Tried adding SQFC decompression (doesn't work)

dedmen · dedmen · commit b4b7f64bbae6 · 2024-02-13T17:20:27.000+01:00
diff --git a/src/lzokay_stream.cpp b/src/lzokay_stream.cpp
@@ -0,0 +1,249 @@
+// I hate this, but I need lzokay reading from a stream
+// this is 99% code from lzokay library from lib folder, just with std::istream used as input
+
+#include <istream>
+
+#include "lzokay.hpp"
+using namespace lzokay;
+
+
+#define NEEDS_IN(count) { \
+  auto curTell = (std::streamoff)inp.tellg(); \
+  if (curTell + (count) > inp_end) { \
+    dst_size = outp - dst; \
+    return EResult::InputOverrun; \
+  }}
+
+#define NEEDS_OUT(count) \
+  if (outp + (count) > outp_end) { \
+    dst_size = outp - dst; \
+    return EResult::OutputOverrun; \
+  }
+
+#define CONSUME_ZERO_BYTE_LENGTH \
+  std::size_t offset; \
+  { \
+    auto old_inp = inp.tellg(); \
+    while (inp.peek() == 0) inp.get(); \
+    offset = inp.tellg() - old_inp; \
+    if (offset > Max255Count) { \
+      dst_size = outp - dst; \
+      return EResult::Error; \
+    } \
+  }
+
+constexpr uint32_t M1Marker = 0x0;
+constexpr uint32_t M2Marker = 0x40;
+constexpr uint32_t M3Marker = 0x20;
+constexpr uint32_t M4Marker = 0x10;
+
+constexpr std::size_t Max255Count = std::size_t(~0) / 255 - 2;
+
+static uint16_t get_le16(std::istream& p) {
+    uint16_t result;
+    p.read((char*) &result, 2);
+    return result;
+}
+
+
+EResult decompressStream(std::istream& src, std::size_t src_size,
+                   uint8_t* dst, std::size_t init_dst_size,
+                   std::size_t& dst_size) {
+  dst_size = init_dst_size;
+
+  if (src_size < 3) {
+    dst_size = 0;
+    return EResult::InputOverrun;
+  }
+
+  std::istream& inp = src;
+  size_t inp_end = src_size;
+  uint8_t* outp = dst;
+  uint8_t* outp_end = dst + dst_size;
+  uint8_t* lbcur;
+  std::size_t lblen;
+  std::size_t state = 0;
+  std::size_t nstate = 0;
+
+  /* First byte encoding */
+  if (inp.peek() >= 22) {
+    /* 22..255 : copy literal string
+     *           length = (byte - 17) = 4..238
+     *           state = 4 [ don't copy extra literals ]
+     *           skip byte
+     */
+    std::size_t len = inp.get() - uint8_t(17);
+    NEEDS_IN(len)
+    NEEDS_OUT(len)
+    for (std::size_t i = 0; i < len; ++i)
+      *outp++ = inp.get();
+    state = 4;
+  } else if (inp.peek() >= 18) {
+    /* 18..21 : copy 0..3 literals
+     *          state = (byte - 17) = 0..3  [ copy <state> literals ]
+     *          skip byte
+     */
+    nstate = inp.get() - uint8_t(17);
+    state = nstate;
+    NEEDS_IN(nstate)
+    NEEDS_OUT(nstate)
+    for (std::size_t i = 0; i < nstate; ++i)
+      *outp++ = inp.get();
+  }
+  /* 0..17 : follow regular instruction encoding, see below. It is worth
+   *         noting that codes 16 and 17 will represent a block copy from
+   *         the dictionary which is empty, and that they will always be
+   *         invalid at this place.
+   */
+
+  while (true) {
+    NEEDS_IN(1)
+    uint8_t inst = inp.get();
+    if (inst & 0xC0) {
+      /* [M2]
+       * 1 L L D D D S S  (128..255)
+       *   Copy 5-8 bytes from block within 2kB distance
+       *   state = S (copy S literals after this block)
+       *   length = 5 + L
+       * Always followed by exactly one byte : H H H H H H H H
+       *   distance = (H << 3) + D + 1
+       *
+       * 0 1 L D D D S S  (64..127)
+       *   Copy 3-4 bytes from block within 2kB distance
+       *   state = S (copy S literals after this block)
+       *   length = 3 + L
+       * Always followed by exactly one byte : H H H H H H H H
+       *   distance = (H << 3) + D + 1
+       */
+      NEEDS_IN(1)
+      lbcur = outp - ((inp.get() << 3) + ((inst >> 2) & 0x7) + 1);
+      lblen = std::size_t(inst >> 5) + 1;
+      nstate = inst & uint8_t(0x3);
+    } else if (inst & M3Marker) {
+      /* [M3]
+       * 0 0 1 L L L L L  (32..63)
+       *   Copy of small block within 16kB distance (preferably less than 34B)
+       *   length = 2 + (L ?: 31 + (zero_bytes * 255) + non_zero_byte)
+       * Always followed by exactly one LE16 :  D D D D D D D D : D D D D D D S S
+       *   distance = D + 1
+       *   state = S (copy S literals after this block)
+       */
+      lblen = std::size_t(inst & uint8_t(0x1f)) + 2;
+      if (lblen == 2) {
+        CONSUME_ZERO_BYTE_LENGTH
+        NEEDS_IN(1)
+        lblen += offset * 255 + 31 + inp.get();
+      }
+      NEEDS_IN(2)
+      nstate = get_le16(inp);
+      //inp += 2;
+      lbcur = outp - ((nstate >> 2) + 1);
+      nstate &= 0x3;
+    } else if (inst & M4Marker) {
+      /* [M4]
+       * 0 0 0 1 H L L L  (16..31)
+       *   Copy of a block within 16..48kB distance (preferably less than 10B)
+       *   length = 2 + (L ?: 7 + (zero_bytes * 255) + non_zero_byte)
+       * Always followed by exactly one LE16 :  D D D D D D D D : D D D D D D S S
+       *   distance = 16384 + (H << 14) + D
+       *   state = S (copy S literals after this block)
+       *   End of stream is reached if distance == 16384
+       */
+      lblen = std::size_t(inst & uint8_t(0x7)) + 2;
+      if (lblen == 2) {
+        CONSUME_ZERO_BYTE_LENGTH
+        NEEDS_IN(1)
+        lblen += offset * 255 + 7 + inp.get();
+      }
+      NEEDS_IN(2)
+      nstate = get_le16(inp);
+      //inp += 2;
+      lbcur = outp - (((inst & 0x8) << 11) + (nstate >> 2));
+      nstate &= 0x3;
+      if (lbcur == outp)
+        break; /* Stream finished */
+      lbcur -= 16384;
+    } else {
+      /* [M1] Depends on the number of literals copied by the last instruction. */
+      if (state == 0) {
+        /* If last instruction did not copy any literal (state == 0), this
+         * encoding will be a copy of 4 or more literal, and must be interpreted
+         * like this :
+         *
+         *    0 0 0 0 L L L L  (0..15)  : copy long literal string
+         *    length = 3 + (L ?: 15 + (zero_bytes * 255) + non_zero_byte)
+         *    state = 4  (no extra literals are copied)
+         */
+        std::size_t len = inst + 3;
+        if (len == 3) {
+          CONSUME_ZERO_BYTE_LENGTH
+          NEEDS_IN(1)
+          len += offset * 255 + 15 + inp.get();
+        }
+        /* copy_literal_run */
+        NEEDS_IN(len)
+        NEEDS_OUT(len)
+        for (std::size_t i = 0; i < len; ++i)
+          *outp++ = inp.get();
+        state = 4;
+        continue;
+      } else if (state != 4) {
+        /* If last instruction used to copy between 1 to 3 literals (encoded in
+         * the instruction's opcode or distance), the instruction is a copy of a
+         * 2-byte block from the dictionary within a 1kB distance. It is worth
+         * noting that this instruction provides little savings since it uses 2
+         * bytes to encode a copy of 2 other bytes but it encodes the number of
+         * following literals for free. It must be interpreted like this :
+         *
+         *    0 0 0 0 D D S S  (0..15)  : copy 2 bytes from <= 1kB distance
+         *    length = 2
+         *    state = S (copy S literals after this block)
+         *  Always followed by exactly one byte : H H H H H H H H
+         *    distance = (H << 2) + D + 1
+         */
+        NEEDS_IN(1)
+        nstate = inst & uint8_t(0x3);
+        lbcur = outp - ((inst >> 2) + (inp.get() << 2) + 1);
+        lblen = 2;
+      } else {
+        /* If last instruction used to copy 4 or more literals (as detected by
+         * state == 4), the instruction becomes a copy of a 3-byte block from the
+         * dictionary from a 2..3kB distance, and must be interpreted like this :
+         *
+         *    0 0 0 0 D D S S  (0..15)  : copy 3 bytes from 2..3 kB distance
+         *    length = 3
+         *    state = S (copy S literals after this block)
+         *  Always followed by exactly one byte : H H H H H H H H
+         *    distance = (H << 2) + D + 2049
+         */
+        NEEDS_IN(1)
+        nstate = inst & uint8_t(0x3);
+        lbcur = outp - ((inst >> 2) + (inp.get() << 2) + 2049);
+        lblen = 3;
+      }
+    }
+    if (lbcur < dst) {
+      dst_size = outp - dst;
+      return EResult::LookbehindOverrun;
+    }
+    NEEDS_IN(nstate)
+    NEEDS_OUT(lblen + nstate)
+    /* Copy lookbehind */
+    for (std::size_t i = 0; i < lblen; ++i)
+      *outp++ = *lbcur++;
+    state = nstate;
+    /* Copy literal */
+    for (std::size_t i = 0; i < nstate; ++i)
+      *outp++ = inp.get();
+  }
+
+  dst_size = outp - dst;
+  if (lblen != 3) /* Ensure terminating M4 was encountered */
+    return EResult::Error;
+  if (inp.tellg() == inp_end)
+    return EResult::Success;
+  else if (inp.tellg() < inp_end)
+    return EResult::InputNotConsumed;
+  else
+    return EResult::InputOverrun;
+}
diff --git a/src/main.cpp b/src/main.cpp
@@ -137,10 +137,29 @@ void processFile(ScriptCompiler& comp, std::filesystem::path path) {
     catch (std::runtime_error& err) {
 
     }
+}
+
+void DecompressSQFC(std::filesystem::path inputPath, std::filesystem::path outputPath)
+{
+    // To use this, also need to edit ScriptSerializer::compiledToBinary and disable compressed serialization
+
+    std::ifstream inputFile(inputPath, std::ifstream::binary);
+    auto compiledData = ScriptSerializer::binaryToCompiled(inputFile);
 
 
+
+    std::stringstream output(std::stringstream::binary | std::stringstream::out);
+    ScriptSerializer::compiledToBinary(compiledData, output);
+
+    auto data = output.str();
+    auto encoded = data; //base64_encode(data);
+    std::ofstream outputFile(outputPath, std::ofstream::binary);
+
+    outputFile.write(encoded.data(), encoded.length());
+    outputFile.flush();
 }
 
+
 int main(int argc, char* argv[]) {
 
     if (std::filesystem::exists("sqfc.lua")) {
diff --git a/src/scriptSerializer.cpp b/src/scriptSerializer.cpp
@@ -4,6 +4,7 @@
 #include <functional>
 #include <zstd.h>
 #include <sstream>
+#include <strstream>
 
 static constexpr const int compressionLevel = 22;
 
@@ -205,7 +206,7 @@ void ScriptSerializer::compiledToBinary(const CompiledCodeData& code, std::ostre
             __debugbreak();
 
         writeT(static_cast<uint32_t>(bufferContent.size()), output); // uncompressed size
-        writeT(static_cast<uint8_t>(2), output); // compression method, always 2
+        writeT(static_cast<uint8_t>(2), output); // compression method, always 2 (That is RV engine stuff)
         output.write((const char*)compressed.get(), compressed_size);
     } else {
         writeT(static_cast<uint8_t>(SerializedBlockType::constant), output);
@@ -227,6 +228,58 @@ void ScriptSerializer::compiledToBinary(const CompiledCodeData& code, std::ostre
     output.flush();
 }
 
+// lzokay_stream.cpp
+extern lzokay::EResult decompressStream(std::istream& src, std::size_t src_size,
+                                        uint8_t* dst, std::size_t init_dst_size,
+                                        std::size_t& dst_size);
+
+struct DecompressedData
+{
+    DecompressedData(std::vector<uint8_t>&& data):
+        uncompressedData(std::move(data)),
+        buf(uncompressedData.data(), uncompressedData.size()),
+        stream(&buf)
+    {}
+
+    std::vector<uint8_t> uncompressedData;
+    std::strstreambuf buf;
+    std::istream stream;
+};
+
+//#TODO make the compress write also a method
+
+// This is shit code, it depends on RVO being applied otherwise the streambuf pointer would change
+DecompressedData DecompressFromStream(std::istream& input)
+{
+    size_t uncompressedSize = readT<uint32_t>(input);
+    readT<uint8_t>(input); // compression method, always 2
+
+    // compressed buffer
+
+    std::vector<uint8_t> uncompressedData;
+    uncompressedData.resize(uncompressedSize);
+
+    const auto pos = input.tellg();
+    input.seekg(0, SEEK_END);
+    const std::streamoff size = input.tellg(); // -pos;
+    input.seekg(pos, SEEK_SET);
+
+    //if (pos == 0x2bc && size == 3413)
+    //{
+    //    std::vector<uint8_t> compData;
+    //    compData.resize(size);
+    //    input.read((char*)compData.data(), size);
+    //    auto res = lzokay::decompress(compData.data(), size - pos, uncompressedData.data(), uncompressedSize, uncompressedSize);
+    //}
+
+    const auto error = decompressStream(input, size, uncompressedData.data(), uncompressedSize, uncompressedSize);
+    if (error < lzokay::EResult::Success)
+        __debugbreak();
+
+    return { std::move(uncompressedData) };
+}
+
+
 CompiledCodeData ScriptSerializer::binaryToCompiled(std::istream& input) {
     CompiledCodeData output;
     output.version = readT<uint32_t>(input);
@@ -241,6 +294,10 @@ CompiledCodeData ScriptSerializer::binaryToCompiled(std::istream& input) {
             case SerializedBlockType::constant: {
                 readConstants(output, input);
             } break;
+            case SerializedBlockType::constantCompressed: {
+                auto decompressedData = DecompressFromStream(input);
+                readConstants(output, decompressedData.stream);
+            } break;
             case SerializedBlockType::locationInfo: {
                 auto locCount = readT<uint16_t>(input);
 
@@ -251,6 +308,19 @@ CompiledCodeData ScriptSerializer::binaryToCompiled(std::istream& input) {
             case SerializedBlockType::code: {
                 output.codeIndex = readT<uint64_t>(input);
             } break;
+            case SerializedBlockType::codeDebug: {
+                __debugbreak(); // not implemented
+            } break;
+            case SerializedBlockType::commandNameDirectory: {
+                // Compressed buffer
+
+                auto decompressedData = DecompressFromStream(input);
+
+                auto numCommandNames = readT<uint16_t>(decompressedData.stream);
+                output.commandNameDirectory.reserve(numCommandNames);
+                for (int i = 0; i < numCommandNames; ++i)
+                    output.commandNameDirectory.emplace_back(readString(decompressedData.stream));
+            } break;
         }
     }
     return output;
@@ -639,8 +709,8 @@ std::vector<char> ScriptSerializer::decompressDataDictionary(const std::vector<c
 }
 
 STRINGTYPE ScriptSerializer::readString(std::istream& input) {
-    uint32_t length;
-    input.read(reinterpret_cast<char*>(&length) + 1, 3);
+    uint32_t length{}; // length is actually uint24
+    input.read(reinterpret_cast<char*>(&length), 3);
 
 
     if constexpr (std::is_same_v<STRINGTYPE, std::string>) {
