14528052b1cd3e6f03e07"
[[package]]
name = "block-buffer"
version = "0.10.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3078c7629b62d3f0439517fa394996acacc5cbc91c5a20d8c658e77abd503a71"
dependencies = [
"generic-array",
]
[[package]]
name = "cfg-if"
version = "1.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
[[package]]
name = "chacha20"
version = "0.9.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c3613f74bd2eac03dad61bd53dbe620703d4371614fe0bc3b9f04dd36fe4e818"
dependencies = [
"cfg-if",
"cipher",
"cpufeatures",
]
[[package]]
name = "cipher"
version = "0.4.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "773f3b9af64447d2ce9850330c473515014aa235e6a783b02db81ff39e4a3dad"
dependencies = [
"crypto-common",
"inout",
"zeroize",
]
[[package]]
name = "cpufeatures"
version = "0.2.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ce420fe07aecd3e67c5f910618fe65e94158f6dcc0adf44e00d69ce2bdfe0fd0"
dependencies = [
"libc",
]
[[package]]
name = "crypto-common"
version = "0.1.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1bfb12502f3fc46cca1bb51ac28df9d618d813cdc3d2f25b9fe775a34af26bb3"
dependencies = [
"generic-array",
"typenum",
]
[[package]]
name = "curve25519-dalek"
version = "4.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e89b8c6a2e4b1f45971ad09761aafb85514a84744b67a95e32c3cc1352d1f65c"
dependencies = [
"cfg-if",
"cpufeatures",
"curve25519-dalek-derive",
"fiat-crypto",
"platforms",
"rustc_version",
"subtle",
"zeroize",
]
[[package]]
name = "curve25519-dalek-derive"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f46882e17999c6cc590af592290432be3bce0428cb0d5f8b6715e4dc7b383eb3"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.39",
]
[[package]]
name = "digest"
version = "0.10.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9ed9a281f7bc9b7576e61468ba615a66a5c8cfdff42420a70aa82701a3b1e292"
dependencies = [
"block-buffer",
"crypto-common",
]
[[package]]
name = "dryoc"
version = "0.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "65f5013c48133363c5a8db6bc74511b8b254680929c7757d9d833dea18c12f13"
dependencies = [
"bitflags",
"chacha20",
"curve25519-dalek",
"generic-array",
"lazy_static",
"libc",
"rand_core",
"salsa20",
"sha2",
"subtle",
"winapi",
"zeroize",
]
[[package]]
name = "err-derive"
version = "0.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c34a887c8df3ed90498c1c437ce21f211c8e27672921a8ffa293cb8d6d4caa9e"
dependencies = [
"proc-macro-error",
"proc-macro2",
"quote",
"rustversion",
"syn 1.0.109",
"synstructure",
]
[[package]]
name = "fiat-crypto"
version = "0.2.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "53a56f0780318174bad1c127063fd0c5fdfb35398e3cd79ffaab931a6c79df80"
[[package]]
name = "generic-array"
version = "0.14.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "85649ca51fd72272d7821adaf274ad91c288277713d9c18820d8499a7ff69e9a"
dependencies = [
"typenum",
"version_check",
]
[[package]]
name = "getrandom"
version = "0.2.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fe9006bed769170c11f845cf00c7c1e9092aeb3f268e007c3e760ac68008070f"
dependencies = [
"cfg-if",
"libc",
"wasi",
]
[[package]]
name = "hex"
version = "0.4.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7f24254aa9a54b5c858eaee2f5bccdb46aaf0e486a595ed5fd8f86ba55232a70"
[[package]]
name = "inout"
version = "0.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a0c10553d664a4d0bcff9f4215d0aac67a639cc68ef660840afe309b807bc9f5"
dependencies = [
"generic-array",
]
[[package]]
name = "lazy_static"
version = "1.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e2abad23fbc42b3700f2f279844dc832adb2b2eb069b2df918f455c4e18cc646"
[[package]]
name = "libc"
version = "0.2.150"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "89d92a4743f9a61002fae18374ed11e7973f530cb3a3255fb354818118b2203c"
[[package]]
name = "memchr"
version = "2.6.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f665ee40bc4a3c5590afb1e9677db74a508659dfd71e126420da8274909a0167"
[[package]]
name = "minimal-lexical"
version = "0.2.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "68354c5c6bd36d73ff3feceb05efa59b6acb7626617f4962be322a825e61f79a"
[[package]]
name = "nettext"
version = "0.4.1"
dependencies = [
"base64",
"dryoc",
"err-derive",
"hex",
"nom",
"serde",
]
[[package]]
name = "nom"
version = "7.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d273983c5a657a70a3e8f2a01329822f3b8c8172b73826411a55751e404a0a4a"
dependencies = [
"memchr",
"minimal-lexical",
]
[[package]]
name = "platforms"
version = "3.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "14e6ab3f592e6fb464fc9712d8d6e6912de6473954635fd76a589d832cffcbb0"
[[package]]
name = "proc-macro-error"
version = "1.0.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "da25490ff9892aab3fcf7c36f08cfb902dd3e71ca0f9f9517bea02a73a5ce38c"
dependencies = [
"proc-macro-error-attr",
"proc-macro2",
"quote",
"syn 1.0.109",
"version_check",
]
[[package]]
name = "proc-macro-error-attr"
version = "1.0.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a1be40180e52ecc98ad80b184934baf3d0d29f979574e439af5a55274b35f869"
dependencies = [
"proc-macro2",
"quote",
"version_check",
]
[[package]]
name = "proc-macro2"
version = "1.0.69"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "134c189feb4956b20f6f547d2cf727d4c0fe06722b20a0eec87ed445a97f92da"
dependencies = [
"unicode-ident",
]
[[package]]
name = "quote"
version = "1.0.33"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5267fca4496028628a95160fc423a33e8b2e6af8a5302579e322e4b520293cae"
dependencies = [
"proc-macro2",
]
[[package]]
name = "rand_core"
version = "0.6.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ec0be4795e2f6a28069bec0b5ff3e2ac9bafc99e6a9a7dc3547996c5c816922c"
dependencies = [
"getrandom",
]
[[package]]
name = "rustc_version"
version = "0.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bfa0f585226d2e68097d4f95d113b15b83a82e819ab25717ec0590d9584ef366"
dependencies = [
"semver",
]
[[package]]
name = "rustversion"
version = "1.0.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7ffc183a10b4478d04cbbbfc96d0873219d962dd5accaff2ffbd4ceb7df837f4"
[[package]]
name = "salsa20"
version = "0.10.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "97a22f5af31f73a954c10289c93e8a50cc23d971e80ee446f1f6f7137a088213"
dependencies = [
"cipher",
]
[[package]]
name = "semver"
version = "1.0.20"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "836fa6a3e1e547f9a2c4040802ec865b5d85f4014efe00555d7090a3dcaa1090"
[[package]]
name = "serde"
version = "1.0.192"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bca2a08484b285dcb282d0f67b26cadc0df8b19f8c12502c13d966bf9482f001"
dependencies = [
"serde_derive",
]
[[package]]
name = "serde_derive"
version = "1.0.192"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d6c7207fbec9faa48073f3e3074cbe553af6ea512d7c21ba46e434e70ea9fbc1"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.39",
]
[[package]]
name = "sha2"
version = "0.10.8"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "793db75ad2bcafc3ffa7c68b215fee268f537982cd901d132f89c6343f3a3dc8"
dependencies = [
"cfg-if",
"cpufeatures",
"digest",
]
[[package]]
name = "subtle"
version = "2.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "81cdd64d312baedb58e21336b31bc043b77e01cc99033ce76ef539f78e965ebc"
[[package]]
name = "syn"
version = "1.0.109"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "72b64191b275b66ffe2469e8af2c1cfe3bafa67b529ead792a6d0160888b4237"
dependencies = [
"proc-macro2",
"quote",
"unicode-ident",
]
[[package]]
name = "syn"
version = "2.0.39"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "23e78b90f2fcf45d3e842032ce32e3f2d1545ba6636271dcbf24fa306d87be7a"
dependencies = [
"proc-macro2",
"quote",
"unicode-ident",
]
[[package]]
name = "synstructure"
version = "0.12.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f36bdaa60a83aca3921b5259d5400cbf5e90fc51931376a9bd4a0eb79aa7210f"
dependencies = [
"proc-macro2",
"quote",
"syn 1.0.109",
"unicode-xid",
]
[[package]]
name = "typenum"
version = "1.17.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "42ff0bf0c66b8238c6f3b578df37d0b7848e55df8577b3f74f92a69acceeb825"
[[package]]
name = "unicode-ident"
version = "1.0.12"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3354b9ac3fae1ff6755cb6db53683adb661634f67557942dea4facebec0fee4b"
[[package]]
name = "unicode-xid"
version = "0.2.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f962df74c8c05a667b5ee8bcf162993134c104e96440b663c8daa176dc772d8c"
[[package]]
name = "version_check"
version = "0.9.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "49874b5167b65d7193b8aba1567f5c7d93d001cafc34600cee003eda787e483f"
[[package]]
name = "wasi"
version = "0.11.0+wasi-snapshot-preview1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9c8d87e72b64a3b4db28d11ce29237c246188f4f51057d65a7eab63b7987e423"
[[package]]
name = "winapi"
version = "0.3.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5c839a674fcd7a98952e593242ea400abe93992746761e38641405d28b00f419"
dependencies = [
"winapi-i686-pc-windows-gnu",
"winapi-x86_64-pc-windows-gnu",
]
[[package]]
name = "winapi-i686-pc-windows-gnu"
version = "0.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6"
[[package]]
name = "winapi-x86_64-pc-windows-gnu"
version = "0.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"
[[package]]
name = "zeroize"
version = "1.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2a0956f1ba7c7909bfb66c2e9e4124ab6f6482560f6628b5aaeba39207c9aad9"
dependencies = [
"zeroize_derive",
]
[[package]]
name = "zeroize_derive"
version = "1.4.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ce36e65b0d2999d2aafac989fb249189a141aee1f53c612c1f37d72631959f69"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.39",
]

View file

@ -2,7 +2,7 @@
name = "nettext" name = "nettext"
description = "A text-based data format for cryptographic network protocols" description = "A text-based data format for cryptographic network protocols"
authors = ["Alex Auvolat <alex@adnab.me>"] authors = ["Alex Auvolat <alex@adnab.me>"]
version = "0.4.0" version = "0.4.1"
edition = "2021" edition = "2021"
license = "AGPL-3.0" license = "AGPL-3.0"
readme = "README.md" readme = "README.md"

View file

@ -82,6 +82,30 @@ Terms can be interpreted in a number of different ways, depending on the context
Terms further have mappings as different data types: Terms further have mappings as different data types:
- BYTES: if the term maps as a STRING, decode it using base64
- INT: if the term maps as a STRING, decode it as an integer written in decimal notation - INT: if the term maps as a STRING, decode it as an integer written in decimal notation
- BYTES: if the term maps as a STRING, decode it using base64. Since a STRING cannot be empty, the string `-` is used to represent an empty byte string.
- Cryptographic data types (see below)
## Cryptographic data types
Cryptographic values such as keys, hashes, signatures, etc. are encoded
as STRING with a prefix indicating the algorithm used, followed by ":",
followed by the base64-encoded value.
Prefixes are as follows:
- `pk.box:` public key for NaCl's box API
- `sk.box:` secret key for NaCl's box API
- `sk.sbox:` secret key for NaCl's secretbox API
- `h.sha256:` sha256 hash
- `h.sha512:` sha512 hash
- `h.sha3:` sha3 hash
- `h.b2:` blake2b hash
- `h.b3:` blake3 hash
- `sig.ed25519:` ed25519 signature
- `pk.ed25519:` ed25519 public signing key
- `sk.ed25519:` ed25519 secret signing key
More can be added.
- HASH, PUBKEY, SECKEY, SIGNATURE, ENCKEY, DECKEY, SYMKEY: a bunch of things that interpret BYTES as specific cryptographic items - HASH, PUBKEY, SECKEY, SIGNATURE, ENCKEY, DECKEY, SYMKEY: a bunch of things that interpret BYTES as specific cryptographic items

View file

@ -1,474 +0,0 @@
use crate::{
is_string_char, is_whitespace, DICT_ASSIGN, DICT_CLOSE, DICT_DELIM, DICT_OPEN, LIST_CLOSE,
LIST_DELIM, LIST_OPEN,
};
use super::*;
pub fn decode(input: &[u8]) -> Result<(Buf<'_>, ITerm), ParseError> {
let mut buf = Buf {
bytes: input.into(),
seqs: Vec::with_capacity(16),
dicts: Vec::with_capacity(16),
terms: Vec::with_capacity(16),
};
let all_buf = IRaw {
start: 0,
end: input.len() as Pos,
};
let term = buf.decode(all_buf)?;
Ok((buf, term))
}
// ================
enum StackItem {
Term(ITerm),
KeyValue(IRaw, ITerm),
}
type Stack = Vec<StackItem>;
impl StackItem {
fn term(self) -> ITerm {
match self {
StackItem::Term(term) => term,
_ => unreachable!(),
}
}
fn kv(self) -> (IRaw, ITerm) {
match self {
StackItem::KeyValue(key, term) => (key, term),
_ => unreachable!(),
}
}
}
impl<'a> Buf<'a> {
pub(crate) fn decode(&mut self, raw: IRaw) -> Result<ITerm, ParseError> {
let mut stack = Vec::with_capacity(16);
let (term, rest) = self.decode_seq(raw, &mut stack)?;
assert!(stack.is_empty());
let rest = self.take_whitespace(rest);
if rest.start < raw.end {
return Err(ParseError::UnexpectedInput(rest.start as usize));
}
Ok(term)
}
fn decode_seq(&mut self, raw: IRaw, stack: &mut Stack) -> Result<(ITerm, IRaw), ParseError> {
let start = self.take_whitespace(raw);
let stack_start = stack.len();
let mut cur_end = start;
let mut next_start = start;
loop {
match self.decode_nonseq_term(next_start, stack) {
Err(_) => break,
Ok((term, rest)) => {
stack.push(StackItem::Term(term));
cur_end = rest;
next_start = self.take_whitespace(rest);
}
}
}
if stack.len() == stack_start {
Err(self.error_at(next_start))
} else if stack.len() == stack_start + 1 {
Ok((stack.pop().unwrap().term(), next_start))
} else {
let seq_raw = IRaw {
start: start.start,
end: cur_end.start,
};
let seq_start = self.seqs.len();
self.seqs
.extend(stack.drain(stack_start..).map(StackItem::term));
let seq = TTerm::RawSeq(
seq_raw,
ISeq {
seq_start: seq_start as Pos,
seq_end: self.seqs.len() as Pos,
},
);
Ok((self.push_term(seq), next_start))
}
}
fn decode_nonseq_term(
&mut self,
raw: IRaw,
stack: &mut Stack,
) -> Result<(ITerm, IRaw), ParseError> {
if let Ok((term, rest)) = self.decode_string(raw) {
Ok((term, rest))
} else if let Ok((term, rest)) = self.decode_list(raw, stack) {
Ok((term, rest))
} else if let Ok((term, rest)) = self.decode_dict(raw, stack) {
Ok((term, rest))
} else {
Err(self.error_at(raw))
}
}
fn decode_list(&mut self, raw: IRaw, stack: &mut Stack) -> Result<(ITerm, IRaw), ParseError> {
let stack_start = stack.len();
let mut cur = self.take_whitespace(self.take_char(raw, LIST_OPEN)?);
while let Ok((term, rest)) = self.decode_seq(cur, stack) {
stack.push(StackItem::Term(term));
cur = self.take_whitespace(rest);
if let Ok(rest) = self.take_char(rest, LIST_DELIM) {
cur = self.take_whitespace(rest);
} else {
break;
}
}
if let Ok(rest) = self.take_char(cur, LIST_CLOSE) {
let seq_raw = IRaw {
start: raw.start,
end: rest.start,
};
let seq_start = self.seqs.len();
self.seqs
.extend(stack.drain(stack_start..).map(StackItem::term));
let seq = TTerm::RawList(
seq_raw,
ISeq {
seq_start: seq_start as Pos,
seq_end: self.seqs.len() as Pos,
},
);
Ok((self.push_term(seq), rest))
} else {
stack.truncate(stack_start);
Err(self.error_at(cur))
}
}
fn decode_dict(&mut self, raw: IRaw, stack: &mut Stack) -> Result<(ITerm, IRaw), ParseError> {
let stack_start = stack.len();
match self.decode_dict_inner(raw, stack) {
Ok(rest) => {
let dict_raw = IRaw {
start: raw.start,
end: rest.start,
};
let dict_start = self.dicts.len();
self.dicts
.extend(stack.drain(stack_start..).map(StackItem::kv));
self.dicts[dict_start..]
.sort_by_key(|(k, _)| (&self.bytes[k.start as usize..k.end as usize], k.start));
for ((k1, _), (k2, _)) in self.dicts[dict_start..]
.iter()
.zip(self.dicts[dict_start + 1..].iter())
{
if self.get_bytes(*k1) == self.get_bytes(*k2) {
return Err(ParseError::DuplicateKey(String::from_utf8(self.get_bytes(*k1).to_vec()).unwrap()));
}
}
let dict = TTerm::RawDict(
dict_raw,
IDict {
dict_start: dict_start as Pos,
dict_end: self.dicts.len() as Pos,
},
);
Ok((self.push_term(dict), rest))
}
Err(e) => {
stack.truncate(stack_start);
Err(e)
}
}
}
fn decode_dict_inner(&mut self, raw: IRaw, stack: &mut Stack) -> Result<IRaw, ParseError> {
let mut cur = self.take_whitespace(self.take_char(raw, DICT_OPEN)?);
while let Ok((key, rest)) = self.take_string(cur) {
cur = self.take_whitespace(rest);
cur = self.take_char(cur, DICT_ASSIGN)?;
let (value, rest) = self.decode_seq(cur, stack)?;
cur = self.take_whitespace(rest);
stack.push(StackItem::KeyValue(key, value));
if let Ok(rest) = self.take_char(cur, DICT_DELIM) {
cur = self.take_whitespace(rest);
} else {
break;
}
}
let rest = self.take_char(cur, DICT_CLOSE)?;
Ok(rest)
}
fn decode_string(&mut self, raw: IRaw) -> Result<(ITerm, IRaw), ParseError> {
let (string_raw, rest) = self.take_string(raw)?;
let term = if (string_raw.end - string_raw.start) as usize <= STR_INLINE_MAX {
let b = self.get_bytes(string_raw);
let mut bytes = [0u8; STR_INLINE_MAX];
bytes[..b.len()].copy_from_slice(b);
TTerm::StrInline(b.len() as u8, bytes)
} else {
TTerm::Str(string_raw)
};
Ok((self.push_term(term), rest))
}
fn take_string(&mut self, raw: IRaw) -> Result<(IRaw, IRaw), ParseError> {
let mut rest = raw;
while rest.start < rest.end {
if is_string_char(self.bytes[rest.start as usize]) {
rest.start += 1;
} else {
break;
}
}
if rest.start > raw.start {
let string_raw = IRaw {
start: raw.start,
end: rest.start,
};
Ok((string_raw, rest))
} else {
Err(self.error_at(rest))
}
}
#[inline]
fn take_char(&self, raw: IRaw, c: u8) -> Result<IRaw, ParseError> {
if raw.start >= raw.end {
Err(ParseError::IncompleteInput)
} else if self.bytes[raw.start as usize] != c {
Err(ParseError::UnexpectedInput(raw.start as usize))
} else {
Ok(IRaw {
start: raw.start + 1,
end: raw.end,
})
}
}
#[inline]
fn take_whitespace(&self, mut raw: IRaw) -> IRaw {
while raw.start < raw.end {
if is_whitespace(self.bytes[raw.start as usize]) {
raw.start += 1;
} else {
break;
}
}
raw
}
#[inline]
pub(crate) fn get_bytes(&self, raw: IRaw) -> &[u8] {
&self.bytes[raw.start as usize..raw.end as usize]
}
#[inline]
fn error_at(&self, raw: IRaw) -> ParseError {
if raw.start < raw.end {
ParseError::UnexpectedInput(raw.start as usize)
} else {
ParseError::IncompleteInput
}
}
#[cfg(test)]
fn debug(&self, i: ITerm) {
use crate::debug as debug_str;
let term = &self.terms[i.0 as usize];
match term {
TTerm::StrInline(len, bytes) => {
eprintln!(
"{} -> {:?} = `{}`",
i.0,
term,
debug_str(&bytes[..*len as usize])
);
}
TTerm::Str(r) => {
eprintln!(
"{} -> {:?} = `{}`",
i.0,
term,
debug_str(&self.bytes[r.start as usize..r.end as usize])
);
}
TTerm::RawSeq(r, s) => {
eprintln!(
"{} -> {:?} = `{}` ((",
i.0,
term,
debug_str(&self.bytes[r.start as usize..r.end as usize])
);
for j in self.seqs[s.seq_start as usize..s.seq_end as usize].iter() {
self.debug(*j);
}
eprintln!("))");
}
TTerm::Seq(s) => {
eprintln!("{} -> {:?} ((", i.0, term);
for j in self.seqs[s.seq_start as usize..s.seq_end as usize].iter() {
self.debug(*j);
}
eprintln!("))");
}
TTerm::RawList(r, l) => {
eprintln!(
"{} -> {:?} = `{}` [[",
i.0,
term,
debug_str(&self.bytes[r.start as usize..r.end as usize])
);
for j in self.seqs[l.seq_start as usize..l.seq_end as usize].iter() {
self.debug(*j);
}
eprintln!("]]");
}
TTerm::List(l) => {
eprintln!("{} -> {:?} [[", i.0, term);
for j in self.seqs[l.seq_start as usize..l.seq_end as usize].iter() {
self.debug(*j);
}
eprintln!("]]");
}
TTerm::RawDict(r, d) => {
eprintln!(
"{} -> {:?} = `{}` {{{{",
i.0,
term,
debug_str(&self.bytes[r.start as usize..r.end as usize])
);
for (k, v) in self.dicts[d.dict_start as usize..d.dict_end as usize].iter() {
eprint!(
"[`{}`] = ",
debug_str(&self.bytes[k.start as usize..k.end as usize])
);
self.debug(*v);
}
eprintln!("}}}}");
}
TTerm::Dict(d) => {
eprintln!("{} -> {:?} {{{{", i.0, term);
for (k, v) in self.dicts[d.dict_start as usize..d.dict_end as usize].iter() {
eprint!(
"[`{}`] = ",
debug_str(&self.bytes[k.start as usize..k.end as usize])
);
self.debug(*v);
}
eprintln!("}}}}");
}
}
}
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn list_and_seq() {
let string = "[ [ h1; h2; h3 ]; hello world [ acc bii; persephonisumiolaodododo dlpa] ]";
eprintln!("{}", string);
let (buf, term) = decode(string.as_bytes()).unwrap();
buf.debug(term);
let [a, b] = buf.list_of(term).unwrap();
assert_eq!(buf.raw(a).unwrap(), b"[ h1; h2; h3 ]");
let l = buf.list(a).unwrap();
assert_eq!(l.len(), 3);
assert_eq!(buf.str(l[0]).unwrap(), "h1");
assert_eq!(buf.str(l[1]).unwrap(), "h2");
assert_eq!(buf.str(l[2]).unwrap(), "h3");
assert_eq!(
buf.raw(b).unwrap(),
b"hello world [ acc bii; persephonisumiolaodododo dlpa]"
);
let [h, w, l] = buf.seq_of(b).unwrap();
assert_eq!(buf.str(h).unwrap(), "hello");
assert_eq!(buf.str(w).unwrap(), "world");
assert_eq!(
buf.raw(l).unwrap(),
b"[ acc bii; persephonisumiolaodododo dlpa]"
);
let [l1, l2] = buf.list_of(l).unwrap();
assert_eq!(buf.raw(l1).unwrap(), b"acc bii");
let s = buf.seq(&l1).unwrap();
assert_eq!(s.len(), 2);
assert_eq!(buf.str(s[0]).unwrap(), "acc");
assert_eq!(buf.str(s[1]).unwrap(), "bii");
assert_eq!(buf.raw(l2).unwrap(), b"persephonisumiolaodododo dlpa");
let [s2a, s2b] = buf.seq_of(l2).unwrap();
assert_eq!(buf.str(s2a).unwrap(), "persephonisumiolaodododo");
assert_eq!(buf.str(s2b).unwrap(), "dlpa");
}
#[test]
fn dict() {
let string =
"[ { a = plop; b = hello world }; ploplop { e=15; d=12 ;c = {key=val;key2=azertyuiopazertyuiopazertyuiop}} ]";
eprintln!("{}", string);
let (buf, term) = decode(string.as_bytes()).unwrap();
buf.debug(term);
let [a, b] = buf.list_of(term).unwrap();
assert_eq!(buf.raw(a).unwrap(), b"{ a = plop; b = hello world }");
let [aa, ab] = buf.dict_of(a, ["a", "b"], false).unwrap();
assert_eq!(buf.raw(aa).unwrap(), b"plop");
assert_eq!(buf.raw(ab).unwrap(), b"hello world");
assert_eq!(
buf.raw(b).unwrap(),
b"ploplop { e=15; d=12 ;c = {key=val;key2=azertyuiopazertyuiopazertyuiop}}"
);
let [ba, bb] = buf.seq_of(b).unwrap();
assert_eq!(buf.str(ba).unwrap(), "ploplop");
assert_eq!(
buf.str(buf.dict_get(bb, "e").unwrap().unwrap()).unwrap(),
"15"
);
let mut iter = buf.dict_iter(bb).unwrap();
let (k1, v1) = iter.next().unwrap();
assert_eq!(k1, "c");
assert_eq!(
buf.raw(v1).unwrap(),
b"{key=val;key2=azertyuiopazertyuiopazertyuiop}"
);
let (k2, v2) = iter.next().unwrap();
assert_eq!(k2, "d");
assert_eq!(buf.str(v2).unwrap(), "12");
let (k3, v3) = iter.next().unwrap();
assert_eq!(k3, "e");
assert_eq!(buf.str(v3).unwrap(), "15");
assert!(iter.next().is_none());
}
}

View file

@ -1,39 +0,0 @@
use err_derive::Error;
use super::*;
#[derive(Debug, Error)]
pub enum ParseError {
#[error(display = "Incomplete input")]
IncompleteInput,
#[error(display = "UnexpectedInput at position {}", _0)]
UnexpectedInput(usize),
#[error(display = "Duplicate dictionnary key: {}", _0)]
DuplicateKey(String),
}
#[derive(Debug, Error)]
pub enum TermError {
#[error(display = "Invalid term index ({:?})", _0)]
InvalidIndex(ITerm),
#[error(display = "Wrong type, expected {}, got {}", _0, _1)]
WrongType(&'static str, &'static str),
#[error(display = "Wrong length, expected {}, got {}", _0, _1)]
WrongLength(usize, usize),
#[error(display = "Wrong dictionnary keys")]
WrongKeys,
#[error(display = "Term does not have a raw representation")]
NoRawRepresentation,
}
#[derive(Debug, Error)]
pub enum ValueError {
#[error(display = "Invalid term index ({:?})", _0)]
InvalidIndex(ITerm),
#[error(display = "Duplicate dictionary key")]
DuplicateKey,
#[error(display = "Bad string (contains invalid characters)")]
BadString,
#[error(display = "Nested sequence")]
SeqInSeq,
}

View file

@ -1,391 +0,0 @@
mod decode;
mod error;
use std::borrow::Cow;
use crate::is_string_char;
pub use error::*;
pub use decode::*;
pub(crate) const STR_INLINE_MAX: usize = 18;
pub(crate) type Pos = u32;
#[derive(Clone, Copy, Debug)]
pub struct ITerm(Pos);
#[derive(Clone, Copy, Debug)]
pub(crate) struct IRaw {
start: Pos,
end: Pos,
}
#[derive(Clone, Copy, Debug)]
pub(crate) struct ISeq {
seq_start: Pos,
seq_end: Pos,
}
#[derive(Clone, Copy, Debug)]
pub(crate) struct IDict {
dict_start: Pos,
dict_end: Pos,
}
#[derive(Debug)]
pub(crate) enum TTerm {
Str(IRaw),
StrInline(u8, [u8; STR_INLINE_MAX]),
RawSeq(IRaw, ISeq),
RawList(IRaw, ISeq),
RawDict(IRaw, IDict),
Seq(ISeq),
List(ISeq),
Dict(IDict),
}
#[derive(Debug)]
pub struct Buf<'a> {
bytes: Cow<'a, [u8]>,
seqs: Vec<ITerm>,
dicts: Vec<(IRaw, ITerm)>,
terms: Vec<TTerm>,
}
impl<'a> Buf<'a> {
pub fn new() -> Self {
Self {
bytes: Default::default(),
seqs: Vec::with_capacity(16),
dicts: Vec::with_capacity(16),
terms: Vec::with_capacity(16),
}
}
// ================ READING FUNCTIONS ==================
pub fn raw(&self, term: ITerm) -> Result<&[u8], TermError> {
match self.get_term(term)? {
TTerm::StrInline(len, bytes) => Ok(&bytes[..*len as usize]),
TTerm::Str(r) | TTerm::RawSeq(r, _) | TTerm::RawList(r, _) | TTerm::RawDict(r, _) => {
Ok(self.get_bytes(*r))
}
_ => Err(TermError::NoRawRepresentation),
}
}
pub fn str(&self, term: ITerm) -> Result<&str, TermError> {
match self.get_term(term)? {
TTerm::StrInline(len, bytes) => {
let bytes = &bytes[..*len as usize];
let s = unsafe { std::str::from_utf8_unchecked(bytes) };
Ok(s)
}
TTerm::Str(r) => {
let bytes = self.get_bytes(*r);
let s = unsafe { std::str::from_utf8_unchecked(bytes) };
Ok(s)
}
t => Err(TermError::WrongType("string", t.typename())),
}
}
pub fn seq<'x>(&'x self, term: &'x ITerm) -> Result<&'x [ITerm], TermError> {
match self.get_term(*term)? {
TTerm::RawSeq(_, s) | TTerm::Seq(s) => {
Ok(&self.seqs[s.seq_start as usize..s.seq_end as usize])
}
_ => Ok(std::slice::from_ref(term)),
}
}
pub fn seq_of<const N: usize>(&self, term: ITerm) -> Result<[ITerm; N], TermError> {
match self.get_term(term)? {
TTerm::RawSeq(_, s) | TTerm::Seq(s) => {
let seq_len = (s.seq_end - s.seq_start) as usize;
if seq_len == N {
let seq = &self.seqs[s.seq_start as usize..s.seq_end as usize];
Ok(seq.try_into().unwrap())
} else {
Err(TermError::WrongLength(N, seq_len))
}
}
t => Err(TermError::WrongType("seq", t.typename())),
}
}
pub fn list(&self, term: ITerm) -> Result<&[ITerm], TermError> {
match self.get_term(term)? {
TTerm::RawList(_, s) | TTerm::List(s) => {
Ok(&self.seqs[s.seq_start as usize..s.seq_end as usize])
}
t => Err(TermError::WrongType("list", t.typename())),
}
}
pub fn list_of<const N: usize>(&self, term: ITerm) -> Result<[ITerm; N], TermError> {
match self.get_term(term)? {
TTerm::RawList(_, s) | TTerm::List(s) => {
let list_len = (s.seq_end - s.seq_start) as usize;
if list_len == N {
let seq = &self.seqs[s.seq_start as usize..s.seq_end as usize];
Ok(seq.try_into().unwrap())
} else {
Err(TermError::WrongLength(N, list_len))
}
}
t => Err(TermError::WrongType("list", t.typename())),
}
}
pub fn dict_get(&self, term: ITerm, key: &str) -> Result<Option<ITerm>, TermError> {
match self.get_term(term)? {
TTerm::RawDict(_, d) | TTerm::Dict(d) => {
let dict = &self.dicts[d.dict_start as usize..d.dict_end as usize];
let pos_opt = dict
.binary_search_by(|(k, _)| self.get_bytes(*k).cmp(key.as_bytes()))
.ok();
Ok(pos_opt.map(|pos| dict[pos].1))
}
t => Err(TermError::WrongType("dict", t.typename())),
}
}
pub fn dict_of<const N: usize>(
&self,
term: ITerm,
keys: [&str; N],
allow_other: bool,
) -> Result<[ITerm; N], TermError> {
match self.get_term(term)? {
TTerm::RawDict(_, d) | TTerm::Dict(d) => {
let dict = &self.dicts[d.dict_start as usize..d.dict_end as usize];
if dict.len() < N || (dict.len() > N && !allow_other) {
return Err(TermError::WrongKeys);
}
let mut ret = [ITerm(0); N];
for i in 0..N {
let pos = dict
.binary_search_by(|(k, _)| self.get_bytes(*k).cmp(keys[i].as_bytes()))
.map_err(|_| TermError::WrongKeys)?;
ret[i] = dict[pos].1;
}
Ok(ret)
}
t => Err(TermError::WrongType("dict", t.typename())),
}
}
pub fn dict_iter(
&self,
term: ITerm,
) -> Result<impl Iterator<Item = (&str, ITerm)> + '_, TermError> {
match self.get_term(term)? {
TTerm::RawDict(_, d) | TTerm::Dict(d) => {
let dict = &self.dicts[d.dict_start as usize..d.dict_end as usize];
let iter = dict.iter().map(|(k, v)| {
(
unsafe { std::str::from_utf8_unchecked(self.get_bytes(*k)) },
*v,
)
});
Ok(iter)
}
t => Err(TermError::WrongType("dict", t.typename())),
}
}
// ================= WRITING FUNCTIONS ================
pub fn push_str(&mut self, s: &str) -> Result<ITerm, ValueError> {
let b = s.as_bytes();
if !b.iter().copied().all(is_string_char) {
return Err(ValueError::BadString);
}
let term = if b.len() <= STR_INLINE_MAX {
let mut bytes = [0u8; STR_INLINE_MAX];
bytes[..b.len()].copy_from_slice(b);
TTerm::StrInline(b.len() as u8, bytes)
} else {
TTerm::Str(self.push_bytes(b))
};
Ok(self.push_term(term))
}
pub fn push_seq(&mut self, iterator: impl Iterator<Item = ITerm>) -> Result<ITerm, ValueError> {
let seq_start = self.seqs.len();
for term in iterator {
match self.terms.get(term.0 as usize) {
None => {
self.seqs.truncate(seq_start);
return Err(ValueError::InvalidIndex(term));
}
Some(TTerm::RawSeq(_, _)) => {
self.seqs.truncate(seq_start);
return Err(ValueError::SeqInSeq);
}
_ => {
self.seqs.push(term);
}
}
}
let seq = ISeq {
seq_start: seq_start as Pos,
seq_end: self.seqs.len() as Pos,
};
Ok(self.push_term(TTerm::Seq(seq)))
}
pub fn push_list(
&mut self,
iterator: impl Iterator<Item = ITerm>,
) -> Result<ITerm, ValueError> {
let list_start = self.seqs.len();
for term in iterator {
match self.terms.get(term.0 as usize) {
None => {
self.seqs.truncate(list_start);
return Err(ValueError::InvalidIndex(term));
}
_ => {
self.seqs.push(term);
}
}
}
let list = ISeq {
seq_start: list_start as Pos,
seq_end: self.seqs.len() as Pos,
};
Ok(self.push_term(TTerm::List(list)))
}
pub fn push_dict<'k>(
&mut self,
iterator: impl Iterator<Item = (&'k str, ITerm)>,
) -> Result<ITerm, ValueError> {
let bytes_start = self.bytes.len();
let dict_start = self.dicts.len();
for (key, term) in iterator {
if !key.as_bytes().iter().copied().all(is_string_char) {
return Err(ValueError::BadString);
}
let key = self.push_bytes(key.as_bytes());
match self.terms.get(term.0 as usize) {
None => {
self.bytes.to_mut().truncate(bytes_start);
self.dicts.truncate(dict_start);
return Err(ValueError::InvalidIndex(term));
}
_ => {
self.dicts.push((key, term));
}
}
}
self.dicts[dict_start..]
.sort_by_key(|(k, _)| (&self.bytes[k.start as usize..k.end as usize], k.start));
for ((k1, _), (k2, _)) in self.dicts[dict_start..]
.iter()
.zip(self.dicts[dict_start + 1..].iter())
{
if self.get_bytes(*k1) == self.get_bytes(*k2) {
self.bytes.to_mut().truncate(bytes_start);
self.dicts.truncate(dict_start);
return Err(ValueError::DuplicateKey);
}
}
let dict = IDict {
dict_start: dict_start as Pos,
dict_end: self.dicts.len() as Pos,
};
Ok(self.push_term(TTerm::Dict(dict)))
}
pub fn push_raw(&mut self, raw: &[u8]) -> Result<ITerm, ParseError> {
let bytes_len = self.bytes.len();
let seqs_len = self.seqs.len();
let dicts_len = self.dicts.len();
let terms_len = self.terms.len();
let raw = self.push_bytes(raw);
let result = self.decode(raw);
if result.is_err() {
// reset to initial state
self.bytes.to_mut().truncate(bytes_len);
self.seqs.truncate(seqs_len);
self.dicts.truncate(dicts_len);
self.terms.truncate(terms_len);
}
if let Err(ParseError::UnexpectedInput(pos)) = result {
Err(ParseError::UnexpectedInput(pos - bytes_len))
} else {
result
}
}
// ==== Internal ====
#[inline]
fn get_term(&self, term: ITerm) -> Result<&TTerm, TermError> {
self.terms
.get(term.0 as usize)
.ok_or(TermError::InvalidIndex(term))
}
#[inline]
fn push_term(&mut self, term: TTerm) -> ITerm {
let ret = ITerm(self.terms.len() as Pos);
self.terms.push(term);
ret
}
#[inline]
fn push_bytes(&mut self, raw: &[u8]) -> IRaw {
let bytes_start = self.bytes.len();
self.bytes.to_mut().extend(raw);
IRaw {
start: bytes_start as Pos,
end: self.bytes.len() as Pos,
}
}
}
impl TTerm {
fn typename(&self) -> &'static str {
match self {
TTerm::Str(_) | TTerm::StrInline(_, _) => "string",
TTerm::RawSeq(_, _) | TTerm::Seq(_) => "seq",
TTerm::RawList(_, _) | TTerm::List(_) => "list",
TTerm::RawDict(_, _) | TTerm::Dict(_) => "dict",
}
}
}
#[cfg(test)]
mod tests {
pub use super::*;
#[test]
fn test_sizeof() {
assert_eq!(std::mem::size_of::<TTerm>(), 20);
}
}

View file

@ -1,48 +1,158 @@
//! Helpers to use cryptographic data types in nettext //! Helpers to use cryptographic data types in nettext
pub use dryoc::*; pub use dryoc;
use dryoc::types::{Bytes, StackByteArray}; use dryoc::types::Bytes;
use dryoc::*;
use crate::dec;
use crate::enc; use crate::enc;
pub type SigningKeyPair = sign::SigningKeyPair<sign::PublicKey, sign::SecretKey>; const BM_HASH: &str = "h.b2";
impl<const N: usize> enc::Encode for StackByteArray<N> { const BM_SIGNATURE: &str = "sig.ed25519";
fn term(&self) -> enc::Result<'_> { const BM_SIGN_KEYPAIR: &str = "sk.ed25519";
Ok(enc::bytes(self.as_slice())) const BM_SIGN_PUBKEY: &str = "pk.ed25519";
// ---- types ----
#[derive(Eq, PartialEq, Clone, Debug)]
pub struct Hash(pub generichash::Hash);
#[derive(Eq, PartialEq, Clone, Debug)]
pub struct Signature(pub sign::Signature);
#[derive(Eq, PartialEq, Clone, Debug)]
pub struct SigningPublicKey(pub sign::PublicKey);
#[derive(PartialEq, Clone, Debug)]
pub struct SigningKeyPair(pub sign::SigningKeyPair<sign::PublicKey, sign::SecretKey>);
impl SigningKeyPair {
/// Return the public key of this keypair
pub fn public_key(&self) -> SigningPublicKey {
SigningPublicKey(self.0.public_key.clone())
} }
} }
impl enc::Encode for sign::SigningKeyPair<sign::PublicKey, sign::SecretKey> { // ---- encoding ----
impl enc::Encode for Hash {
fn term(&self) -> enc::Result<'_> { fn term(&self) -> enc::Result<'_> {
Ok(enc::bytes(self.secret_key.as_slice())) enc::marked_bytes(BM_HASH, self.0.as_slice())
} }
} }
// ---- helpers ---- impl enc::Encode for Signature {
fn term(&self) -> enc::Result<'_> {
enc::marked_bytes(BM_SIGNATURE, self.0.as_slice())
}
}
impl enc::Encode for SigningPublicKey {
fn term(&self) -> enc::Result<'_> {
enc::marked_bytes(BM_SIGN_PUBKEY, self.0.as_slice())
}
}
impl enc::Encode for SigningKeyPair {
fn term(&self) -> enc::Result<'_> {
enc::marked_bytes(BM_SIGN_KEYPAIR, self.0.secret_key.as_slice())
}
}
// ---- calculating hashes, signatures, etc ----
/// Compute the hash of a payload with default dryoc parameters and optionnal key /// Compute the hash of a payload with default dryoc parameters and optionnal key
pub fn compute_hash(bytes: &[u8], key: Option<&[u8; 32]>) -> generichash::Hash { pub fn compute_hash(bytes: &[u8], key: Option<&[u8; 32]>) -> Hash {
generichash::GenericHash::hash_with_defaults(bytes, key).unwrap() Hash(generichash::GenericHash::hash_with_defaults(bytes, key).unwrap())
}
/// Generate a new signing keypair
pub fn gen_signing_keypair() -> SigningKeyPair {
SigningKeyPair(sign::SigningKeyPair::gen_with_defaults())
} }
/// Compute the ed25519 signature of a message using a secret key /// Compute the ed25519 signature of a message using a secret key
pub fn compute_signature(message: &[u8], secret_key: &sign::SecretKey) -> sign::Signature { pub fn compute_signature(message: &[u8], keypair: &SigningKeyPair) -> Signature {
SigningKeyPair::from_secret_key(secret_key.clone()) Signature(
keypair
.0
.sign_with_defaults(message) .sign_with_defaults(message)
.unwrap() .unwrap()
.into_parts() .into_parts()
.0 .0,
)
} }
/// Verify the ed25519 signature of a message using a public key /// Verify the ed25519 signature of a message using a public key
pub fn verify_signature( pub fn verify_signature(
signature: sign::Signature, signature: &Signature,
message: Vec<u8>, message: &[u8],
public_key: &sign::PublicKey, public_key: &SigningPublicKey,
) -> bool { ) -> bool {
sign::SignedMessage::from_parts(signature.clone(), message) sign::SignedMessage::from_parts(signature.0.clone(), message.to_vec())
.verify(public_key) .verify(&public_key.0)
.is_ok() .is_ok()
} }
// ---- decode helpers ----
pub trait CryptoDec {
/// Try to interpret this string as a Blake2b512 digest
/// (32-bytes base64 encoded, prefixed by `h.b2:`)
///
/// Example:
///
/// ```
/// use nettext::dec::decode;
/// use nettext::crypto::{compute_hash, CryptoDec};
///
/// let term = decode(b"{
/// message = hello;
/// hash = h.b2:Mk3PAn3UowqTLEQfNlol6GsXPe-kuOWJSCU0cbgbcs8;
/// }").unwrap();
/// let [msg, hash] = term.dict_of(["message", "hash"], false).unwrap();
/// let expected_hash = compute_hash(msg.raw(), None);
/// assert_eq!(hash.hash().unwrap(), expected_hash);
/// ```
fn hash(&self) -> Result<Hash, dec::TypeError>;
/// Try to interpret this string as an ed25519 signature
/// (64 bytes base64 encoded, prefixed by `sig.ed25519:`)
fn signature(&self) -> Result<Signature, dec::TypeError>;
/// Try to interpret this string as an ed25519 keypair
/// (64 bytes base64 encoded, prefixed by `sk.ed25519:`)
fn keypair(&self) -> Result<SigningKeyPair, dec::TypeError>;
/// Try to interpret this string as an ed25519 public key
/// (32 bytes base64 encoded, prefixed by `pk.ed25519:`)
fn public_key(&self) -> Result<SigningPublicKey, dec::TypeError>;
}
impl<'a, 'b> CryptoDec for dec::Term<'a, 'b> {
fn hash(&self) -> Result<Hash, dec::TypeError> {
Ok(Hash(generichash::Hash::from(
self.marked_bytes_exact(BM_HASH)?,
)))
}
/// Try to interpret this string as an ed25519 signature (64 bytes base64 encoded)
fn signature(&self) -> Result<Signature, dec::TypeError> {
Ok(Signature(sign::Signature::from(
self.marked_bytes_exact(BM_SIGNATURE)?,
)))
}
fn keypair(&self) -> Result<SigningKeyPair, dec::TypeError> {
let secret_key = sign::SecretKey::from(self.marked_bytes_exact(BM_SIGN_KEYPAIR)?);
Ok(SigningKeyPair(sign::SigningKeyPair::from_secret_key(
secret_key,
)))
}
fn public_key(&self) -> Result<SigningPublicKey, dec::TypeError> {
Ok(SigningPublicKey(sign::PublicKey::from(
self.marked_bytes_exact(BM_SIGN_PUBKEY)?,
)))
}
}

View file

@ -8,6 +8,9 @@ pub enum TypeError {
/// The term could not be decoded in the given type /// The term could not be decoded in the given type
#[error(display = "Not a {}", _0)] #[error(display = "Not a {}", _0)]
WrongType(&'static str), WrongType(&'static str),
/// The term did not have the correct marker
#[error(display = "Byte marker was not {}", _0)]
WrongMarker(&'static str),
/// The term is not an array of the requested length /// The term is not an array of the requested length
#[error(display = "Expected {} items, got {}", _0, _1)] #[error(display = "Expected {} items, got {}", _0, _1)]

View file

@ -5,9 +5,6 @@ mod error;
use std::collections::HashMap; use std::collections::HashMap;
#[cfg(any(feature = "dryoc"))]
use crate::crypto;
use crate::debug; use crate::debug;
pub use decode::*; pub use decode::*;
@ -420,7 +417,7 @@ impl<'a, 'b> Term<'a, 'b> {
}; };
match self.0.mkref() { match self.0.mkref() {
AnyTerm::Str(encoded) => { AnyTerm::Str(encoded) => {
if encoded == b"." { if encoded == b"-" {
Ok(vec![]) Ok(vec![])
} else { } else {
decode(encoded) decode(encoded)
@ -442,68 +439,40 @@ impl<'a, 'b> Term<'a, 'b> {
} }
/// Try to interpret this string as base64-encoded bytes, /// Try to interpret this string as base64-encoded bytes,
/// with an exact length. /// with a marker prefix and an exact byte length.
/// This is typically used for cryptographic data types such as hashes,
/// keys, signatures, ...
/// ///
/// Example: /// Example:
/// ///
/// ``` /// ```
/// use nettext::dec::decode; /// use nettext::dec::decode;
/// ///
/// let term = decode(b"aGVsbG8sIHdvcmxkIQ").unwrap(); /// let term = decode(b"test:aGVsbG8sIHdvcmxkIQ").unwrap();
/// assert_eq!(&term.bytes_exact::<13>().unwrap(), b"hello, world!"); /// assert_eq!(&term.marked_bytes_exact::<13>("test").unwrap(), b"hello, world!");
/// ``` /// ```
pub fn bytes_exact<const N: usize>(&self) -> Result<[u8; N], TypeError> { pub fn marked_bytes_exact<const N: usize>(
let bytes = self.bytes()?; &self,
marker: &'static str,
) -> Result<[u8; N], TypeError> {
let mkr = marker.as_bytes();
match &self.0 {
AnyTerm::Str(s)
if s.len() >= mkr.len() + 2 && &s[..mkr.len()] == mkr && s[mkr.len()] == b':' =>
{
let bytes = match &s[mkr.len() + 1..] {
b"-" => vec![],
bytes => base64::decode_config(bytes, base64::URL_SAFE_NO_PAD)
.map_err(|_| TypeError::WrongType("BYTES"))?,
};
let bytes_len = bytes.len(); let bytes_len = bytes.len();
bytes bytes
.try_into() .try_into()
.map_err(|_| TypeError::WrongLength(bytes_len, N)) .map_err(|_| TypeError::WrongLength(bytes_len, N))
} }
} AnyTerm::Str(_) => Err(TypeError::WrongMarker(marker)),
_ => Err(TypeError::WrongType("BYTES")),
// ---- CRYPTO HELPERS ----
#[cfg(feature = "dryoc")]
impl<'a, 'b> Term<'a, 'b> {
/// Try to interpret this string as a Blake2b512 digest (32-bytes base64 encoded)
///
/// Example:
///
/// ```
/// use nettext::dec::decode;
/// use nettext::crypto::generichash::GenericHash;
///
/// let term = decode(b"{
/// message = hello;
/// hash = Mk3PAn3UowqTLEQfNlol6GsXPe-kuOWJSCU0cbgbcs8;
/// }").unwrap();
/// let [msg, hash] = term.dict_of(["message", "hash"], false).unwrap();
/// let expected_hash = GenericHash::hash_with_defaults(msg.raw(), None::<&Vec<u8>>).unwrap();
/// assert_eq!(hash.hash().unwrap(), expected_hash);
/// ```
pub fn hash(&self) -> Result<crypto::generichash::Hash, TypeError> {
Ok(crypto::generichash::Hash::from(self.bytes_exact()?))
} }
/// Try to interpret this string as an ed25519 keypair (64 bytes base64 encoded)
pub fn keypair(&self) -> Result<crypto::SigningKeyPair, TypeError> {
let secret_key = crypto::sign::SecretKey::from(self.bytes_exact()?);
Ok(crypto::SigningKeyPair::from_secret_key(secret_key))
}
/// Try to interpret this string as an ed25519 public key (32 bytes base64 encoded)
pub fn public_key(&self) -> Result<crypto::sign::PublicKey, TypeError> {
Ok(crypto::sign::PublicKey::from(self.bytes_exact()?))
}
/// Try to interpret this string as an ed25519 secret key (32 bytes base64 encoded)
pub fn secret_key(&self) -> Result<crypto::sign::SecretKey, TypeError> {
Ok(crypto::sign::SecretKey::from(self.bytes_exact()?))
}
/// Try to interpret this string as an ed25519 signature (64 bytes base64 encoded)
pub fn signature(&self) -> Result<crypto::sign::Signature, TypeError> {
Ok(crypto::sign::Signature::from(self.bytes_exact()?))
} }
} }

View file

@ -24,7 +24,7 @@ use std::collections::HashMap;
use crate::dec::{self, decode}; use crate::dec::{self, decode};
use crate::*; use crate::*;
use crate::{is_string_char, is_whitespace, switch64, BytesEncoding}; use crate::{is_string_char, is_whitespace};
pub use error::Error; pub use error::Error;
@ -100,41 +100,36 @@ pub fn raw(bytes: &[u8]) -> Result<'_> {
/// Term corresponding to a byte slice, /// Term corresponding to a byte slice,
/// encoding using base64 url-safe encoding without padding. /// encoding using base64 url-safe encoding without padding.
/// Since empty strings are not possible in nettext, /// Since empty strings are not possible in nettext,
/// an empty byte string is encoded as an empty list (`[]`). /// an empty byte string is encoded as the special string `-`.
/// ///
/// Example: /// Example:
/// ///
/// ``` /// ```
/// use nettext::enc::*; /// use nettext::enc::*;
/// ///
/// assert_eq!(bytes(b"").encode(), b"-");
/// assert_eq!(bytes(b"hello, world!").encode(), b"aGVsbG8sIHdvcmxkIQ"); /// assert_eq!(bytes(b"hello, world!").encode(), b"aGVsbG8sIHdvcmxkIQ");
/// ``` /// ```
pub fn bytes(bytes: &[u8]) -> Term<'static> { pub fn bytes(bytes: &[u8]) -> Term<'static> {
bytes_format(bytes, BytesEncoding::Base64 { split: false }) if bytes.is_empty() {
Term(T::Str(b"-"))
} else {
Term(T::OwnedStr(
base64::encode_config(bytes, base64::URL_SAFE_NO_PAD).into_bytes(),
))
}
} }
/// Same as `bytes()`, but splits the byte slice in 48-byte chunks /// Same as `bytes()`, but splits the byte slice in 48-byte chunks
/// and encodes each chunk separately, putting them in a sequence of terms. /// and encodes each chunk separately, putting them in a sequence of terms.
/// Usefull for long byte slices to have cleaner representations, /// Usefull for long byte slices to have cleaner representations.
/// mainly usefull for dictionnary keys.
pub fn bytes_split(bytes: &[u8]) -> Term<'static> { pub fn bytes_split(bytes: &[u8]) -> Term<'static> {
bytes_format(bytes, BytesEncoding::Base64 { split: true }) if bytes.is_empty() {
} Term(T::Str(b"-"))
} else {
pub fn bytes_format(bytes: &[u8], encoding: BytesEncoding) -> Term<'static> {
match encoding {
BytesEncoding::Base64 { .. } | BytesEncoding::Hex { .. } if bytes.is_empty() => {
Term(T::List(vec![]))
}
BytesEncoding::Base64 { split: false } => Term(T::OwnedStr(
base64::encode_config(bytes, base64::URL_SAFE_NO_PAD).into_bytes(),
)),
BytesEncoding::Base64 { split: true } => {
let chunks = bytes let chunks = bytes
.chunks(48) .chunks(48)
.map(|b| { .map(|b| T::OwnedStr(base64::encode_config(b, base64::URL_SAFE_NO_PAD).into_bytes()))
T::OwnedStr(base64::encode_config(b, base64::URL_SAFE_NO_PAD).into_bytes())
})
.collect::<Vec<_>>(); .collect::<Vec<_>>();
if chunks.len() > 1 { if chunks.len() > 1 {
Term(T::Seq(chunks)) Term(T::Seq(chunks))
@ -142,21 +137,39 @@ pub fn bytes_format(bytes: &[u8], encoding: BytesEncoding) -> Term<'static> {
Term(chunks.into_iter().next().unwrap()) Term(chunks.into_iter().next().unwrap())
} }
} }
BytesEncoding::Hex { split: false } => Term(T::OwnedStr(hex::encode(bytes).into_bytes())), }
BytesEncoding::Hex { split: true } => {
let chunks = bytes /// Term corresponding to a byte slice,
.chunks(32) /// encoding using base64 url-safe encoding without padding,
.map(|b| T::OwnedStr(hex::encode(b).into_bytes())) /// with a prefix used to identify its content type.
.collect::<Vec<_>>(); /// The marker prefix is typically used in crypto settings to identify
if chunks.len() > 1 { /// a cryptographic protocol or algorithm; it may not contain the `:` character.
Term(T::Seq(chunks)) ///
/// Example:
///
/// ```
/// use nettext::enc::*;
///
/// assert_eq!(marked_bytes("mytype", b"").unwrap().encode(), b"mytype:-");
/// assert_eq!(marked_bytes("mytype", b"hello, world!").unwrap().encode(), b"mytype:aGVsbG8sIHdvcmxkIQ");
/// ```
pub fn marked_bytes(marker: &str, bytes: &[u8]) -> Result<'static> {
for c in marker.as_bytes().iter() {
if !is_string_char(*c) || *c == b':' {
return Err(Error::InvalidCharacter(*c));
}
}
if bytes.is_empty() {
Ok(Term(T::OwnedStr(format!("{}:-", marker).into_bytes())))
} else { } else {
Term(chunks.into_iter().next().unwrap()) Ok(Term(T::OwnedStr(
} format!(
} "{}:{}",
BytesEncoding::Switch64 { allow_whitespace } => { marker,
Term(T::OwnedStr(switch64::encode(bytes, allow_whitespace))) base64::encode_config(bytes, base64::URL_SAFE_NO_PAD)
} )
.into_bytes(),
)))
} }
} }

View file

@ -1,13 +1,12 @@
//! A text-based data format for cryptographic network protocols. //! A text-based data format for cryptographic network protocols.
//! //!
//! ``` //! ```
//! #[cfg(feature = "dryoc")] {
//! use nettext::enc::*; //! use nettext::enc::*;
//! use nettext::dec::*; //! use nettext::dec::*;
//! use nettext::crypto::*; //! use nettext::crypto::*;
//! //!
//! let final_payload = { //! let final_payload = {
//! let keypair = SigningKeyPair::gen_with_defaults(); //! let keypair = gen_signing_keypair();
//! //!
//! // Encode a fist object that represents a payload that will be hashed and signed //! // Encode a fist object that represents a payload that will be hashed and signed
//! let signed_payload = seq([ //! let signed_payload = seq([
@ -19,12 +18,12 @@
//! ("c", raw(b"{ a = 12; b = 42 }").unwrap()), //! ("c", raw(b"{ a = 12; b = 42 }").unwrap()),
//! ("d", bytes_split(&((0..128u8).collect::<Vec<_>>()))), //! ("d", bytes_split(&((0..128u8).collect::<Vec<_>>()))),
//! ]).unwrap(), //! ]).unwrap(),
//! keypair.public_key.term().unwrap(), //! keypair.public_key().term().unwrap(),
//! ]).unwrap().encode(); //! ]).unwrap().encode();
//! eprintln!("{}", std::str::from_utf8(&signed_payload).unwrap()); //! eprintln!("{}", std::str::from_utf8(&signed_payload).unwrap());
//! //!
//! let hash = compute_hash(&signed_payload, None); //! let hash = compute_hash(&signed_payload, None);
//! let sign = compute_signature(&signed_payload[..], &keypair.secret_key); //! let sign = compute_signature(&signed_payload[..], &keypair);
//! //!
//! // Encode a second object that represents the signed and hashed payload //! // Encode a second object that represents the signed and hashed payload
//! dict([ //! dict([
@ -47,11 +46,10 @@
//! //!
//! let [verb, arg1, arg2, pubkey] = object2.seq_of().unwrap(); //! let [verb, arg1, arg2, pubkey] = object2.seq_of().unwrap();
//! let pubkey = pubkey.public_key().unwrap(); //! let pubkey = pubkey.public_key().unwrap();
//! assert!(verify_signature(signature, payload.raw().to_vec(), &pubkey)); //! assert!(verify_signature(&signature, payload.raw(), &pubkey));
//! //!
//! assert_eq!(verb.string().unwrap(), "CALL"); //! assert_eq!(verb.string().unwrap(), "CALL");
//! assert_eq!(arg1.string().unwrap(), "myfunction"); //! assert_eq!(arg1.string().unwrap(), "myfunction");
//! }
//! ``` //! ```
//! //!
//! The value of `signed_payload` would be as follows: //! The value of `signed_payload` would be as follows:
@ -64,13 +62,13 @@
//! d = AAECAwQFBgcICQoLDA0ODxAREhMUFRYXGBkaGxwdHh8gISIjJCUmJygpKissLS4v //! d = AAECAwQFBgcICQoLDA0ODxAREhMUFRYXGBkaGxwdHh8gISIjJCUmJygpKissLS4v
//! MDEyMzQ1Njc4OTo7PD0-P0BBQkNERUZHSElKS0xNTk9QUVJTVFVWV1hZWltcXV5f //! MDEyMzQ1Njc4OTo7PD0-P0BBQkNERUZHSElKS0xNTk9QUVJTVFVWV1hZWltcXV5f
//! YGFiY2RlZmdoaWprbG1ub3BxcnN0dXZ3eHl6e3x9fn8; //! YGFiY2RlZmdoaWprbG1ub3BxcnN0dXZ3eHl6e3x9fn8;
//! } ZCkE-mTMlK3355u_0UzabRbSNcNO3CWAur7dAhglYtI //! } pk.ed25519:inYgWFyL_BzZTsXNKp71r2aVct_3Izi_bkerbzOiz94
//! ``` //! ```
//! //!
//! And the value of `final_payload` would be as follows: //! And the value of `final_payload` would be as follows:
//! ```raw //! ```raw
//! { //! {
//! hash = fTTk8Hm0HLGwaskCIqFBzRVMrVTeXGetmNBK2X3pNyY; //! hash = h.b2:B1AnRocS90DmqxynGyvvBNuh-brucNO7-5hrsGplJr0;
//! payload = CALL myfunction { //! payload = CALL myfunction {
//! a = hello; //! a = hello;
//! b = world; //! b = world;
@ -78,18 +76,16 @@
//! d = AAECAwQFBgcICQoLDA0ODxAREhMUFRYXGBkaGxwdHh8gISIjJCUmJygpKissLS4v //! d = AAECAwQFBgcICQoLDA0ODxAREhMUFRYXGBkaGxwdHh8gISIjJCUmJygpKissLS4v
//! MDEyMzQ1Njc4OTo7PD0-P0BBQkNERUZHSElKS0xNTk9QUVJTVFVWV1hZWltcXV5f //! MDEyMzQ1Njc4OTo7PD0-P0BBQkNERUZHSElKS0xNTk9QUVJTVFVWV1hZWltcXV5f
//! YGFiY2RlZmdoaWprbG1ub3BxcnN0dXZ3eHl6e3x9fn8; //! YGFiY2RlZmdoaWprbG1ub3BxcnN0dXZ3eHl6e3x9fn8;
//! } ZCkE-mTMlK3355u_0UzabRbSNcNO3CWAur7dAhglYtI; //! } pk.ed25519:inYgWFyL_BzZTsXNKp71r2aVct_3Izi_bkerbzOiz94;
//! signature = XPMrlhAIMfZb6a5Fh5F_ZaEf61olJ1hK4I2kh7vEPT1n20S-943X5cH35bb0Bfwkvy_ENfOTbb3ep1zn2lSIBg; //! signature = sig.ed25519:LvLC1gHxNxUH44HHQRO-zWtLM4WyXhiYLFr94qTdI311Wa-kmgZsaWqSWe3jcjkS4PnsWSNt5apgbhR68cWWCg;
//! } //! }
//! ``` //! ```
//! //!
//! Note that the value of `text1` is embedded as-is inside `text2`. This is what allows us //! Note that the value of `text1` is embedded as-is inside `text2`. This is what allows us
//! to check the hash and the signature: the raw representation of the term hasn't changed. //! to check the hash and the signature: the raw representation of the term hasn't changed.
pub mod buf;
pub mod dec; pub mod dec;
pub mod enc; pub mod enc;
pub mod switch64;
#[cfg(feature = "dryoc")] #[cfg(feature = "dryoc")]
pub mod crypto; pub mod crypto;
@ -97,35 +93,6 @@ pub mod crypto;
#[cfg(feature = "serde")] #[cfg(feature = "serde")]
pub mod serde; pub mod serde;
/// Possible encodings for byte strings in NetText
#[derive(Clone, Copy)]
pub enum BytesEncoding {
/// Base64 encoding (default)
Base64 { split: bool },
/// Hexadecimal encoding
Hex { split: bool },
/// Switch64 encoding, a mix of plain text and base64
Switch64 { allow_whitespace: bool },
}
impl Default for BytesEncoding {
fn default() -> Self {
BytesEncoding::Base64 { split: true }
}
}
impl BytesEncoding {
pub fn without_whitespace(&self) -> Self {
match self {
BytesEncoding::Base64 { .. } => BytesEncoding::Base64 { split: false },
BytesEncoding::Hex { .. } => BytesEncoding::Hex { split: false },
BytesEncoding::Switch64 { .. } => BytesEncoding::Switch64 {
allow_whitespace: false,
},
}
}
}
// ---- syntactic elements of the data format ---- // ---- syntactic elements of the data format ----
pub(crate) const DICT_OPEN: u8 = b'{'; pub(crate) const DICT_OPEN: u8 = b'{';
@ -135,12 +102,9 @@ pub(crate) const DICT_DELIM: u8 = b';';
pub(crate) const LIST_OPEN: u8 = b'['; pub(crate) const LIST_OPEN: u8 = b'[';
pub(crate) const LIST_CLOSE: u8 = b']'; pub(crate) const LIST_CLOSE: u8 = b']';
pub(crate) const LIST_DELIM: u8 = b';'; pub(crate) const LIST_DELIM: u8 = b';';
const BASE_EXTRA_CHARS: &[u8] = br#".,:?!@$^<>|&#"'_-+*/%"#; const BASE_EXTRA_CHARS: &[u8] = b".,:?!@$^<>|&#'_-+*/%";
const STR_EXTRA_CHARS: &[u8] = b"\\"; const STR_EXTRA_CHARS: &[u8] = b"\\";
pub(crate) const SWITCH64_SEPARATOR: u8 = b'\\';
pub(crate) const SWITCH64_EXTRA_CHARS: &[u8] = BASE_EXTRA_CHARS;
#[inline] #[inline]
pub(crate) fn is_string_char(c: u8) -> bool { pub(crate) fn is_string_char(c: u8) -> bool {
c.is_ascii_alphanumeric() || BASE_EXTRA_CHARS.contains(&c) || STR_EXTRA_CHARS.contains(&c) c.is_ascii_alphanumeric() || BASE_EXTRA_CHARS.contains(&c) || STR_EXTRA_CHARS.contains(&c)

View file

@ -4,7 +4,6 @@ mod de;
mod error; mod error;
mod ser; mod ser;
pub use crate::BytesEncoding;
pub use de::{from_bytes, from_term, Deserializer}; pub use de::{from_bytes, from_term, Deserializer};
pub use error::{Error, Result}; pub use error::{Error, Result};
pub use ser::{to_bytes, to_term, Serializer}; pub use ser::{to_bytes, to_term, Serializer};
@ -32,17 +31,6 @@ mod tests {
eprintln!("Serialized (concise): {}", ser_concise); eprintln!("Serialized (concise): {}", ser_concise);
assert_eq!(ser_concise, expected_concise); assert_eq!(ser_concise, expected_concise);
assert_eq!(from_bytes::<T>(ser_concise.as_bytes()).unwrap(), input); assert_eq!(from_bytes::<T>(ser_concise.as_bytes()).unwrap(), input);
// let ser_str_hex = input
// .serialize(&mut Serializer {
// string_format: BytesEncoding::Switch64 {
// allow_whitespace: true,
// },
// bytes_format: BytesEncoding::Hex { split: true },
// })
// .unwrap()
// .encode();
// panic!("{}", debug(&ser_str_hex));
} }
#[test] #[test]

View file

@ -2,15 +2,11 @@ use serde::{ser, Serialize};
use crate::enc::*; use crate::enc::*;
use crate::serde::error::{Error, Result}; use crate::serde::error::{Error, Result};
use crate::BytesEncoding;
use serde::ser::Error as SerError; use serde::ser::Error as SerError;
/// Serde serializer for nettext /// Serde serializer for nettext
#[derive(Clone, Copy, Default)] #[derive(Clone, Copy, Default)]
pub struct Serializer { pub struct Serializer;
pub string_format: BytesEncoding,
pub bytes_format: BytesEncoding,
}
/// Serialize value to nettext encoder term /// Serialize value to nettext encoder term
pub fn to_term<T>(value: &T) -> Result<Term<'static>> pub fn to_term<T>(value: &T) -> Result<Term<'static>>
@ -94,11 +90,11 @@ impl<'a> ser::Serializer for &'a mut Serializer {
} }
fn serialize_str(self, v: &str) -> Result<Self::Ok> { fn serialize_str(self, v: &str) -> Result<Self::Ok> {
Ok(bytes_format(v.as_bytes(), self.string_format)) Ok(bytes(v.as_bytes()))
} }
fn serialize_bytes(self, v: &[u8]) -> Result<Self::Ok> { fn serialize_bytes(self, v: &[u8]) -> Result<Self::Ok> {
Ok(bytes_format(v, self.bytes_format)) Ok(bytes(v))
} }
fn serialize_none(self) -> Result<Self::Ok> { fn serialize_none(self) -> Result<Self::Ok> {
@ -313,10 +309,7 @@ impl ser::SerializeMap for MapSerializer {
where where
T: ?Sized + Serialize, T: ?Sized + Serialize,
{ {
let mut ser = Serializer { let mut ser = Serializer;
string_format: self.ser.string_format.without_whitespace(),
bytes_format: self.ser.bytes_format.without_whitespace(),
};
self.next = Some(key.serialize(&mut ser)?.encode()); self.next = Some(key.serialize(&mut ser)?.encode());
Ok(()) Ok(())
} }

View file

@ -1,131 +0,0 @@
//! The Switch64 encoding for text strings
//!
//! Allowed characters are encoded as-is.
//! Others are encoded using base64.
//! Plain parts and base64-encoded parts are separated by a backslasah `\`
use crate::{SWITCH64_EXTRA_CHARS, SWITCH64_SEPARATOR};
pub fn encode(bytes: &[u8], allow_whitespace: bool) -> Vec<u8> {
let mut output = Vec::with_capacity(bytes.len());
let mut pos = 0;
while pos < bytes.len() {
// Determine how many bytes to copy as-is
let cnt = bytes[pos..]
.iter()
.take_while(|c| is_valid_plaintext_char(**c, allow_whitespace))
.count();
// Copy those bytes as-is
output.extend_from_slice(&bytes[pos..pos + cnt]);
pos += cnt;
// If some bytes remain, switch to base64 encoding
if pos < bytes.len() {
output.push(SWITCH64_SEPARATOR);
} else {
break;
}
// Count how many bytes to write as base64
// We stop at the first position where we find three consecutive
// characters to encode as-is
let mut b64end = bytes.len();
for i in pos..bytes.len() {
if i + 3 > bytes.len() {
break;
}
if bytes[i..i + 3]
.iter()
.all(|c| is_valid_plaintext_char(*c, allow_whitespace))
{
b64end = i;
break;
}
}
output.extend_from_slice(
base64::encode_config(&bytes[pos..b64end], base64::URL_SAFE_NO_PAD).as_bytes(),
);
pos = b64end;
if pos < bytes.len() {
output.push(SWITCH64_SEPARATOR);
}
}
output
}
pub fn decode(bytes: &[u8]) -> Result<Vec<u8>, base64::DecodeError> {
let mut output = Vec::with_capacity(bytes.len());
let mut pos = 0;
while pos < bytes.len() {
let cnt = bytes[pos..]
.iter()
.take_while(|c| **c != SWITCH64_SEPARATOR)
.count();
output.extend_from_slice(&bytes[pos..pos + cnt]);
pos += cnt + 1;
if pos >= bytes.len() {
break;
}
let cnt = bytes[pos..]
.iter()
.take_while(|c| **c != SWITCH64_SEPARATOR)
.count();
output.extend_from_slice(&base64::decode_config(
&bytes[pos..pos + cnt],
base64::URL_SAFE_NO_PAD,
)?);
pos += cnt + 1;
}
Ok(output)
}
#[inline]
fn is_valid_plaintext_char(c: u8, allow_whitespace: bool) -> bool {
c.is_ascii_alphanumeric()
|| (allow_whitespace && c.is_ascii_whitespace())
|| SWITCH64_EXTRA_CHARS.contains(&c)
}
#[cfg(test)]
mod tests {
use super::*;
use crate::debug;
#[test]
fn test_encode() {
assert_eq!(debug(&encode(&b"hello world"[..], true)), "hello world");
assert_eq!(debug(&encode(&b"hello, world!"[..], true)), "hello, world!");
}
#[test]
fn test_decode() {
assert_eq!(debug(&decode(&b"hello world"[..]).unwrap()), "hello world");
assert_eq!(
debug(&decode(&b"hello\\LA\\ world\\IQ"[..]).unwrap()),
"hello, world!"
);
assert_eq!(debug(&decode(&b"\\LDssQCQ7OA"[..]).unwrap()), ",;,@$;8");
}
#[test]
fn test_encdec() {
for s in [
br#"assert_eq!(debug(&decode(&b"hello\\LA\\ world\\IQ"[..]).unwrap()), "hello, world!");"#.to_vec(),
br#"- a list, which may contain any number of any kind of terms (can be mixed)"#.to_vec(),
base64::decode("dVcG5EzJqGP/2ZGkVu4ewzfAug1W96tb2KiBOVyPUXfw8uD34DEepW/PPqRzi0HL").unwrap(),
br#",;,@$;8"#.to_vec()
] {
assert_eq!(decode(&encode(&s, true)).unwrap(), s);
assert_eq!(decode(&encode(&s, false)).unwrap(), s);
}
}
}