implement simplified prob update, update unpack.c

c_unpacker/.gitignore

@@ -0,0 +1,5 @@
+unpack
+unpack_bitstream
+unpack_debug
+*.upk
+

c_unpacker/Makefile

@@ -0,0 +1,10 @@
+all: unpack unpack_bitstream
+
+unpack: main.c unpack.c
+	cc -O2 -o unpack main.c unpack.c
+
+unpack_bitstream: main.c unpack.c
+	cc -O2 -D UPKR_BITSTREAM -o unpack_bitstream main.c unpack.c
+	
+unpack_debug: main.c unpack.c
+	cc -g -o unpack_debug main.c unpack.c

c_unpacker/main.c

@@ -0,0 +1,25 @@
+#include <stdio.h>
+#include <stdlib.h>
+
+int upkr_unpack(void* destination, void* compressed_data);
+
+int main(int argn, char** argv) {
+  void* input_buffer = malloc(1024*1024);
+  void* output_buffer = malloc(1024*1024);
+  
+  FILE* in_file = fopen(argv[1], "rb");
+  int in_size = fread(input_buffer, 1, 1024*1024, in_file);
+  fclose(in_file);
+  
+  printf("Compressed size: %d\n", in_size);
+  
+  int out_size = upkr_unpack(output_buffer, input_buffer);
+  
+  printf("Uncompressed size: %d\n", out_size);
+  
+  FILE* out_file = fopen(argv[2], "wb");
+  fwrite(output_buffer, 1, out_size, out_file);
+  fclose(out_file);
+  
+  return 0;
+}

c_unpacker/readme.txt

@@ -0,0 +1,4 @@
+a very simple unpacker in c, as a reference for people wanting to implement their own unpacker.
+absolutely not production ready, it makes no effort to ensure the output buffer can actually
+hold the uncompressed data.
+!!! Never run on untrusted input !!!

c_unpacker/unpack.c

@@ -0,0 +1,98 @@
+typedef unsigned char u8;
+typedef unsigned short u16;
+typedef unsigned long u32;
+
+u8* upkr_data_ptr;
+u8 upkr_probs[1 + 255 + 1 + 2*32 + 2*32]; 
+#ifdef UPKR_BITSTREAM
+u16 upkr_state;
+u8 upkr_current_byte;
+int upkr_bits_left;
+#else
+u32 upkr_state;
+#endif
+
+int upkr_decode_bit(int context_index) {
+#ifdef UPKR_BITSTREAM
+    while(upkr_state < 32768) {
+        if(upkr_bits_left == 0) {
+            upkr_current_byte = *upkr_data_ptr++;
+            upkr_bits_left = 8;
+        }
+        upkr_state = (upkr_state << 1) + (upkr_current_byte >> 7);
+        upkr_current_byte <<= 1;
+        --upkr_bits_left;
+    }
+#else
+    while(upkr_state < 4096) {
+        upkr_state = (upkr_state << 8) | *upkr_data_ptr++;
+    }
+#endif
+   
+    int prob = upkr_probs[context_index];
+    int bit = (upkr_state & 255) >= prob ? 1 : 0;
+    
+    int prob_offset = 16;
+    int state_offset = 0;
+    int state_scale = prob;
+    if(bit) {
+        state_offset = -prob;
+        state_scale = 256 - prob;
+        prob_offset = 0;
+    }
+    upkr_state = state_offset + state_scale * (upkr_state >> 8) + (upkr_state & 255);
+    upkr_probs[context_index] = prob_offset + prob - ((prob + 8) >> 4); 
+
+    return bit;
+}
+
+int upkr_decode_length(int context_index) {
+    int length = 0;
+    int bit_pos = 0;
+    while(upkr_decode_bit(context_index)) {
+        length |= upkr_decode_bit(context_index + 1) << bit_pos++;
+        context_index += 2;
+    }
+    return length | (1 << bit_pos);
+}
+
+int upkr_unpack(void* destination, void* compressed_data) {
+    upkr_data_ptr = (u8*)compressed_data;
+    upkr_state = 0;
+#ifdef UPKR_BITSTREAM
+    upkr_bits_left = 0;
+#endif
+    for(int i = 0; i < sizeof(upkr_probs); ++i)
+        upkr_probs[i] = 128;
+    
+    u8* write_ptr = (u8*)destination;
+    
+    int prev_was_match = 0;
+    int offset = 0;
+    for(;;) {
+        if(upkr_decode_bit(0)) {
+            if(prev_was_match || upkr_decode_bit(256)) {
+                offset = upkr_decode_length(257) - 1;
+                if(offset == 0) {
+                    break;
+                }
+            }
+            int length = upkr_decode_length(257 + 64);
+            while(length--) {
+                *write_ptr = write_ptr[-offset];
+                ++write_ptr;
+            }
+            prev_was_match = 1;
+        } else {
+            int byte = 1;
+            while(byte < 256) {
+                int bit = upkr_decode_bit(byte);
+                byte = (byte << 1) + bit;
+            }
+            *write_ptr++ = byte;
+            prev_was_match = 0;
+        }
+    }
+    
+    return write_ptr - (u8*)destination;
+}

src/context_state.rs

@@ -1,8 +1,8 @@
-use crate::rans::{PROB_BITS, ONE_PROB};
+use crate::rans::{ONE_PROB, PROB_BITS};
 
 const INIT_PROB: u16 = 1 << (PROB_BITS - 1);
-const UPDATE_RATE: u32 = 4;
-const UPDATE_ADD: u32 = 8;
+const UPDATE_RATE: i32 = 4;
+const UPDATE_ADD: i32 = 8;
 
 #[derive(Clone)]
 pub struct ContextState {
@@ -33,10 +33,13 @@ impl<'a> Context<'a> {
 
     pub fn update(&mut self, bit: bool) {
         let old = self.state.contexts[self.index];
-        self.state.contexts[self.index] = if bit {
-            old + ((ONE_PROB - old as u32 + UPDATE_ADD) >> UPDATE_RATE) as u8
+        let offset = if !bit {
+            ONE_PROB as i32 >> UPDATE_RATE
         } else {
-            old - ((old as u32 + UPDATE_ADD) >> UPDATE_RATE) as u8
+            0
         };
+
+        self.state.contexts[self.index] =
+            (offset + old as i32 - ((old as i32 + UPDATE_ADD) >> UPDATE_RATE)) as u8;
     }
 }

src/lib.rs

@@ -21,3 +21,12 @@ pub fn pack(
         parsing_packer::pack(data, level, use_bitstream, progress_callback)
     }
 }
+
+pub fn compressed_size(mut data: &[u8]) -> f32 {
+    let mut state = 0;
+    while state < 4096 {
+        state = (state << 8) | data[0] as u32;
+        data = &data[1..];
+    }
+    data.len() as f32 + (state as f32).log2() / 8.
+}

src/main.rs

@@ -58,7 +58,10 @@ fn main() -> Result<()> {
 
 fn print_help() {
     eprintln!("Usage:");
-    eprintln!("  upkr pack [-l level(0-9)] <infile> <outfile>");
-    eprintln!("  upkr unpack <infile> <outfile>");
+    eprintln!("  upkr pack [-b] [-l level(0-9)] <infile> <outfile>");
+    eprintln!("  upkr unpack [-b] <infile> <outfile>");
+    eprintln!();
+    eprintln!(" -b, --bitstream     bitstream mode");
+    eprintln!(" -l, --level N       compression level 0-9");
     std::process::exit(1);
 }

src/rans.rs

@@ -38,15 +38,15 @@ impl RansCoder {
         let mut state = 1 << l_bits;
 
         let mut byte = 0u8;
-        let mut bit = 8;
+        let mut bit = 0;
         let mut flush_state: Box<dyn FnMut(&mut u32)> = if self.use_bitstream {
             Box::new(|state: &mut u32| {
-                bit -= 1;
                 byte |= ((*state & 1) as u8) << bit;
-                if bit == 0 {
+                bit += 1;
+                if bit == 8 {
                     buffer.push(byte);
                     byte = 0;
-                    bit = 8;
+                    bit = 0;
                 }
                 *state >>= 1;
             })
@@ -61,7 +61,7 @@ impl RansCoder {
         let max_state_factor: u32 = 1 << (l_bits + num_flush_bits - PROB_BITS);
         for step in self.bits.into_iter().rev() {
             let prob = step as u32 & 32767;
-            let (start, prob) = if step & 32768 != 0 {
+            let (start, prob) = if step & 32768 == 0 {
                 (0, prob)
             } else {
                 (prob, ONE_PROB - prob)
@@ -118,7 +118,7 @@ impl CostCounter {
 
 impl EntropyCoder for CostCounter {
     fn encode_bit(&mut self, bit: bool, prob: u16) {
-        let prob = if bit {
+        let prob = if !bit {
             prob as u32
         } else {
             ONE_PROB - prob as u32
@@ -163,8 +163,8 @@ impl<'a> RansDecoder<'a> {
                     self.data = &self.data[1..];
                     self.bits_left = 8;
                 }
-                self.state = (self.state << 1) | (self.byte & 1) as u32;
-                self.byte >>= 1;
+                self.state = (self.state << 1) | (self.byte >> 7) as u32;
+                self.byte <<= 1;
                 self.bits_left -= 1;
             }
         } else {
@@ -174,12 +174,12 @@ impl<'a> RansDecoder<'a> {
             }
         }
 
-        let bit = (self.state & PROB_MASK) < prob;
+        let bit = (self.state & PROB_MASK) >= prob;
 
         let (start, prob) = if bit {
-            (0, prob)
-        } else {
             (prob, ONE_PROB - prob)
+        } else {
+            (0, prob)
         };
         self.state = prob * (self.state >> PROB_BITS) + (self.state & PROB_MASK) - start;