Skip to content

Fix inflate() flush modes and implement state.data_type #113

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.

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
merged 4 commits into from
Jun 13, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
112 changes: 112 additions & 0 deletions libz-rs-sys/src/tests/inflate.rs
Original file line number Diff line number Diff line change
Expand Up @@ -1698,3 +1698,115 @@ fn op_len_edge_case() {

assert_eq!(output_rs, output_ng);
}

// Fills the provided buffer with pseudorandom bytes based on the given seed
// Duplicates bytes by `step` in a row
fn prng_bytes(seed: u64, bytes: &mut [u8], step: usize) {
const M: u64 = 2u64.pow(32);
const A: u64 = 1664525;
const C: u64 = 1013904223;
let mut state = seed;
for chunk in bytes.chunks_mut(4 * step) {
state = (A * state + C) % M;
let rand_bytes = state.to_le_bytes();
for (i, byte) in chunk.iter_mut().enumerate() {
*byte = rand_bytes[i / step];
}
}
}

#[test]
fn test_inflate_flush_block() {
let window_bits = -15; // Raw
const CHUNK: usize = 16384;

// Create a compressed vector of random data that's bigger then the zlib block size
let mut data = vec![0u8; 160000];
prng_bytes(314159, &mut data, 4);
let config = DeflateConfig {
window_bits,
..DeflateConfig::default()
};
let mut output = vec![0u8; 80000];
// Compress the data
let (compressed_data, return_code) = compress_slice(&mut output, &data, config);
assert_eq!(return_code, ReturnCode::Ok);

// Log the stream positions and data_type output from libz
let mut zlib_log = Vec::new();
{
let mut stream = MaybeUninit::<libz_sys::z_stream>::zeroed();

let ret = unsafe {
libz_sys::inflateInit2_(
stream.as_mut_ptr(),
window_bits,
libz_sys::zlibVersion(),
core::mem::size_of::<libz_sys::z_stream>() as c_int,
)
};
assert_eq!(ReturnCode::from(ret), ReturnCode::Ok);

let mut output = vec![0u8; CHUNK * 2];
let stream = unsafe { stream.assume_init_mut() };
stream.next_in = compressed_data.as_ptr() as *mut u8;
stream.avail_in = compressed_data.len() as _;
loop {
stream.next_out = output.as_mut_ptr();
stream.avail_out = output.len() as _;

let ret = unsafe { libz_sys::inflate(stream, InflateFlush::Block as i32) };

let log = format!(
"In:{} Out:{} DT:{}",
stream.avail_in, stream.avail_out, stream.data_type
);
zlib_log.push(log);

assert_eq!(ReturnCode::from(ret), ReturnCode::Ok);

if stream.avail_in == 0 {
break;
}
}
}

// Log the stream positions and data_type output from libz_rs and compare
{
let mut stream = MaybeUninit::<z_stream>::zeroed();

let ret = unsafe {
inflateInit2_(
stream.as_mut_ptr(),
window_bits,
zlibVersion(),
core::mem::size_of::<z_stream>() as c_int,
)
};
assert_eq!(ReturnCode::from(ret), ReturnCode::Ok);

let mut output = vec![0u8; CHUNK * 2];
let stream = unsafe { stream.assume_init_mut() };
stream.next_in = compressed_data.as_ptr() as *mut u8;
stream.avail_in = compressed_data.len() as _;
loop {
stream.next_out = output.as_mut_ptr();
stream.avail_out = output.len() as _;

let ret = unsafe { inflate(stream, InflateFlush::Block as i32) };

let log = format!(
"In:{} Out:{} DT:{}",
stream.avail_in, stream.avail_out, stream.data_type
);
// Compare log entries
assert_eq!(zlib_log.remove(0), log);

assert_eq!(ReturnCode::from(ret), ReturnCode::Ok);

if stream.avail_in == 0 {
break;
}
}
}
}
63 changes: 48 additions & 15 deletions zlib-rs/src/inflate.rs
Original file line number Diff line number Diff line change
Expand Up @@ -252,6 +252,7 @@ pub enum Mode {
Stored,
CopyBlock,
Check,
Len_,
Len,
Lit,
LenExt,
Expand Down Expand Up @@ -505,6 +506,7 @@ impl<'a> State<'a> {
Mode::CopyBlock => self.copy_block(),
Mode::Check => self.check(),
Mode::Len => self.len(),
Mode::Len_ => self.len_(),
Mode::LenExt => self.len_ext(),
Mode::Lit => self.lit(),
Mode::Dist => self.dist(),
Expand Down Expand Up @@ -965,13 +967,14 @@ impl<'a> State<'a> {
bits: 5,
};

self.mode = Mode::Len_;

self.bit_reader.drop_bits(2);

if let InflateFlush::Trees = self.flush {
self.inflate_leave(ReturnCode::Ok)
} else {
self.mode = Mode::Len;
self.len()
self.len_()
}
}
2 => {
Expand Down Expand Up @@ -1046,6 +1049,11 @@ impl<'a> State<'a> {
self.type_()
}

fn len_(&mut self) -> ReturnCode {
self.mode = Mode::Len;
self.len()
}

fn len(&mut self) -> ReturnCode {
let avail_in = self.bit_reader.bytes_remaining();
let avail_out = self.writer.remaining();
Expand Down Expand Up @@ -1463,13 +1471,13 @@ impl<'a> State<'a> {
self.dist_table.bits = root;
self.dist_table.codes = Codes::Dist;

self.mode = Mode::Len;
self.mode = Mode::Len_;

if matches!(self.flush, InflateFlush::Trees) {
return self.inflate_leave(ReturnCode::Ok);
}

self.len()
self.len_()
}

fn dict_id(&mut self) -> ReturnCode {
Expand Down Expand Up @@ -1499,7 +1507,7 @@ impl<'a> State<'a> {
}

fn bad(&mut self, msg: &'static str) -> ReturnCode {
#[cfg(feature = "std")]
#[cfg(all(feature = "std", test))]
dbg!(msg);
self.error_message = Some(msg);
self.inflate_leave(ReturnCode::DataError)
Expand All @@ -1511,6 +1519,22 @@ impl<'a> State<'a> {
// actual logic is in `inflate` itself
return_code
}

/// Stored in the `z_stream.data_type` field
fn decoding_state(&self) -> i32 {
let bit_reader_bits = self.bit_reader.bits_in_buffer() as i32;
debug_assert!(bit_reader_bits < 64);

let last = if self.last { 64 } else { 0 };

let mode = match self.mode {
Mode::Type => 128,
Mode::Len_ | Mode::CopyBlock => 256,
_ => 0,
};

bit_reader_bits | last | mode
}
}

fn inflate_fast_help(state: &mut State, _start: usize) -> ReturnCode {
Expand Down Expand Up @@ -1881,6 +1905,13 @@ pub unsafe fn inflate(stream: &mut InflateStream, flush: InflateFlush) -> Return

let state = &mut stream.state;

// skip check
if let Mode::Type = state.mode {
state.mode = Mode::TypeDo;
}

state.flush = flush;

state.bit_reader.update_slice(source_slice);
state.writer = ReadBuf::new(dest_slice);

Expand Down Expand Up @@ -1922,32 +1953,34 @@ pub unsafe fn inflate(stream: &mut InflateStream, flush: InflateFlush) -> Return
if must_update_window {
'blk: {
// initialize the window if needed
if stream.state.window.size() == 0 {
match Window::new_in(&stream.alloc, stream.state.wbits) {
Some(window) => stream.state.window = window,
if state.window.size() == 0 {
match Window::new_in(&stream.alloc, state.wbits) {
Some(window) => state.window = window,
None => {
stream.state.mode = Mode::Mem;
state.mode = Mode::Mem;
err = ReturnCode::MemError;
break 'blk;
}
}
}

stream.state.window.extend(
&stream.state.writer.filled()[..out_written],
stream.state.flags,
state.window.extend(
&state.writer.filled()[..out_written],
state.flags,
update_checksum,
&mut stream.state.checksum,
&mut stream.state.crc_fold,
&mut state.checksum,
&mut state.crc_fold,
);
}
}

if let Some(msg) = stream.state.error_message {
if let Some(msg) = state.error_message {
assert!(msg.ends_with(|c| c == '\0'));
stream.msg = msg.as_ptr() as *mut u8 as *mut core::ffi::c_char;
}

stream.data_type = state.decoding_state();

if ((in_read == 0 && out_written == 0) || flush == InflateFlush::Finish as _)
&& err == (ReturnCode::Ok as _)
{
Expand Down
Loading