Skip to content

Minor code simplifications #107

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
oschwald merged 7 commits into from
Feb 14, 2026
Merged

Minor code simplifications #107

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
9204487
Use shared separator constant in Reader
oschwald Feb 11, 2026
1309f39
Deduplicate fixed-width integer decoding
oschwald Feb 11, 2026
46580ac
Unify array path navigation branches
oschwald Feb 11, 2026
2e0459b
Extract child-node push helper
oschwald Feb 11, 2026
d8f6abf
Collapse repeated is_empty implementations
oschwald Feb 11, 2026
a7b47dc
Factor reader test setup helpers
oschwald Feb 11, 2026
0659747
Simplify pointer traversal and keep fast paths
oschwald Feb 11, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
196 changes: 94 additions & 102 deletions src/decoder.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,6 +40,30 @@ fn to_usize(base: u8, bytes: &[u8]) -> usize {
.fold(base as usize, |acc, &b| (acc << 8) | b as usize)
}

macro_rules! decode_int_like {
($name:ident, $ty:ty, $max_size:expr, $label:literal, $zero:expr) => {
fn $name(&mut self, size: usize) -> DecodeResult<$ty> {
match size {
s if s <= $max_size => {
let new_offset = self
.current_ptr
.checked_add(size)
.filter(|&offset| offset <= self.buf.len())
.ok_or_else(|| {
self.invalid_db_error(&format!("{} of size {}", $label, size))
})?;
let value = self.buf[self.current_ptr..new_offset]
.iter()
.fold($zero, |acc, &b| (acc << 8) | <$ty>::from(b));
self.current_ptr = new_offset;
Ok(value)
}
s => Err(self.invalid_db_error(&format!("{} of size {}", $label, s))),
}
}
};
}
Comment thread
coderabbitai[bot] marked this conversation as resolved.

enum Value<'a, 'de> {
Any { prev_ptr: usize },
Bytes(&'de [u8]),
Expand Down Expand Up @@ -262,78 +286,26 @@ impl<'de> Decoder<'de> {
Ok(float_value)
}

fn decode_uint64(&mut self, size: usize) -> DecodeResult<u64> {
match size {
s if s <= 8 => {
let new_offset = self.current_ptr + size;

let value = self.buf[self.current_ptr..new_offset]
.iter()
.fold(0_u64, |acc, &b| (acc << 8) | u64::from(b));
self.current_ptr = new_offset;
Ok(value)
}
s => Err(self.invalid_db_error(&format!("u64 of size {s}"))),
}
}

fn decode_uint128(&mut self, size: usize) -> DecodeResult<u128> {
match size {
s if s <= 16 => {
let new_offset = self.current_ptr + size;

let value = self.buf[self.current_ptr..new_offset]
.iter()
.fold(0_u128, |acc, &b| (acc << 8) | u128::from(b));
self.current_ptr = new_offset;
Ok(value)
}
s => Err(self.invalid_db_error(&format!("u128 of size {s}"))),
}
}

fn decode_uint32(&mut self, size: usize) -> DecodeResult<u32> {
match size {
s if s <= 4 => {
let new_offset = self.current_ptr + size;

let value = self.buf[self.current_ptr..new_offset]
.iter()
.fold(0_u32, |acc, &b| (acc << 8) | u32::from(b));
self.current_ptr = new_offset;
Ok(value)
}
s => Err(self.invalid_db_error(&format!("u32 of size {s}"))),
}
}

fn decode_uint16(&mut self, size: usize) -> DecodeResult<u16> {
match size {
s if s <= 2 => {
let new_offset = self.current_ptr + size;

let value = self.buf[self.current_ptr..new_offset]
.iter()
.fold(0_u16, |acc, &b| (acc << 8) | u16::from(b));
self.current_ptr = new_offset;
Ok(value)
}
s => Err(self.invalid_db_error(&format!("u16 of size {s}"))),
}
}
decode_int_like!(decode_uint64, u64, 8, "u64", 0_u64);
decode_int_like!(decode_uint128, u128, 16, "u128", 0_u128);
decode_int_like!(decode_uint32, u32, 4, "u32", 0_u32);
decode_int_like!(decode_uint16, u16, 2, "u16", 0_u16);

fn decode_int(&mut self, size: usize) -> DecodeResult<i32> {
match size {
s if s <= 4 => {
let new_offset = self.current_ptr + size;

let new_offset = self
.current_ptr
.checked_add(size)
.filter(|&offset| offset <= self.buf.len())
.ok_or_else(|| self.invalid_db_error(&format!("i32 of size {}", size)))?;
let value = self.buf[self.current_ptr..new_offset]
.iter()
.fold(0_i32, |acc, &b| (acc << 8) | i32::from(b));
.fold(0_u32, |acc, &b| (acc << 8) | u32::from(b));
self.current_ptr = new_offset;
Ok(value)
Ok(value as i32)
}
s => Err(self.invalid_db_error(&format!("i32 of size {s}"))),
s => Err(self.invalid_db_error(&format!("i32 of size {}", s))),
}
}

Expand Down Expand Up @@ -407,67 +379,87 @@ impl<'de> Decoder<'de> {

/// Consumes a map header, returning its size. Follows pointers.
pub(crate) fn consume_map_header(&mut self) -> DecodeResult<usize> {
let (size, type_num) = self.size_and_type();
if type_num == TYPE_POINTER {
let new_ptr = self.decode_pointer(size);
self.current_ptr = new_ptr;
self.consume_map_header()
} else if type_num == TYPE_MAP {
Ok(size)
} else {
Err(self.decode_error(&format!("expected map, got type {type_num}")))
}
self.consume_typed_header(TYPE_MAP, "map")
}

/// Consumes an array header, returning its size. Follows pointers.
pub(crate) fn consume_array_header(&mut self) -> DecodeResult<usize> {
self.consume_typed_header(TYPE_ARRAY, "array")
}

/// Consumes a header of the expected type, following one pointer.
fn consume_typed_header(&mut self, expected_type: u8, label: &str) -> DecodeResult<usize> {
let (size, type_num) = self.size_and_type();
if type_num == TYPE_POINTER {
let new_ptr = self.decode_pointer(size);
self.current_ptr = new_ptr;
self.consume_array_header()
} else if type_num == TYPE_ARRAY {
self.current_ptr = self.decode_pointer(size);
let (size, type_num) = self.size_and_type();
if type_num == TYPE_POINTER {
return Err(self.invalid_db_error("pointer points to another pointer"));
}
if type_num == expected_type {
return Ok(size);
}
return Err(self.decode_error(&format!("expected {label}, got type {type_num}")));
}
if type_num == expected_type {
Ok(size)
} else {
Err(self.decode_error(&format!("expected array, got type {type_num}")))
Err(self.decode_error(&format!("expected {label}, got type {type_num}")))
}
}
Comment thread
coderabbitai[bot] marked this conversation as resolved.

/// Gets size and type, following any pointers.
fn size_and_type_following_pointers(&mut self) -> DecodeResult<(usize, u8)> {
let (size, type_num) = self.size_and_type();
if type_num != TYPE_POINTER {
return Ok((size, type_num));
}

self.current_ptr = self.decode_pointer(size);
let (size, type_num) = self.size_and_type();
if type_num == TYPE_POINTER {
// Pointer - follow it
let new_ptr = self.decode_pointer(size);
self.current_ptr = new_ptr;
self.size_and_type_following_pointers()
} else {
Ok((size, type_num))
return Err(self.invalid_db_error("pointer points to another pointer"));
}

Ok((size, type_num))
}

#[inline(always)]
fn read_string_bytes(&mut self, size: usize) -> DecodeResult<&'de [u8]> {
let new_offset = self
.current_ptr
.checked_add(size)
.ok_or_else(|| self.invalid_db_error("string length exceeds buffer"))?;
if new_offset > self.buf.len() {
return Err(self.invalid_db_error("string length exceeds buffer"));
}
let bytes = &self.buf[self.current_ptr..new_offset];
self.current_ptr = new_offset;
Ok(bytes)
}
Comment thread
coderabbitai[bot] marked this conversation as resolved.

/// Reads a string's bytes directly, following pointers if needed.
/// Does NOT validate UTF-8.
pub(crate) fn read_str_as_bytes(&mut self) -> DecodeResult<&'de [u8]> {
let (size, type_num) = self.size_and_type();
if type_num == TYPE_POINTER {
// Pointer
let new_ptr = self.decode_pointer(size);
let saved_ptr = self.current_ptr;
self.current_ptr = new_ptr;
let result = self.read_str_as_bytes();
self.current_ptr = saved_ptr;
result
} else if type_num == TYPE_STRING {
let new_offset = self.current_ptr + size;
if new_offset > self.buf.len() {
return Err(self.invalid_db_error("string length exceeds buffer"));
match type_num {
TYPE_POINTER => {
let new_ptr = self.decode_pointer(size);
let saved_ptr = self.current_ptr;
self.current_ptr = new_ptr;
let (size, type_num) = self.size_and_type();
let result = if type_num == TYPE_POINTER {
Err(self.invalid_db_error("pointer points to another pointer"))
} else if type_num == TYPE_STRING {
self.read_string_bytes(size)
} else {
Err(self.invalid_db_error(&format!("expected string, got type {type_num}")))
};
self.current_ptr = saved_ptr;
result
}
let bytes = &self.buf[self.current_ptr..new_offset];
self.current_ptr = new_offset;
Ok(bytes)
} else {
Err(self.invalid_db_error(&format!("expected string, got type {type_num}")))
TYPE_STRING => self.read_string_bytes(size),
_ => Err(self.invalid_db_error(&format!("expected string, got type {type_num}"))),
}
}

Expand Down
Loading