Source code

Revision control

Copy as Markdown

Other Tools

//! Utility functions and types for encoding and decoding Protobuf types.
//!
//! Meant to be used only from `Message` implementations.
#![allow(clippy::implicit_hasher, clippy::ptr_arg)]
use alloc::collections::BTreeMap;
use alloc::format;
use alloc::string::String;
use alloc::vec::Vec;
use core::cmp::min;
use core::convert::TryFrom;
use core::mem;
use core::str;
use core::u32;
use core::usize;
use ::bytes::{Buf, BufMut, Bytes};
use crate::DecodeError;
use crate::Message;
/// Encodes an integer value into LEB128 variable length format, and writes it to the buffer.
/// The buffer must have enough remaining space (maximum 10 bytes).
#[inline]
pub fn encode_varint<B>(mut value: u64, buf: &mut B)
where
B: BufMut,
{
loop {
if value < 0x80 {
buf.put_u8(value as u8);
break;
} else {
buf.put_u8(((value & 0x7F) | 0x80) as u8);
value >>= 7;
}
}
}
/// Decodes a LEB128-encoded variable length integer from the buffer.
#[inline]
pub fn decode_varint<B>(buf: &mut B) -> Result<u64, DecodeError>
where
B: Buf,
{
let bytes = buf.chunk();
let len = bytes.len();
if len == 0 {
return Err(DecodeError::new("invalid varint"));
}
let byte = bytes[0];
if byte < 0x80 {
buf.advance(1);
Ok(u64::from(byte))
} else if len > 10 || bytes[len - 1] < 0x80 {
let (value, advance) = decode_varint_slice(bytes)?;
buf.advance(advance);
Ok(value)
} else {
decode_varint_slow(buf)
}
}
/// Decodes a LEB128-encoded variable length integer from the slice, returning the value and the
/// number of bytes read.
///
/// Based loosely on [`ReadVarint64FromArray`][1] with a varint overflow check from
/// [`ConsumeVarint`][2].
///
/// ## Safety
///
/// The caller must ensure that `bytes` is non-empty and either `bytes.len() >= 10` or the last
/// element in bytes is < `0x80`.
///
#[inline]
fn decode_varint_slice(bytes: &[u8]) -> Result<(u64, usize), DecodeError> {
// Fully unrolled varint decoding loop. Splitting into 32-bit pieces gives better performance.
// Use assertions to ensure memory safety, but it should always be optimized after inline.
assert!(!bytes.is_empty());
assert!(bytes.len() > 10 || bytes[bytes.len() - 1] < 0x80);
let mut b: u8 = unsafe { *bytes.get_unchecked(0) };
let mut part0: u32 = u32::from(b);
if b < 0x80 {
return Ok((u64::from(part0), 1));
};
part0 -= 0x80;
b = unsafe { *bytes.get_unchecked(1) };
part0 += u32::from(b) << 7;
if b < 0x80 {
return Ok((u64::from(part0), 2));
};
part0 -= 0x80 << 7;
b = unsafe { *bytes.get_unchecked(2) };
part0 += u32::from(b) << 14;
if b < 0x80 {
return Ok((u64::from(part0), 3));
};
part0 -= 0x80 << 14;
b = unsafe { *bytes.get_unchecked(3) };
part0 += u32::from(b) << 21;
if b < 0x80 {
return Ok((u64::from(part0), 4));
};
part0 -= 0x80 << 21;
let value = u64::from(part0);
b = unsafe { *bytes.get_unchecked(4) };
let mut part1: u32 = u32::from(b);
if b < 0x80 {
return Ok((value + (u64::from(part1) << 28), 5));
};
part1 -= 0x80;
b = unsafe { *bytes.get_unchecked(5) };
part1 += u32::from(b) << 7;
if b < 0x80 {
return Ok((value + (u64::from(part1) << 28), 6));
};
part1 -= 0x80 << 7;
b = unsafe { *bytes.get_unchecked(6) };
part1 += u32::from(b) << 14;
if b < 0x80 {
return Ok((value + (u64::from(part1) << 28), 7));
};
part1 -= 0x80 << 14;
b = unsafe { *bytes.get_unchecked(7) };
part1 += u32::from(b) << 21;
if b < 0x80 {
return Ok((value + (u64::from(part1) << 28), 8));
};
part1 -= 0x80 << 21;
let value = value + ((u64::from(part1)) << 28);
b = unsafe { *bytes.get_unchecked(8) };
let mut part2: u32 = u32::from(b);
if b < 0x80 {
return Ok((value + (u64::from(part2) << 56), 9));
};
part2 -= 0x80;
b = unsafe { *bytes.get_unchecked(9) };
part2 += u32::from(b) << 7;
// Check for u64::MAX overflow. See [`ConsumeVarint`][1] for details.
if b < 0x02 {
return Ok((value + (u64::from(part2) << 56), 10));
};
// We have overrun the maximum size of a varint (10 bytes) or the final byte caused an overflow.
// Assume the data is corrupt.
Err(DecodeError::new("invalid varint"))
}
/// Decodes a LEB128-encoded variable length integer from the buffer, advancing the buffer as
/// necessary.
///
/// Contains a varint overflow check from [`ConsumeVarint`][1].
///
#[inline(never)]
#[cold]
fn decode_varint_slow<B>(buf: &mut B) -> Result<u64, DecodeError>
where
B: Buf,
{
let mut value = 0;
for count in 0..min(10, buf.remaining()) {
let byte = buf.get_u8();
value |= u64::from(byte & 0x7F) << (count * 7);
if byte <= 0x7F {
// Check for u64::MAX overflow. See [`ConsumeVarint`][1] for details.
if count == 9 && byte >= 0x02 {
return Err(DecodeError::new("invalid varint"));
} else {
return Ok(value);
}
}
}
Err(DecodeError::new("invalid varint"))
}
/// Additional information passed to every decode/merge function.
///
/// The context should be passed by value and can be freely cloned. When passing
/// to a function which is decoding a nested object, then use `enter_recursion`.
#[derive(Clone, Debug)]
#[cfg_attr(feature = "no-recursion-limit", derive(Default))]
pub struct DecodeContext {
/// How many times we can recurse in the current decode stack before we hit
/// the recursion limit.
///
/// The recursion limit is defined by `RECURSION_LIMIT` and cannot be
/// customized. The recursion limit can be ignored by building the Prost
/// crate with the `no-recursion-limit` feature.
#[cfg(not(feature = "no-recursion-limit"))]
recurse_count: u32,
}
#[cfg(not(feature = "no-recursion-limit"))]
impl Default for DecodeContext {
#[inline]
fn default() -> DecodeContext {
DecodeContext {
recurse_count: crate::RECURSION_LIMIT,
}
}
}
impl DecodeContext {
/// Call this function before recursively decoding.
///
/// There is no `exit` function since this function creates a new `DecodeContext`
/// to be used at the next level of recursion. Continue to use the old context
// at the previous level of recursion.
#[cfg(not(feature = "no-recursion-limit"))]
#[inline]
pub(crate) fn enter_recursion(&self) -> DecodeContext {
DecodeContext {
recurse_count: self.recurse_count - 1,
}
}
#[cfg(feature = "no-recursion-limit")]
#[inline]
pub(crate) fn enter_recursion(&self) -> DecodeContext {
DecodeContext {}
}
/// Checks whether the recursion limit has been reached in the stack of
/// decodes described by the `DecodeContext` at `self.ctx`.
///
/// Returns `Ok<()>` if it is ok to continue recursing.
/// Returns `Err<DecodeError>` if the recursion limit has been reached.
#[cfg(not(feature = "no-recursion-limit"))]
#[inline]
pub(crate) fn limit_reached(&self) -> Result<(), DecodeError> {
if self.recurse_count == 0 {
Err(DecodeError::new("recursion limit reached"))
} else {
Ok(())
}
}
#[cfg(feature = "no-recursion-limit")]
#[inline]
#[allow(clippy::unnecessary_wraps)] // needed in other features
pub(crate) fn limit_reached(&self) -> Result<(), DecodeError> {
Ok(())
}
}
/// Returns the encoded length of the value in LEB128 variable length format.
/// The returned value will be between 1 and 10, inclusive.
#[inline]
pub fn encoded_len_varint(value: u64) -> usize {
// Based on [VarintSize64][1].
((((value | 1).leading_zeros() ^ 63) * 9 + 73) / 64) as usize
}
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
#[repr(u8)]
pub enum WireType {
Varint = 0,
SixtyFourBit = 1,
LengthDelimited = 2,
StartGroup = 3,
EndGroup = 4,
ThirtyTwoBit = 5,
}
pub const MIN_TAG: u32 = 1;
pub const MAX_TAG: u32 = (1 << 29) - 1;
impl TryFrom<u64> for WireType {
type Error = DecodeError;
#[inline]
fn try_from(value: u64) -> Result<Self, Self::Error> {
match value {
0 => Ok(WireType::Varint),
1 => Ok(WireType::SixtyFourBit),
2 => Ok(WireType::LengthDelimited),
3 => Ok(WireType::StartGroup),
4 => Ok(WireType::EndGroup),
5 => Ok(WireType::ThirtyTwoBit),
_ => Err(DecodeError::new(format!(
"invalid wire type value: {}",
value
))),
}
}
}
/// Encodes a Protobuf field key, which consists of a wire type designator and
/// the field tag.
#[inline]
pub fn encode_key<B>(tag: u32, wire_type: WireType, buf: &mut B)
where
B: BufMut,
{
debug_assert!((MIN_TAG..=MAX_TAG).contains(&tag));
let key = (tag << 3) | wire_type as u32;
encode_varint(u64::from(key), buf);
}
/// Decodes a Protobuf field key, which consists of a wire type designator and
/// the field tag.
#[inline(always)]
pub fn decode_key<B>(buf: &mut B) -> Result<(u32, WireType), DecodeError>
where
B: Buf,
{
let key = decode_varint(buf)?;
if key > u64::from(u32::MAX) {
return Err(DecodeError::new(format!("invalid key value: {}", key)));
}
let wire_type = WireType::try_from(key & 0x07)?;
let tag = key as u32 >> 3;
if tag < MIN_TAG {
return Err(DecodeError::new("invalid tag value: 0"));
}
Ok((tag, wire_type))
}
/// Returns the width of an encoded Protobuf field key with the given tag.
/// The returned width will be between 1 and 5 bytes (inclusive).
#[inline]
pub fn key_len(tag: u32) -> usize {
encoded_len_varint(u64::from(tag << 3))
}
/// Checks that the expected wire type matches the actual wire type,
/// or returns an error result.
#[inline]
pub fn check_wire_type(expected: WireType, actual: WireType) -> Result<(), DecodeError> {
if expected != actual {
return Err(DecodeError::new(format!(
"invalid wire type: {:?} (expected {:?})",
actual, expected
)));
}
Ok(())
}
/// Helper function which abstracts reading a length delimiter prefix followed
/// by decoding values until the length of bytes is exhausted.
pub fn merge_loop<T, M, B>(
value: &mut T,
buf: &mut B,
ctx: DecodeContext,
mut merge: M,
) -> Result<(), DecodeError>
where
M: FnMut(&mut T, &mut B, DecodeContext) -> Result<(), DecodeError>,
B: Buf,
{
let len = decode_varint(buf)?;
let remaining = buf.remaining();
if len > remaining as u64 {
return Err(DecodeError::new("buffer underflow"));
}
let limit = remaining - len as usize;
while buf.remaining() > limit {
merge(value, buf, ctx.clone())?;
}
if buf.remaining() != limit {
return Err(DecodeError::new("delimited length exceeded"));
}
Ok(())
}
pub fn skip_field<B>(
wire_type: WireType,
tag: u32,
buf: &mut B,
ctx: DecodeContext,
) -> Result<(), DecodeError>
where
B: Buf,
{
ctx.limit_reached()?;
let len = match wire_type {
WireType::Varint => decode_varint(buf).map(|_| 0)?,
WireType::ThirtyTwoBit => 4,
WireType::SixtyFourBit => 8,
WireType::LengthDelimited => decode_varint(buf)?,
WireType::StartGroup => loop {
let (inner_tag, inner_wire_type) = decode_key(buf)?;
match inner_wire_type {
WireType::EndGroup => {
if inner_tag != tag {
return Err(DecodeError::new("unexpected end group tag"));
}
break 0;
}
_ => skip_field(inner_wire_type, inner_tag, buf, ctx.enter_recursion())?,
}
},
WireType::EndGroup => return Err(DecodeError::new("unexpected end group tag")),
};
if len > buf.remaining() as u64 {
return Err(DecodeError::new("buffer underflow"));
}
buf.advance(len as usize);
Ok(())
}
/// Helper macro which emits an `encode_repeated` function for the type.
macro_rules! encode_repeated {
($ty:ty) => {
pub fn encode_repeated<B>(tag: u32, values: &[$ty], buf: &mut B)
where
B: BufMut,
{
for value in values {
encode(tag, value, buf);
}
}
};
}
/// Helper macro which emits a `merge_repeated` function for the numeric type.
macro_rules! merge_repeated_numeric {
($ty:ty,
$wire_type:expr,
$merge:ident,
$merge_repeated:ident) => {
pub fn $merge_repeated<B>(
wire_type: WireType,
values: &mut Vec<$ty>,
buf: &mut B,
ctx: DecodeContext,
) -> Result<(), DecodeError>
where
B: Buf,
{
if wire_type == WireType::LengthDelimited {
// Packed.
merge_loop(values, buf, ctx, |values, buf, ctx| {
let mut value = Default::default();
$merge($wire_type, &mut value, buf, ctx)?;
values.push(value);
Ok(())
})
} else {
// Unpacked.
check_wire_type($wire_type, wire_type)?;
let mut value = Default::default();
$merge(wire_type, &mut value, buf, ctx)?;
values.push(value);
Ok(())
}
}
};
}
/// Macro which emits a module containing a set of encoding functions for a
/// variable width numeric type.
macro_rules! varint {
($ty:ty,
$proto_ty:ident) => (
varint!($ty,
$proto_ty,
to_uint64(value) { *value as u64 },
from_uint64(value) { value as $ty });
);
($ty:ty,
$proto_ty:ident,
to_uint64($to_uint64_value:ident) $to_uint64:expr,
from_uint64($from_uint64_value:ident) $from_uint64:expr) => (
pub mod $proto_ty {
use crate::encoding::*;
pub fn encode<B>(tag: u32, $to_uint64_value: &$ty, buf: &mut B) where B: BufMut {
encode_key(tag, WireType::Varint, buf);
encode_varint($to_uint64, buf);
}
pub fn merge<B>(wire_type: WireType, value: &mut $ty, buf: &mut B, _ctx: DecodeContext) -> Result<(), DecodeError> where B: Buf {
check_wire_type(WireType::Varint, wire_type)?;
let $from_uint64_value = decode_varint(buf)?;
*value = $from_uint64;
Ok(())
}
encode_repeated!($ty);
pub fn encode_packed<B>(tag: u32, values: &[$ty], buf: &mut B) where B: BufMut {
if values.is_empty() { return; }
encode_key(tag, WireType::LengthDelimited, buf);
let len: usize = values.iter().map(|$to_uint64_value| {
encoded_len_varint($to_uint64)
}).sum();
encode_varint(len as u64, buf);
for $to_uint64_value in values {
encode_varint($to_uint64, buf);
}
}
merge_repeated_numeric!($ty, WireType::Varint, merge, merge_repeated);
#[inline]
pub fn encoded_len(tag: u32, $to_uint64_value: &$ty) -> usize {
key_len(tag) + encoded_len_varint($to_uint64)
}
#[inline]
pub fn encoded_len_repeated(tag: u32, values: &[$ty]) -> usize {
key_len(tag) * values.len() + values.iter().map(|$to_uint64_value| {
encoded_len_varint($to_uint64)
}).sum::<usize>()
}
#[inline]
pub fn encoded_len_packed(tag: u32, values: &[$ty]) -> usize {
if values.is_empty() {
0
} else {
let len = values.iter()
.map(|$to_uint64_value| encoded_len_varint($to_uint64))
.sum::<usize>();
key_len(tag) + encoded_len_varint(len as u64) + len
}
}
#[cfg(test)]
mod test {
use proptest::prelude::*;
use crate::encoding::$proto_ty::*;
use crate::encoding::test::{
check_collection_type,
check_type,
};
proptest! {
#[test]
fn check(value: $ty, tag in MIN_TAG..=MAX_TAG) {
check_type(value, tag, WireType::Varint,
encode, merge, encoded_len)?;
}
#[test]
fn check_repeated(value: Vec<$ty>, tag in MIN_TAG..=MAX_TAG) {
check_collection_type(value, tag, WireType::Varint,
encode_repeated, merge_repeated,
encoded_len_repeated)?;
}
#[test]
fn check_packed(value: Vec<$ty>, tag in MIN_TAG..=MAX_TAG) {
check_type(value, tag, WireType::LengthDelimited,
encode_packed, merge_repeated,
encoded_len_packed)?;
}
}
}
}
);
}
varint!(bool, bool,
to_uint64(value) u64::from(*value),
from_uint64(value) value != 0);
varint!(i32, int32);
varint!(i64, int64);
varint!(u32, uint32);
varint!(u64, uint64);
varint!(i32, sint32,
to_uint64(value) {
((value << 1) ^ (value >> 31)) as u32 as u64
},
from_uint64(value) {
let value = value as u32;
((value >> 1) as i32) ^ (-((value & 1) as i32))
});
varint!(i64, sint64,
to_uint64(value) {
((value << 1) ^ (value >> 63)) as u64
},
from_uint64(value) {
((value >> 1) as i64) ^ (-((value & 1) as i64))
});
/// Macro which emits a module containing a set of encoding functions for a
/// fixed width numeric type.
macro_rules! fixed_width {
($ty:ty,
$width:expr,
$wire_type:expr,
$proto_ty:ident,
$put:ident,
$get:ident) => {
pub mod $proto_ty {
use crate::encoding::*;
pub fn encode<B>(tag: u32, value: &$ty, buf: &mut B)
where
B: BufMut,
{
encode_key(tag, $wire_type, buf);
buf.$put(*value);
}
pub fn merge<B>(
wire_type: WireType,
value: &mut $ty,
buf: &mut B,
_ctx: DecodeContext,
) -> Result<(), DecodeError>
where
B: Buf,
{
check_wire_type($wire_type, wire_type)?;
if buf.remaining() < $width {
return Err(DecodeError::new("buffer underflow"));
}
*value = buf.$get();
Ok(())
}
encode_repeated!($ty);
pub fn encode_packed<B>(tag: u32, values: &[$ty], buf: &mut B)
where
B: BufMut,
{
if values.is_empty() {
return;
}
encode_key(tag, WireType::LengthDelimited, buf);
let len = values.len() as u64 * $width;
encode_varint(len as u64, buf);
for value in values {
buf.$put(*value);
}
}
merge_repeated_numeric!($ty, $wire_type, merge, merge_repeated);
#[inline]
pub fn encoded_len(tag: u32, _: &$ty) -> usize {
key_len(tag) + $width
}
#[inline]
pub fn encoded_len_repeated(tag: u32, values: &[$ty]) -> usize {
(key_len(tag) + $width) * values.len()
}
#[inline]
pub fn encoded_len_packed(tag: u32, values: &[$ty]) -> usize {
if values.is_empty() {
0
} else {
let len = $width * values.len();
key_len(tag) + encoded_len_varint(len as u64) + len
}
}
#[cfg(test)]
mod test {
use proptest::prelude::*;
use super::super::test::{check_collection_type, check_type};
use super::*;
proptest! {
#[test]
fn check(value: $ty, tag in MIN_TAG..=MAX_TAG) {
check_type(value, tag, $wire_type,
encode, merge, encoded_len)?;
}
#[test]
fn check_repeated(value: Vec<$ty>, tag in MIN_TAG..=MAX_TAG) {
check_collection_type(value, tag, $wire_type,
encode_repeated, merge_repeated,
encoded_len_repeated)?;
}
#[test]
fn check_packed(value: Vec<$ty>, tag in MIN_TAG..=MAX_TAG) {
check_type(value, tag, WireType::LengthDelimited,
encode_packed, merge_repeated,
encoded_len_packed)?;
}
}
}
}
};
}
fixed_width!(
f32,
4,
WireType::ThirtyTwoBit,
float,
put_f32_le,
get_f32_le
);
fixed_width!(
f64,
8,
WireType::SixtyFourBit,
double,
put_f64_le,
get_f64_le
);
fixed_width!(
u32,
4,
WireType::ThirtyTwoBit,
fixed32,
put_u32_le,
get_u32_le
);
fixed_width!(
u64,
8,
WireType::SixtyFourBit,
fixed64,
put_u64_le,
get_u64_le
);
fixed_width!(
i32,
4,
WireType::ThirtyTwoBit,
sfixed32,
put_i32_le,
get_i32_le
);
fixed_width!(
i64,
8,
WireType::SixtyFourBit,
sfixed64,
put_i64_le,
get_i64_le
);
/// Macro which emits encoding functions for a length-delimited type.
macro_rules! length_delimited {
($ty:ty) => {
encode_repeated!($ty);
pub fn merge_repeated<B>(
wire_type: WireType,
values: &mut Vec<$ty>,
buf: &mut B,
ctx: DecodeContext,
) -> Result<(), DecodeError>
where
B: Buf,
{
check_wire_type(WireType::LengthDelimited, wire_type)?;
let mut value = Default::default();
merge(wire_type, &mut value, buf, ctx)?;
values.push(value);
Ok(())
}
#[inline]
pub fn encoded_len(tag: u32, value: &$ty) -> usize {
key_len(tag) + encoded_len_varint(value.len() as u64) + value.len()
}
#[inline]
pub fn encoded_len_repeated(tag: u32, values: &[$ty]) -> usize {
key_len(tag) * values.len()
+ values
.iter()
.map(|value| encoded_len_varint(value.len() as u64) + value.len())
.sum::<usize>()
}
};
}
pub mod string {
use super::*;
pub fn encode<B>(tag: u32, value: &String, buf: &mut B)
where
B: BufMut,
{
encode_key(tag, WireType::LengthDelimited, buf);
encode_varint(value.len() as u64, buf);
buf.put_slice(value.as_bytes());
}
pub fn merge<B>(
wire_type: WireType,
value: &mut String,
buf: &mut B,
ctx: DecodeContext,
) -> Result<(), DecodeError>
where
B: Buf,
{
// ## Unsafety
//
// `string::merge` reuses `bytes::merge`, with an additional check of utf-8
// well-formedness. If the utf-8 is not well-formed, or if any other error occurs, then the
// string is cleared, so as to avoid leaking a string field with invalid data.
//
// This implementation uses the unsafe `String::as_mut_vec` method instead of the safe
// alternative of temporarily swapping an empty `String` into the field, because it results
// in up to 10% better performance on the protobuf message decoding benchmarks.
//
// It's required when using `String::as_mut_vec` that invalid utf-8 data not be leaked into
// the backing `String`. To enforce this, even in the event of a panic in `bytes::merge` or
// in the buf implementation, a drop guard is used.
unsafe {
struct DropGuard<'a>(&'a mut Vec<u8>);
impl<'a> Drop for DropGuard<'a> {
#[inline]
fn drop(&mut self) {
self.0.clear();
}
}
let drop_guard = DropGuard(value.as_mut_vec());
bytes::merge_one_copy(wire_type, drop_guard.0, buf, ctx)?;
match str::from_utf8(drop_guard.0) {
Ok(_) => {
// Success; do not clear the bytes.
mem::forget(drop_guard);
Ok(())
}
Err(_) => Err(DecodeError::new(
"invalid string value: data is not UTF-8 encoded",
)),
}
}
}
length_delimited!(String);
#[cfg(test)]
mod test {
use proptest::prelude::*;
use super::super::test::{check_collection_type, check_type};
use super::*;
proptest! {
#[test]
fn check(value: String, tag in MIN_TAG..=MAX_TAG) {
super::test::check_type(value, tag, WireType::LengthDelimited,
encode, merge, encoded_len)?;
}
#[test]
fn check_repeated(value: Vec<String>, tag in MIN_TAG..=MAX_TAG) {
super::test::check_collection_type(value, tag, WireType::LengthDelimited,
encode_repeated, merge_repeated,
encoded_len_repeated)?;
}
}
}
}
pub trait BytesAdapter: sealed::BytesAdapter {}
mod sealed {
use super::{Buf, BufMut};
pub trait BytesAdapter: Default + Sized + 'static {
fn len(&self) -> usize;
/// Replace contents of this buffer with the contents of another buffer.
fn replace_with<B>(&mut self, buf: B)
where
B: Buf;
/// Appends this buffer to the (contents of) other buffer.
fn append_to<B>(&self, buf: &mut B)
where
B: BufMut;
fn is_empty(&self) -> bool {
self.len() == 0
}
}
}
impl BytesAdapter for Bytes {}
impl sealed::BytesAdapter for Bytes {
fn len(&self) -> usize {
Buf::remaining(self)
}
fn replace_with<B>(&mut self, mut buf: B)
where
B: Buf,
{
*self = buf.copy_to_bytes(buf.remaining());
}
fn append_to<B>(&self, buf: &mut B)
where
B: BufMut,
{
buf.put(self.clone())
}
}
impl BytesAdapter for Vec<u8> {}
impl sealed::BytesAdapter for Vec<u8> {
fn len(&self) -> usize {
Vec::len(self)
}
fn replace_with<B>(&mut self, buf: B)
where
B: Buf,
{
self.clear();
self.reserve(buf.remaining());
self.put(buf);
}
fn append_to<B>(&self, buf: &mut B)
where
B: BufMut,
{
buf.put(self.as_slice())
}
}
pub mod bytes {
use super::*;
pub fn encode<A, B>(tag: u32, value: &A, buf: &mut B)
where
A: BytesAdapter,
B: BufMut,
{
encode_key(tag, WireType::LengthDelimited, buf);
encode_varint(value.len() as u64, buf);
value.append_to(buf);
}
pub fn merge<A, B>(
wire_type: WireType,
value: &mut A,
buf: &mut B,
_ctx: DecodeContext,
) -> Result<(), DecodeError>
where
A: BytesAdapter,
B: Buf,
{
check_wire_type(WireType::LengthDelimited, wire_type)?;
let len = decode_varint(buf)?;
if len > buf.remaining() as u64 {
return Err(DecodeError::new("buffer underflow"));
}
let len = len as usize;
// Clear the existing value. This follows from the following rule in the encoding guide[1]:
//
// > Normally, an encoded message would never have more than one instance of a non-repeated
// > field. However, parsers are expected to handle the case in which they do. For numeric
// > types and strings, if the same field appears multiple times, the parser accepts the
// > last value it sees.
//
//
// This is intended for A and B both being Bytes so it is zero-copy.
// Some combinations of A and B types may cause a double-copy,
// in which case merge_one_copy() should be used instead.
value.replace_with(buf.copy_to_bytes(len));
Ok(())
}
pub(super) fn merge_one_copy<A, B>(
wire_type: WireType,
value: &mut A,
buf: &mut B,
_ctx: DecodeContext,
) -> Result<(), DecodeError>
where
A: BytesAdapter,
B: Buf,
{
check_wire_type(WireType::LengthDelimited, wire_type)?;
let len = decode_varint(buf)?;
if len > buf.remaining() as u64 {
return Err(DecodeError::new("buffer underflow"));
}
let len = len as usize;
// If we must copy, make sure to copy only once.
value.replace_with(buf.take(len));
Ok(())
}
length_delimited!(impl BytesAdapter);
#[cfg(test)]
mod test {
use proptest::prelude::*;
use super::super::test::{check_collection_type, check_type};
use super::*;
proptest! {
#[test]
fn check_vec(value: Vec<u8>, tag in MIN_TAG..=MAX_TAG) {
super::test::check_type::<Vec<u8>, Vec<u8>>(value, tag, WireType::LengthDelimited,
encode, merge, encoded_len)?;
}
#[test]
fn check_bytes(value: Vec<u8>, tag in MIN_TAG..=MAX_TAG) {
let value = Bytes::from(value);
super::test::check_type::<Bytes, Bytes>(value, tag, WireType::LengthDelimited,
encode, merge, encoded_len)?;
}
#[test]
fn check_repeated_vec(value: Vec<Vec<u8>>, tag in MIN_TAG..=MAX_TAG) {
super::test::check_collection_type(value, tag, WireType::LengthDelimited,
encode_repeated, merge_repeated,
encoded_len_repeated)?;
}
#[test]
fn check_repeated_bytes(value: Vec<Vec<u8>>, tag in MIN_TAG..=MAX_TAG) {
let value = value.into_iter().map(Bytes::from).collect();
super::test::check_collection_type(value, tag, WireType::LengthDelimited,
encode_repeated, merge_repeated,
encoded_len_repeated)?;
}
}
}
}
pub mod message {
use super::*;
pub fn encode<M, B>(tag: u32, msg: &M, buf: &mut B)
where
M: Message,
B: BufMut,
{
encode_key(tag, WireType::LengthDelimited, buf);
encode_varint(msg.encoded_len() as u64, buf);
msg.encode_raw(buf);
}
pub fn merge<M, B>(
wire_type: WireType,
msg: &mut M,
buf: &mut B,
ctx: DecodeContext,
) -> Result<(), DecodeError>
where
M: Message,
B: Buf,
{
check_wire_type(WireType::LengthDelimited, wire_type)?;
ctx.limit_reached()?;
merge_loop(
msg,
buf,
ctx.enter_recursion(),
|msg: &mut M, buf: &mut B, ctx| {
let (tag, wire_type) = decode_key(buf)?;
msg.merge_field(tag, wire_type, buf, ctx)
},
)
}
pub fn encode_repeated<M, B>(tag: u32, messages: &[M], buf: &mut B)
where
M: Message,
B: BufMut,
{
for msg in messages {
encode(tag, msg, buf);
}
}
pub fn merge_repeated<M, B>(
wire_type: WireType,
messages: &mut Vec<M>,
buf: &mut B,
ctx: DecodeContext,
) -> Result<(), DecodeError>
where
M: Message + Default,
B: Buf,
{
check_wire_type(WireType::LengthDelimited, wire_type)?;
let mut msg = M::default();
merge(WireType::LengthDelimited, &mut msg, buf, ctx)?;
messages.push(msg);
Ok(())
}
#[inline]
pub fn encoded_len<M>(tag: u32, msg: &M) -> usize
where
M: Message,
{
let len = msg.encoded_len();
key_len(tag) + encoded_len_varint(len as u64) + len
}
#[inline]
pub fn encoded_len_repeated<M>(tag: u32, messages: &[M]) -> usize
where
M: Message,
{
key_len(tag) * messages.len()
+ messages
.iter()
.map(Message::encoded_len)
.map(|len| len + encoded_len_varint(len as u64))
.sum::<usize>()
}
}
pub mod group {
use super::*;
pub fn encode<M, B>(tag: u32, msg: &M, buf: &mut B)
where
M: Message,
B: BufMut,
{
encode_key(tag, WireType::StartGroup, buf);
msg.encode_raw(buf);
encode_key(tag, WireType::EndGroup, buf);
}
pub fn merge<M, B>(
tag: u32,
wire_type: WireType,
msg: &mut M,
buf: &mut B,
ctx: DecodeContext,
) -> Result<(), DecodeError>
where
M: Message,
B: Buf,
{
check_wire_type(WireType::StartGroup, wire_type)?;
ctx.limit_reached()?;
loop {
let (field_tag, field_wire_type) = decode_key(buf)?;
if field_wire_type == WireType::EndGroup {
if field_tag != tag {
return Err(DecodeError::new("unexpected end group tag"));
}
return Ok(());
}
M::merge_field(msg, field_tag, field_wire_type, buf, ctx.enter_recursion())?;
}
}
pub fn encode_repeated<M, B>(tag: u32, messages: &[M], buf: &mut B)
where
M: Message,
B: BufMut,
{
for msg in messages {
encode(tag, msg, buf);
}
}
pub fn merge_repeated<M, B>(
tag: u32,
wire_type: WireType,
messages: &mut Vec<M>,
buf: &mut B,
ctx: DecodeContext,
) -> Result<(), DecodeError>
where
M: Message + Default,
B: Buf,
{
check_wire_type(WireType::StartGroup, wire_type)?;
let mut msg = M::default();
merge(tag, WireType::StartGroup, &mut msg, buf, ctx)?;
messages.push(msg);
Ok(())
}
#[inline]
pub fn encoded_len<M>(tag: u32, msg: &M) -> usize
where
M: Message,
{
2 * key_len(tag) + msg.encoded_len()
}
#[inline]
pub fn encoded_len_repeated<M>(tag: u32, messages: &[M]) -> usize
where
M: Message,
{
2 * key_len(tag) * messages.len() + messages.iter().map(Message::encoded_len).sum::<usize>()
}
}
/// Rust doesn't have a `Map` trait, so macros are currently the best way to be
/// generic over `HashMap` and `BTreeMap`.
macro_rules! map {
($map_ty:ident) => {
use crate::encoding::*;
use core::hash::Hash;
/// Generic protobuf map encode function.
pub fn encode<K, V, B, KE, KL, VE, VL>(
key_encode: KE,
key_encoded_len: KL,
val_encode: VE,
val_encoded_len: VL,
tag: u32,
values: &$map_ty<K, V>,
buf: &mut B,
) where
K: Default + Eq + Hash + Ord,
V: Default + PartialEq,
B: BufMut,
KE: Fn(u32, &K, &mut B),
KL: Fn(u32, &K) -> usize,
VE: Fn(u32, &V, &mut B),
VL: Fn(u32, &V) -> usize,
{
encode_with_default(
key_encode,
key_encoded_len,
val_encode,
val_encoded_len,
&V::default(),
tag,
values,
buf,
)
}
/// Generic protobuf map merge function.
pub fn merge<K, V, B, KM, VM>(
key_merge: KM,
val_merge: VM,
values: &mut $map_ty<K, V>,
buf: &mut B,
ctx: DecodeContext,
) -> Result<(), DecodeError>
where
K: Default + Eq + Hash + Ord,
V: Default,
B: Buf,
KM: Fn(WireType, &mut K, &mut B, DecodeContext) -> Result<(), DecodeError>,
VM: Fn(WireType, &mut V, &mut B, DecodeContext) -> Result<(), DecodeError>,
{
merge_with_default(key_merge, val_merge, V::default(), values, buf, ctx)
}
/// Generic protobuf map encode function.
pub fn encoded_len<K, V, KL, VL>(
key_encoded_len: KL,
val_encoded_len: VL,
tag: u32,
values: &$map_ty<K, V>,
) -> usize
where
K: Default + Eq + Hash + Ord,
V: Default + PartialEq,
KL: Fn(u32, &K) -> usize,
VL: Fn(u32, &V) -> usize,
{
encoded_len_with_default(key_encoded_len, val_encoded_len, &V::default(), tag, values)
}
/// Generic protobuf map encode function with an overridden value default.
///
/// This is necessary because enumeration values can have a default value other
/// than 0 in proto2.
pub fn encode_with_default<K, V, B, KE, KL, VE, VL>(
key_encode: KE,
key_encoded_len: KL,
val_encode: VE,
val_encoded_len: VL,
val_default: &V,
tag: u32,
values: &$map_ty<K, V>,
buf: &mut B,
) where
K: Default + Eq + Hash + Ord,
V: PartialEq,
B: BufMut,
KE: Fn(u32, &K, &mut B),
KL: Fn(u32, &K) -> usize,
VE: Fn(u32, &V, &mut B),
VL: Fn(u32, &V) -> usize,
{
for (key, val) in values.iter() {
let skip_key = key == &K::default();
let skip_val = val == val_default;
let len = (if skip_key { 0 } else { key_encoded_len(1, key) })
+ (if skip_val { 0 } else { val_encoded_len(2, val) });
encode_key(tag, WireType::LengthDelimited, buf);
encode_varint(len as u64, buf);
if !skip_key {
key_encode(1, key, buf);
}
if !skip_val {
val_encode(2, val, buf);
}
}
}
/// Generic protobuf map merge function with an overridden value default.
///
/// This is necessary because enumeration values can have a default value other
/// than 0 in proto2.
pub fn merge_with_default<K, V, B, KM, VM>(
key_merge: KM,
val_merge: VM,
val_default: V,
values: &mut $map_ty<K, V>,
buf: &mut B,
ctx: DecodeContext,
) -> Result<(), DecodeError>
where
K: Default + Eq + Hash + Ord,
B: Buf,
KM: Fn(WireType, &mut K, &mut B, DecodeContext) -> Result<(), DecodeError>,
VM: Fn(WireType, &mut V, &mut B, DecodeContext) -> Result<(), DecodeError>,
{
let mut key = Default::default();
let mut val = val_default;
ctx.limit_reached()?;
merge_loop(
&mut (&mut key, &mut val),
buf,
ctx.enter_recursion(),
|&mut (ref mut key, ref mut val), buf, ctx| {
let (tag, wire_type) = decode_key(buf)?;
match tag {
1 => key_merge(wire_type, key, buf, ctx),
2 => val_merge(wire_type, val, buf, ctx),
_ => skip_field(wire_type, tag, buf, ctx),
}
},
)?;
values.insert(key, val);
Ok(())
}
/// Generic protobuf map encode function with an overridden value default.
///
/// This is necessary because enumeration values can have a default value other
/// than 0 in proto2.
pub fn encoded_len_with_default<K, V, KL, VL>(
key_encoded_len: KL,
val_encoded_len: VL,
val_default: &V,
tag: u32,
values: &$map_ty<K, V>,
) -> usize
where
K: Default + Eq + Hash + Ord,
V: PartialEq,
KL: Fn(u32, &K) -> usize,
VL: Fn(u32, &V) -> usize,
{
key_len(tag) * values.len()
+ values
.iter()
.map(|(key, val)| {
let len = (if key == &K::default() {
0
} else {
key_encoded_len(1, key)
}) + (if val == val_default {
0
} else {
val_encoded_len(2, val)
});
encoded_len_varint(len as u64) + len
})
.sum::<usize>()
}
};
}
#[cfg(feature = "std")]
pub mod hash_map {
use std::collections::HashMap;
map!(HashMap);
}
pub mod btree_map {
map!(BTreeMap);
}
#[cfg(test)]
mod test {
use alloc::string::ToString;
use core::borrow::Borrow;
use core::fmt::Debug;
use core::u64;
use ::bytes::{Bytes, BytesMut};
use proptest::{prelude::*, test_runner::TestCaseResult};
use crate::encoding::*;
pub fn check_type<T, B>(
value: T,
tag: u32,
wire_type: WireType,
encode: fn(u32, &B, &mut BytesMut),
merge: fn(WireType, &mut T, &mut Bytes, DecodeContext) -> Result<(), DecodeError>,
encoded_len: fn(u32, &B) -> usize,
) -> TestCaseResult
where
T: Debug + Default + PartialEq + Borrow<B>,
B: ?Sized,
{
prop_assume!((MIN_TAG..=MAX_TAG).contains(&tag));
let expected_len = encoded_len(tag, value.borrow());
let mut buf = BytesMut::with_capacity(expected_len);
encode(tag, value.borrow(), &mut buf);
let mut buf = buf.freeze();
prop_assert_eq!(
buf.remaining(),
expected_len,
"encoded_len wrong; expected: {}, actual: {}",
expected_len,
buf.remaining()
);
if !buf.has_remaining() {
// Short circuit for empty packed values.
return Ok(());
}
let (decoded_tag, decoded_wire_type) =
decode_key(&mut buf).map_err(|error| TestCaseError::fail(error.to_string()))?;
prop_assert_eq!(
tag,
decoded_tag,
"decoded tag does not match; expected: {}, actual: {}",
tag,
decoded_tag
);
prop_assert_eq!(
wire_type,
decoded_wire_type,
"decoded wire type does not match; expected: {:?}, actual: {:?}",
wire_type,
decoded_wire_type,
);
match wire_type {
WireType::SixtyFourBit if buf.remaining() != 8 => Err(TestCaseError::fail(format!(
"64bit wire type illegal remaining: {}, tag: {}",
buf.remaining(),
tag
))),
WireType::ThirtyTwoBit if buf.remaining() != 4 => Err(TestCaseError::fail(format!(
"32bit wire type illegal remaining: {}, tag: {}",
buf.remaining(),
tag
))),
_ => Ok(()),
}?;
let mut roundtrip_value = T::default();
merge(
wire_type,
&mut roundtrip_value,
&mut buf,
DecodeContext::default(),
)
.map_err(|error| TestCaseError::fail(error.to_string()))?;
prop_assert!(
!buf.has_remaining(),
"expected buffer to be empty, remaining: {}",
buf.remaining()
);
prop_assert_eq!(value, roundtrip_value);
Ok(())
}
pub fn check_collection_type<T, B, E, M, L>(
value: T,
tag: u32,
wire_type: WireType,
encode: E,
mut merge: M,
encoded_len: L,
) -> TestCaseResult
where
T: Debug + Default + PartialEq + Borrow<B>,
B: ?Sized,
E: FnOnce(u32, &B, &mut BytesMut),
M: FnMut(WireType, &mut T, &mut Bytes, DecodeContext) -> Result<(), DecodeError>,
L: FnOnce(u32, &B) -> usize,
{
prop_assume!((MIN_TAG..=MAX_TAG).contains(&tag));
let expected_len = encoded_len(tag, value.borrow());
let mut buf = BytesMut::with_capacity(expected_len);
encode(tag, value.borrow(), &mut buf);
let mut buf = buf.freeze();
prop_assert_eq!(
buf.remaining(),
expected_len,
"encoded_len wrong; expected: {}, actual: {}",
expected_len,
buf.remaining()
);
let mut roundtrip_value = Default::default();
while buf.has_remaining() {
let (decoded_tag, decoded_wire_type) =
decode_key(&mut buf).map_err(|error| TestCaseError::fail(error.to_string()))?;
prop_assert_eq!(
tag,
decoded_tag,
"decoded tag does not match; expected: {}, actual: {}",
tag,
decoded_tag
);
prop_assert_eq!(
wire_type,
decoded_wire_type,
"decoded wire type does not match; expected: {:?}, actual: {:?}",
wire_type,
decoded_wire_type
);
merge(
wire_type,
&mut roundtrip_value,
&mut buf,
DecodeContext::default(),
)
.map_err(|error| TestCaseError::fail(error.to_string()))?;
}
prop_assert_eq!(value, roundtrip_value);
Ok(())
}
#[test]
fn string_merge_invalid_utf8() {
let mut s = String::new();
let buf = b"\x02\x80\x80";
let r = string::merge(
WireType::LengthDelimited,
&mut s,
&mut &buf[..],
DecodeContext::default(),
);
r.expect_err("must be an error");
assert!(s.is_empty());
}
#[test]
fn varint() {
fn check(value: u64, mut encoded: &[u8]) {
// Small buffer.
let mut buf = Vec::with_capacity(1);
encode_varint(value, &mut buf);
assert_eq!(buf, encoded);
// Large buffer.
let mut buf = Vec::with_capacity(100);
encode_varint(value, &mut buf);
assert_eq!(buf, encoded);
assert_eq!(encoded_len_varint(value), encoded.len());
let roundtrip_value =
decode_varint(&mut <&[u8]>::clone(&encoded)).expect("decoding failed");
assert_eq!(value, roundtrip_value);
let roundtrip_value = decode_varint_slow(&mut encoded).expect("slow decoding failed");
assert_eq!(value, roundtrip_value);
}
check(2u64.pow(0) - 1, &[0x00]);
check(2u64.pow(0), &[0x01]);
check(2u64.pow(7) - 1, &[0x7F]);
check(2u64.pow(7), &[0x80, 0x01]);
check(300, &[0xAC, 0x02]);
check(2u64.pow(14) - 1, &[0xFF, 0x7F]);
check(2u64.pow(14), &[0x80, 0x80, 0x01]);
check(2u64.pow(21) - 1, &[0xFF, 0xFF, 0x7F]);
check(2u64.pow(21), &[0x80, 0x80, 0x80, 0x01]);
check(2u64.pow(28) - 1, &[0xFF, 0xFF, 0xFF, 0x7F]);
check(2u64.pow(28), &[0x80, 0x80, 0x80, 0x80, 0x01]);
check(2u64.pow(35) - 1, &[0xFF, 0xFF, 0xFF, 0xFF, 0x7F]);
check(2u64.pow(35), &[0x80, 0x80, 0x80, 0x80, 0x80, 0x01]);
check(2u64.pow(42) - 1, &[0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x7F]);
check(2u64.pow(42), &[0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x01]);
check(
2u64.pow(49) - 1,
&[0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x7F],
);
check(
2u64.pow(49),
&[0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x01],
);
check(
2u64.pow(56) - 1,
&[0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x7F],
);
check(
2u64.pow(56),
&[0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x01],
);
check(
2u64.pow(63) - 1,
&[0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x7F],
);
check(
2u64.pow(63),
&[0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x01],
);
check(
u64::MAX,
&[0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x01],
);
}
#[test]
fn varint_overflow() {
let mut u64_max_plus_one: &[u8] =
&[0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x02];
decode_varint(&mut u64_max_plus_one).expect_err("decoding u64::MAX + 1 succeeded");
decode_varint_slow(&mut u64_max_plus_one)
.expect_err("slow decoding u64::MAX + 1 succeeded");
}
/// This big bowl o' macro soup generates an encoding property test for each combination of map
/// type, scalar map key, and value type.
/// TODO: these tests take a long time to compile, can this be improved?
#[cfg(feature = "std")]
macro_rules! map_tests {
(keys: $keys:tt,
vals: $vals:tt) => {
mod hash_map {
map_tests!(@private HashMap, hash_map, $keys, $vals);
}
mod btree_map {
map_tests!(@private BTreeMap, btree_map, $keys, $vals);
}
};
(@private $map_type:ident,
$mod_name:ident,
[$(($key_ty:ty, $key_proto:ident)),*],
$vals:tt) => {
$(
mod $key_proto {
use std::collections::$map_type;
use proptest::prelude::*;
use crate::encoding::*;
use crate::encoding::test::check_collection_type;
map_tests!(@private $map_type, $mod_name, ($key_ty, $key_proto), $vals);
}
)*
};
(@private $map_type:ident,
$mod_name:ident,
($key_ty:ty, $key_proto:ident),
[$(($val_ty:ty, $val_proto:ident)),*]) => {
$(
proptest! {
#[test]
fn $val_proto(values: $map_type<$key_ty, $val_ty>, tag in MIN_TAG..=MAX_TAG) {
check_collection_type(values, tag, WireType::LengthDelimited,
|tag, values, buf| {
$mod_name::encode($key_proto::encode,
$key_proto::encoded_len,
$val_proto::encode,
$val_proto::encoded_len,
tag,
values,
buf)
},
|wire_type, values, buf, ctx| {
check_wire_type(WireType::LengthDelimited, wire_type)?;
$mod_name::merge($key_proto::merge,
$val_proto::merge,
values,
buf,
ctx)
},
|tag, values| {
$mod_name::encoded_len($key_proto::encoded_len,
$val_proto::encoded_len,
tag,
values)
})?;
}
}
)*
};
}
#[cfg(feature = "std")]
map_tests!(keys: [
(i32, int32),
(i64, int64),
(u32, uint32),
(u64, uint64),
(i32, sint32),
(i64, sint64),
(u32, fixed32),
(u64, fixed64),
(i32, sfixed32),
(i64, sfixed64),
(bool, bool),
(String, string)
],
vals: [
(f32, float),
(f64, double),
(i32, int32),
(i64, int64),
(u32, uint32),
(u64, uint64),
(i32, sint32),
(i64, sint64),
(u32, fixed32),
(u64, fixed64),
(i32, sfixed32),
(i64, sfixed64),
(bool, bool),
(String, string),
(Vec<u8>, bytes)
]);
}