oozle 0.1.5

Open source Kraken, Mermaid, Selkie, Leviathan, LZNA, Bitknit decompressor
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260

use anyhow::{Error, Result};
use log::{debug, trace, warn};

use crate::ffi;

use super::common::{copy_whole_match, get_crc};
use super::header::Header;
use super::quantum_header::QuantumHeader;
use super::DecoderType;

pub struct Decoder {
    pub input_read: u32,
    pub output_written: u32,
    scratch: [u8; 0x6C000],
    header: Header,
    quantum_header: QuantumHeader,
}

impl Decoder {
    pub fn parse_header(&mut self, input: &[u8]) -> Result<usize> {
        self.header.parse(input)
    }

    pub fn parse_quantum_header(&mut self, input: &[u8]) -> Result<usize> {
        self.quantum_header.parse(input, self.header.use_checksums)
    }

    pub fn parse_lzna_quantum_header(&mut self, input: &[u8], output_len: usize) -> Result<usize> {
        self.quantum_header
            .parse_lzna(input, self.header.use_checksums, output_len)
    }

    pub unsafe fn decode_step(
        &mut self,
        output: &mut [u8],
        output_offset: usize,
        input: &[u8],
    ) -> Result<()> {
        let mut input_offset: usize = 0;

        // If the output buffer is less than 0x40000, parse the header
        if output_offset < 0x40000 {
            trace!("Parsing header");
            input_offset += self.parse_header(input)?;
        }

        // Get type of decoder
        let is_kraken_decoder = self.header.decoder_type == DecoderType::Kraken
            || self.header.decoder_type == DecoderType::Mermaid
            || self.header.decoder_type == DecoderType::Leviathan;

        // Calculate the output length
        let output_len = std::cmp::min(
            output.len() - output_offset,
            if is_kraken_decoder { 0x40000 } else { 0x4000 },
        );

        // Check if the data is compressed
        trace!("Checking if data is compressed");
        if self.header.uncompressed {
            // Check if the input buffer is large enough and return if not
            if input.len() - input_offset < output_len {
                self.input_read = 0;
                self.output_written = 0;
                return Ok(());
            }

            // Copy the data to the output buffer
            output[output_offset..output_offset + output_len]
                .copy_from_slice(&input[input_offset..input_offset + output_len]);
            self.input_read = (input_offset + output_len) as u32;
            self.output_written = output_len as u32;
            return Ok(());
        }

        // If the decoder is a Kraken decoder, parse the quantum header
        if is_kraken_decoder {
            trace!("Parsing quantum header");
            input_offset += self.parse_quantum_header(&input[input_offset..])?;
        }
        // Otherwise, parse the LZNA quantum header
        else {
            trace!("Parsing LZNA quantum header");
            input_offset += self.parse_lzna_quantum_header(&input[input_offset..], output_len)?;
        }

        // Check if the input buffer has enough data to decompress and return if not
        trace!("Checking if input buffer has enough data to decompress");
        if input.len() - input_offset < self.quantum_header.compressed_size as usize {
            self.input_read = 0;
            self.output_written = 0;
            return Ok(());
        }

        // Check if the output buffer is large enough for the decompressed data
        // and return an error if not
        trace!("Checking if output buffer is large enough for decompressed data");
        if output_len < self.quantum_header.compressed_size as usize {
            return Err(Error::msg("Output buffer is too small"));
        }

        // If the compressed size is 0, the data should either be parsed as a whole match
        // or filled with the checksum
        trace!("Checking if compressed size is 0");
        if self.quantum_header.compressed_size == 0 {
            if self.quantum_header.whole_match_distance != 0 {
                // Check if the whole match distance is too large
                if self.quantum_header.whole_match_distance > output_offset as u32 {
                    return Err(Error::msg("Whole match distance is too large"));
                }

                copy_whole_match(
                    output,
                    output_offset,
                    self.quantum_header.whole_match_distance as usize,
                );
            } else {
                // TODO: Verify that this is the correct analogy to
                // memset(output + output_offset, checksum, output_len);
                output[output_offset..output_offset + output_len]
                    .fill(self.quantum_header.checksum as u8);
            }

            self.input_read = input_offset as u32;
            self.output_written = output_len as u32;
            return Ok(());
        }

        // Get the CRC checksum of the compressed data
        trace!("Getting CRC checksum of compressed data");
        let crc = get_crc(
            &input[input_offset..input_offset + self.quantum_header.compressed_size as usize],
        );

        // Check if the checksum is correct
        trace!("Checking if checksum is correct");
        if self.header.use_checksums && (crc & 0xFFFFFF) != self.quantum_header.checksum {
            return Err(Error::msg("Checksum mismatch"));
        }

        // If the compressed size is the same as the output length, the data is not compressed.
        // Copy the data to the output buffer and return
        trace!("Checking if compressed size is the same as output length");
        if self.quantum_header.compressed_size == output_len as u32 {
            output[output_offset..output_offset + output_len]
                .copy_from_slice(&input[input_offset..input_offset + output_len]);

            self.input_read = (input_offset + output_len) as u32;
            self.output_written = output_len as u32;
            return Ok(());
        }

        // Decompress the data using the correct decoder
        trace!("Decompressing data using correct decoder");
        debug!(
            "Decompressing {} bytes to {} bytes using {:?}",
            self.quantum_header.compressed_size, output_len, self.header.decoder_type
        );
        debug!(
            "Input offset: {} Output offset: {}",
            input_offset, output_offset
        );
        let read_bytes: i32 = match self.header.decoder_type {
            DecoderType::Kraken => ffi::Kraken_DecodeQuantum(
                output.as_mut_ptr().add(output_offset),
                output.as_mut_ptr().add(output_offset + output_len),
                output.as_mut_ptr(),
                input.as_ptr().add(input_offset),
                input
                    .as_ptr()
                    .add(input_offset + self.quantum_header.compressed_size as usize),
                self.scratch.as_mut_ptr(),
                self.scratch.as_mut_ptr().add(self.scratch.len()),
            ),
            DecoderType::Mermaid => ffi::Mermaid_DecodeQuantum(
                output.as_mut_ptr().add(output_offset),
                output.as_mut_ptr().add(output_offset + output_len),
                output.as_mut_ptr(),
                input.as_ptr().add(input_offset),
                input
                    .as_ptr()
                    .add(input_offset + self.quantum_header.compressed_size as usize),
                self.scratch.as_mut_ptr(),
                self.scratch.as_mut_ptr().add(self.scratch.len()),
            ),
            DecoderType::Leviathan => ffi::Leviathan_DecodeQuantum(
                output.as_mut_ptr().add(output_offset),
                output.as_mut_ptr().add(output_offset + output_len),
                output.as_mut_ptr(),
                input.as_ptr().add(input_offset),
                input
                    .as_ptr()
                    .add(input_offset + self.quantum_header.compressed_size as usize),
                self.scratch.as_mut_ptr(),
                self.scratch.as_mut_ptr().add(self.scratch.len()),
            ),
            DecoderType::LZNA => {
                if self.header.restart_decoder {
                    self.header.restart_decoder = false;
                    ffi::LZNA_InitLookup(self.scratch.as_mut_ptr() as *mut ffi::LznaState);
                }
                ffi::LZNA_DecodeQuantum(
                    output.as_mut_ptr().add(output_offset),
                    output.as_mut_ptr().add(output_offset + output_len),
                    output.as_mut_ptr(),
                    input.as_ptr().add(input_offset),
                    input
                        .as_ptr()
                        .add(input_offset + self.quantum_header.compressed_size as usize),
                    self.scratch.as_mut_ptr() as *mut ffi::LznaState,
                )
            }
            DecoderType::Bitknit => {
                if self.header.restart_decoder {
                    self.header.restart_decoder = false;
                    ffi::BitknitState_Init(self.scratch.as_mut_ptr() as *mut ffi::BitknitState);
                }
                ffi::Bitknit_Decode(
                    input.as_ptr().add(input_offset),
                    input
                        .as_ptr()
                        .add(input_offset + self.quantum_header.compressed_size as usize),
                    output.as_mut_ptr().add(output_offset),
                    output.as_mut_ptr().add(output_offset + output_len),
                    output.as_mut_ptr(),
                    self.scratch.as_mut_ptr() as *mut ffi::BitknitState,
                ) as i32
            }
            _ => return Err(Error::msg("Unknown decoder type")),
        };

        // Check if the decompressed size is correct
        trace!("Checking if decompressed size is correct");
        if read_bytes != self.quantum_header.compressed_size as i32 {
            let message = format!(
                "Decompressed size mismatch: {} != {}",
                read_bytes, self.quantum_header.compressed_size
            );
            warn!("{}", message);
            return Err(Error::msg(message));
        }

        self.input_read = input_offset as u32 + read_bytes as u32;
        self.output_written = output_len as u32;

        Ok(())
    }
}

impl Default for Decoder {
    fn default() -> Self {
        Self {
            input_read: 0,
            output_written: 0,
            scratch: [0; 0x6C000],
            header: Header::default(),
            quantum_header: QuantumHeader::default(),
        }
    }
}