From f817dc9254d3943b0ef0519290b79eaa846b4984 Mon Sep 17 00:00:00 2001
From: Dennis Ranke <dennis.ranke@gmail.com>
Date: Thu, 8 Sep 2022 23:42:03 +0200
Subject: [PATCH] first try of c decompressor, not working yet

---
 c_unpacker/.gitignore |  5 +++
 c_unpacker/Makefile   | 10 +++++
 c_unpacker/main.c     | 25 +++++++++++
 c_unpacker/unpack.c   | 97 +++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 137 insertions(+)
 create mode 100644 c_unpacker/.gitignore
 create mode 100644 c_unpacker/Makefile
 create mode 100644 c_unpacker/main.c
 create mode 100644 c_unpacker/unpack.c

diff --git a/c_unpacker/.gitignore b/c_unpacker/.gitignore
new file mode 100644
index 0000000..1251f15
--- /dev/null
+++ b/c_unpacker/.gitignore
@@ -0,0 +1,5 @@
+unpack
+unpack_bitstream
+unpack_debug
+*.upk
+
diff --git a/c_unpacker/Makefile b/c_unpacker/Makefile
new file mode 100644
index 0000000..0838f11
--- /dev/null
+++ b/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
diff --git a/c_unpacker/main.c b/c_unpacker/main.c
new file mode 100644
index 0000000..d9d9c9a
--- /dev/null
+++ b/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;
+}
diff --git a/c_unpacker/unpack.c b/c_unpacker/unpack.c
new file mode 100644
index 0000000..2318ed5
--- /dev/null
+++ b/c_unpacker/unpack.c
@@ -0,0 +1,97 @@
+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 & 1);
+        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;
+    
+    if(bit) {
+        upkr_state = prob * (upkr_state >> 8) + (upkr_state & 255);
+        upkr_probs[context_index] = prob + ((255 - prob + 8) >> 4);
+    } else {
+        upkr_state = (255 - prob) * (upkr_state >> 8) + (upkr_state & 255) - prob;
+        upkr_probs[context_index] = 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-- > 1) {
+                *write_ptr = write_ptr[-offset];
+                ++write_ptr;
+            }
+            prev_was_match = 1;
+        } else {
+            int context_index = 1;
+            int byte = 0;
+            for(int i = 0; i < 8; ++i) {
+                int bit = upkr_decode_bit(context_index);
+                context_index = (context_index << 1) + bit;
+                byte |= bit << i;
+            }
+            *write_ptr++ = byte;
+            prev_was_match = 0;
+        }
+    }
+    
+    return write_ptr - (u8*)destination;
+}