Skip to content
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

change implementation of hash to use vtable #8

Merged
merged 2 commits into from
Jan 5, 2024
Merged

change implementation of hash to use vtable #8

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9cb72b5
change implementation of hash to use vtable
mcroomp Jan 4, 2024
e51c898
reformat
mcroomp Jan 5, 2024
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
157 changes: 81 additions & 76 deletions src/complevel_estimator.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,16 +7,18 @@
/// This module is design to detect the appropriate overall parameters for the preflate compressor.
/// Getting the parameters correct means that the resulting diff between the deflate stream
/// and the predicted deflate stream will be as small as possible.
use crate::hash_algorithm::{
HashAlgorithm, LibdeflateRotatingHash4, MiniZHash, ZlibNGHash, ZlibRotatingHash,
MINIZ_LEVEL1_HASH_SIZE_MASK,
use crate::{
hash_algorithm::{
HashAlgorithm, LibdeflateRotatingHash4, MiniZHash, RotatingHashTrait, ZlibNGHash,
ZlibRotatingHash, MINIZ_LEVEL1_HASH_SIZE_MASK,
},
hash_chain::{DictionaryAddPolicy, HashChain, MAX_UPDATE_HASH_BATCH},
preflate_constants,
preflate_input::PreflateInput,
preflate_parse_config::{FAST_PREFLATE_PARSER_SETTINGS, SLOW_PREFLATE_PARSER_SETTINGS},
preflate_token::{BlockType, PreflateToken, PreflateTokenBlock, PreflateTokenReference},
skip_length_estimator::estimate_skip_length,
};
use crate::hash_chain::{DictionaryAddPolicy, HashChain, MAX_UPDATE_HASH_BATCH};
use crate::preflate_constants;
use crate::preflate_input::PreflateInput;
use crate::preflate_parse_config::{FAST_PREFLATE_PARSER_SETTINGS, SLOW_PREFLATE_PARSER_SETTINGS};
use crate::preflate_token::{BlockType, PreflateToken, PreflateTokenBlock, PreflateTokenReference};
use crate::skip_length_estimator::estimate_skip_length;

#[derive(Default)]
pub struct CompLevelInfo {
Expand All @@ -37,18 +39,64 @@ pub struct CompLevelInfo {
pub max_chain: u32,
}

enum HashChainType {
Zlib(HashChain<ZlibRotatingHash>),
MiniZ(HashChain<MiniZHash>),
LibFlate4(HashChain<LibdeflateRotatingHash4>),
ZlibNG(HashChain<ZlibNGHash>),
/// vtable for invoking the hash chain functions on specific implementation
/// of hash algorithm
trait HashChainInvoke {
fn invoke_update_hash(
&mut self,
len: u32,
input: &PreflateInput,
add_policy: DictionaryAddPolicy,
);

fn invoke_match_depth(
&mut self,
token: PreflateTokenReference,
window_size: u32,
input: &PreflateInput,
) -> u32;
}

/// holds the hashchain for a specific hash algorithm
struct HashChainHolder<H: RotatingHashTrait> {
hash_chain: HashChain<H>,
}

impl<H: RotatingHashTrait + 'static> HashChainHolder<H> {
fn new(hash_shift: u32, hash_mask: u16, input: &PreflateInput<'_>) -> Box<dyn HashChainInvoke> {
Box::new(HashChainHolder::<H> {
hash_chain: HashChain::<H>::new(hash_shift, hash_mask, input),
})
}
}

impl<H: RotatingHashTrait> HashChainInvoke for HashChainHolder<H> {
fn invoke_update_hash(
&mut self,
len: u32,
input: &PreflateInput,
add_policy: DictionaryAddPolicy,
) {
self.hash_chain
.update_hash_with_policy::<true>(len, input, add_policy)
}

fn invoke_match_depth(
&mut self,
token: PreflateTokenReference,
window_size: u32,
input: &PreflateInput,
) -> u32 {
self.hash_chain.match_depth(&token, window_size, input)
}
}

struct CandidateInfo {
hash_algorithm: HashAlgorithm,
hash_mask: u16,
hash_shift: u32,
add_policy: DictionaryAddPolicy,
hash_chain: HashChainType,
hash_chain: Box<dyn HashChainInvoke>,

longest_dist_at_hop_0: u32,
longest_dist_at_hop_1_plus: u32,
Expand All @@ -67,71 +115,36 @@ impl CandidateInfo {
hash_mask,
hash_shift,
add_policy,
hash_algorithm,
hash_chain: match hash_algorithm {
HashAlgorithm::Zlib => HashChainType::Zlib(HashChain::<ZlibRotatingHash>::new(
hash_shift, hash_mask, &input,
)),
HashAlgorithm::Zlib => {
HashChainHolder::<ZlibRotatingHash>::new(hash_shift, hash_mask, input)
}
HashAlgorithm::MiniZFast => {
HashChainType::MiniZ(HashChain::<MiniZHash>::new(hash_shift, hash_mask, &input))
HashChainHolder::<MiniZHash>::new(hash_shift, hash_mask, input)
}
HashAlgorithm::Libdeflate4 => {
HashChainType::LibFlate4(HashChain::<LibdeflateRotatingHash4>::new(
hash_shift, hash_mask, &input,
))
HashChainHolder::<LibdeflateRotatingHash4>::new(hash_shift, hash_mask, input)
}
HashAlgorithm::ZlibNG => {
HashChainHolder::<ZlibNGHash>::new(hash_shift, hash_mask, input)
}
HashAlgorithm::ZlibNG => HashChainType::ZlibNG(HashChain::<ZlibNGHash>::new(
hash_shift, hash_mask, &input,
)),
},
longest_dist_at_hop_0: 0,
longest_dist_at_hop_1_plus: 0,
max_chain_found: 0,
}
}

fn invoke_update_hash(
&mut self,
len: u32,
input: &PreflateInput,
add_policy: DictionaryAddPolicy,
) {
match self.hash_chain {
HashChainType::Zlib(ref mut h) => {
h.update_hash_with_policy::<true>(len, input, add_policy)
}
HashChainType::MiniZ(ref mut h) => {
h.update_hash_with_policy::<true>(len, input, add_policy)
}
HashChainType::LibFlate4(ref mut h) => {
h.update_hash_with_policy::<true>(len, input, add_policy)
}
HashChainType::ZlibNG(ref mut h) => {
h.update_hash_with_policy::<true>(len, input, add_policy)
}
}
}

fn invoke_match_depth(
&mut self,
token: &PreflateTokenReference,
window_size: u32,
input: &PreflateInput,
) -> u32 {
match self.hash_chain {
HashChainType::Zlib(ref mut h) => h.match_depth(token, window_size, input),
HashChainType::MiniZ(ref mut h) => h.match_depth(token, window_size, input),
HashChainType::LibFlate4(ref mut h) => h.match_depth(token, window_size, input),
HashChainType::ZlibNG(ref mut h) => h.match_depth(token, window_size, input),
}
}

fn match_depth(
&mut self,
token: &PreflateTokenReference,
token: PreflateTokenReference,
window_size: u32,
input: &PreflateInput,
) -> bool {
let mdepth = self.invoke_match_depth(token, window_size, input);
let mdepth = self
.hash_chain
.invoke_match_depth(token, window_size, input);

// remove element if the match was impossible due to matching the
// the hash depth or because in fast mode we can't match partial words
Expand Down Expand Up @@ -183,22 +196,14 @@ impl CandidateInfo {
}

fn hash_algorithm(&self) -> HashAlgorithm {
match self.hash_chain {
HashChainType::Zlib(_) => HashAlgorithm::Zlib,
HashChainType::MiniZ(_) => HashAlgorithm::MiniZFast,
HashChainType::LibFlate4(_) => HashAlgorithm::Libdeflate4,
HashChainType::ZlibNG(_) => HashAlgorithm::ZlibNG,
}
self.hash_algorithm
}
}

struct CompLevelEstimatorState<'a> {
input: PreflateInput<'a>,

// fast compressor candidates, depending on the hash shift and mask
// and what length of matches we should skip adding to the hash table.
// As we look at the data, we remove candidates that have impossible
// matches, and at the end we pick the best candidate.
/// candidates for checking for which hash algorithm to use
candidates: Vec<Box<CandidateInfo>>,

blocks: &'a Vec<PreflateTokenBlock>,
Expand Down Expand Up @@ -291,7 +296,7 @@ impl<'a> CompLevelEstimatorState<'a> {
let batch_len = std::cmp::min(length, MAX_UPDATE_HASH_BATCH);

for i in &mut self.candidates {
i.invoke_update_hash(
i.hash_chain.invoke_update_hash(
batch_len,
&self.input,
if override_add_policy {
Expand All @@ -307,7 +312,7 @@ impl<'a> CompLevelEstimatorState<'a> {
}
}

fn check_match(&mut self, token: &PreflateTokenReference) {
fn check_match(&mut self, token: PreflateTokenReference) {
self.reference_count += 1;

if self.input.pos() < token.dist() || self.candidates.is_empty() {
Expand Down Expand Up @@ -342,7 +347,7 @@ impl<'a> CompLevelEstimatorState<'a> {
PreflateToken::Literal => {
self.update_hash(1, true);
}
PreflateToken::Reference(r) => {
&PreflateToken::Reference(r) => {
self.check_match(r);
self.update_hash(r.len(), false);
}
Expand Down
2 changes: 0 additions & 2 deletions src/hash_chain.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,8 +32,6 @@ pub enum DictionaryAddPolicy {
AddFirstAndLast(u16),
}

pub trait HashChainTrait: Default {}

#[derive(Default, Copy, Clone, Eq, PartialEq, Debug)]
struct InternalPosition {
pos: u16,
Expand Down
6 changes: 4 additions & 2 deletions src/process.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -233,6 +233,8 @@ fn analyze_compressed_data_fast(

let params = estimate_preflate_parameters(&contents.plain_text, &contents.blocks).unwrap();

println!("params: {:?}", params);

params.write(&mut cabac_encoder);
encode_mispredictions(&contents, &params, &mut cabac_encoder).unwrap();

Expand Down Expand Up @@ -420,7 +422,7 @@ fn verify_docx() {
// test binary deflate generated by starcontrol
#[test]
fn verify_savegame() {
do_analyze(None, &read_file("savegame.deflate"), true);
do_analyze(None, &read_file("savegame.deflate"), false);
}

#[test]
Expand Down Expand Up @@ -564,7 +566,7 @@ fn verify_miniz_compressed_1() {
//let minusheader = &v[2..v.len() - 4];
//let crc = Some(u32::from_le_bytes([v[v.len() - 4], v[v.len() - 3], v[v.len() - 2], v[v.len() - 1]]));

do_analyze(None, &v, true);
do_analyze(None, &v, false);
}

#[test]
Expand Down