chmin.cpp

来自「由7-zip提供的压缩、解压缩程序」· C++ 代码 · 共 925 行 · 第 1/2 页

// Archive/ChmIn.cpp

#include "StdAfx.h"

#include "Common/StringConvert.h"
#include "Common/MyCom.h"
#include "Common/UTFConvert.h"
#include "Common/IntToString.h"
#include "Windows/Defs.h"

#include "../../Common/LimitedStreams.h"
#include "ChmIn.h"

namespace NArchive{
namespace NChm{

// define CHM_LOW, if you want to see low level items
// #define CHM_LOW

static const GUID kChmLzxGuid = 
  { 0x7FC28940, 0x9D31, 0x11D0, 0x9B, 0x27, 0x00, 0xA0, 0xC9, 0x1E, 0x9C, 0x7C };
static const GUID kHelp2LzxGuid = 
  { 0x0A9007C6, 0x4076, 0x11D3, 0x87, 0x89, 0x00, 0x00, 0xF8, 0x10, 0x57, 0x54 };
static const GUID kDesGuid = 
  { 0x67F6E4A2, 0x60BF, 0x11D3, 0x85, 0x40, 0x00, 0xC0, 0x4F, 0x58, 0xC3, 0xCF };

static bool AreGuidsEqual(REFGUID g1, REFGUID g2)
{ 
  if (g1.Data1 != g2.Data1 || 
      g1.Data2 != g2.Data2 ||
      g1.Data3 != g2.Data3)
    return false;
  for (int i = 0; i < 8; i++)
    if (g1.Data4[i] != g2.Data4[i])
      return false;
  return true;
}

static char GetHex(Byte value)
{
  return (value < 10) ? ('0' + value) : ('A' + (value - 10));
}

static void PrintByte(Byte b, AString &s)
{
  s += GetHex(b >> 4);
  s += GetHex(b & 0xF);
}

static void PrintUInt16(UInt16 v, AString &s)
{
  PrintByte((Byte)(v >> 8), s);
  PrintByte((Byte)v, s);
}

static void PrintUInt32(UInt32 v, AString &s)
{
  PrintUInt16((UInt16)(v >> 16), s);
  PrintUInt16((UInt16)v, s);
}

AString CMethodInfo::GetGuidString() const
{ 
  AString s;
  s += '{';
  PrintUInt32(Guid.Data1, s);
  s += '-';
  PrintUInt16(Guid.Data2, s);
  s += '-';
  PrintUInt16(Guid.Data3, s);
  s += '-';
  PrintByte(Guid.Data4[0], s);
  PrintByte(Guid.Data4[1], s);
  s += '-';
  for (int i = 2; i < 8; i++)
    PrintByte(Guid.Data4[i], s);
  s += '}';
  return s;
}

bool CMethodInfo::IsLzx() const
{
  if (AreGuidsEqual(Guid, kChmLzxGuid))
    return true;
  return AreGuidsEqual(Guid, kHelp2LzxGuid);
}

bool CMethodInfo::IsDes() const
{
  return AreGuidsEqual(Guid, kDesGuid);
}

UString CMethodInfo::GetName() const
{
  UString s;
  if (IsLzx())
  {
    s = L"LZX:";
    UInt32 numDictBits = LzxInfo.GetNumDictBits();
    wchar_t temp[32];
    ConvertUInt64ToString(numDictBits, temp);
    s += temp;
  }
  else
  {
    AString s2;
    if (IsDes())
      s2 = "DES";
    else
    {
      s2 = GetGuidString();
      if (ControlData.GetCapacity() > 0)
      {
        s2 += ":";
        for (size_t i = 0; i < ControlData.GetCapacity(); i++)
          PrintByte(ControlData[i], s2);
      }
    }
    ConvertUTF8ToUnicode(s2, s);
  }
  return s;
}

bool CSectionInfo::IsLzx() const
{
  if (Methods.Size() != 1)
    return false;
  return Methods[0].IsLzx();
}

UString CSectionInfo::GetMethodName() const
{
  UString s;
  if (!IsLzx())
  {
    UString temp;
    if (ConvertUTF8ToUnicode(Name, temp))
      s += temp;
    s += L": ";
  }
  for (int i = 0; i < Methods.Size(); i++)
  {
    if (i != 0)
      s += L" ";
    s += Methods[i].GetName();
  }
  return s;
}

Byte CInArchive::ReadByte()
{
  Byte b;
  if (!_inBuffer.ReadByte(b))
    throw 1;
  return b;
}

void CInArchive::Skeep(size_t size)
{
  while (size-- != 0)
    ReadByte();
}

void CInArchive::ReadBytes(Byte *data, UInt32 size)
{
  for (UInt32 i = 0; i < size; i++)
    data[i] = ReadByte();
}

UInt16 CInArchive::ReadUInt16()
{
  UInt16 value = 0;
  for (int i = 0; i < 2; i++)
    value |= ((UInt16)(ReadByte()) << (8 * i));
  return value;
}

UInt32 CInArchive::ReadUInt32()
{
  UInt32 value = 0;
  for (int i = 0; i < 4; i++)
    value |= ((UInt32)(ReadByte()) << (8 * i));
  return value;
}

UInt64 CInArchive::ReadUInt64()
{
  UInt64 value = 0;
  for (int i = 0; i < 8; i++)
    value |= ((UInt64)(ReadByte()) << (8 * i));
  return value;
}

UInt64 CInArchive::ReadEncInt()
{
  UInt64 val = 0;;
  for (int i = 0; i < 10; i++)
  {
    Byte b = ReadByte();
    val |= (b & 0x7F);
    if (b < 0x80)
      return val;
    val <<= 7;
  }
  throw 1;
}

void CInArchive::ReadGUID(GUID &g)
{
  g.Data1 = ReadUInt32();
  g.Data2 = ReadUInt16();
  g.Data3 = ReadUInt16();
  ReadBytes(g.Data4, 8);
}

void CInArchive::ReadString(int size, AString &s)
{
  s.Empty();
  while(size-- != 0)
  {
    char c = (char)ReadByte();
    if (c == 0)
    {
      Skeep(size);
      return;
    }
    s += c;
  }
}

void CInArchive::ReadUString(int size, UString &s)
{
  s.Empty();
  while(size-- != 0)
  {
    wchar_t c = ReadUInt16();
    if (c == 0)
    {
      Skeep(2 * size);
      return;
    }
    s += c;
  }
}

HRESULT CInArchive::ReadChunk(IInStream *inStream, UInt64 pos, UInt64 size)
{
  RINOK(inStream->Seek(pos, STREAM_SEEK_SET, NULL));
  CLimitedSequentialInStream *streamSpec = new CLimitedSequentialInStream;
  CMyComPtr<ISequentialInStream> limitedStream(streamSpec);
  streamSpec->Init(inStream, size);
  _inBuffer.SetStream(limitedStream);
  _inBuffer.Init();
  return S_OK;
}

HRESULT CInArchive::ReadDirEntry(CDatabase &database)
{
  CItem item;
  UInt64 nameLength = ReadEncInt();
  if (nameLength == 0 || nameLength >= 0x10000000)
    return S_FALSE;
  ReadString((int)nameLength, item.Name);
  item.Section = ReadEncInt();
  item.Offset = ReadEncInt();
  item.Size = ReadEncInt();
  database.Items.Add(item);
  return S_OK;
}

HRESULT CInArchive::OpenChm(IInStream *inStream, CDatabase &database)
{
  UInt32 headerSize = ReadUInt32();
  if (headerSize != 0x60)
    return S_FALSE;
  UInt32 unknown1 = ReadUInt32();
  if (unknown1 != 0 && unknown1 != 1) // it's 0 in one .sll file
    return S_FALSE;
  UInt32 timeStamp = ReadUInt32();
      // Considered as a big-endian DWORD, it appears to contain seconds (MSB) and 
      // fractional seconds (second byte). 
      // The third and fourth bytes may contain even more fractional bits.  
      // The 4 least significant bits in the last byte are constant.
  UInt32 lang = ReadUInt32();
  GUID g;
  ReadGUID(g); // {7C01FD10-7BAA-11D0-9E0C-00A0-C922-E6EC}
  ReadGUID(g); // {7C01FD11-7BAA-11D0-9E0C-00A0-C922-E6EC}
  const int kNumSections = 2;
  UInt64 sectionOffsets[kNumSections];
  UInt64 sectionSizes[kNumSections];
  int i;
  for (i = 0; i < kNumSections; i++)
  {
    sectionOffsets[i] = ReadUInt64();
    sectionSizes[i] = ReadUInt64();
  }
  // if (chmVersion == 3)
    database.ContentOffset = ReadUInt64();
  /*
  else
    database.ContentOffset = _startPosition + 0x58
  */

  /*
  // Section 0
  ReadChunk(inStream, sectionOffsets[0], sectionSizes[0]);
  if (sectionSizes[0] != 0x18)
    return S_FALSE;
  ReadUInt32(); // unknown:  01FE
  ReadUInt32(); // unknown:  0
  UInt64 fileSize = ReadUInt64();
  ReadUInt32(); // unknown:  0
  ReadUInt32(); // unknown:  0
  */

  // Section 1: The Directory Listing
  ReadChunk(inStream, sectionOffsets[1], sectionSizes[1]);
  if (ReadUInt32() != NHeader::kItspSignature)
    return S_FALSE;
  if (ReadUInt32() != 1) // version
    return S_FALSE;
  UInt32 dirHeaderSize = ReadUInt32();
  ReadUInt32(); // 0x0A (unknown)
  UInt32 dirChunkSize = ReadUInt32(); // $1000
  if (dirChunkSize < 32)
    return S_FALSE;
  UInt32 density = ReadUInt32(); //  "Density" of quickref section, usually 2.
  UInt32 depth = ReadUInt32(); //  Depth of the index tree: 1 there is no index, 
                               // 2 if there is one level of PMGI chunks.

  UInt32 chunkNumber = ReadUInt32(); //  Chunk number of root index chunk, -1 if there is none
                                     // (though at least one file has 0 despite there being no 
                                     // index chunk, probably a bug.) 
  UInt32 firstPmglChunkNumber = ReadUInt32(); // Chunk number of first PMGL (listing) chunk
  UInt32 lastPmglChunkNumber = ReadUInt32();  // Chunk number of last PMGL (listing) chunk
  ReadUInt32(); // -1 (unknown)
  UInt32 numDirChunks = ReadUInt32(); // Number of directory chunks (total)
  UInt32 windowsLangId = ReadUInt32();
  ReadGUID(g);  // {5D02926A-212E-11D0-9DF9-00A0C922E6EC}
  ReadUInt32(); // 0x54 (This is the length again)
  ReadUInt32(); // -1 (unknown)
  ReadUInt32(); // -1 (unknown)
  ReadUInt32(); // -1 (unknown)

  for (UInt32 ci = 0; ci < numDirChunks; ci++)
  {
    UInt64 chunkPos = _inBuffer.GetProcessedSize();
    if (ReadUInt32() == NHeader::kPmglSignature)
    {
      // The quickref area is written backwards from the end of the chunk. 
      // One quickref entry exists for every n entries in the file, where n 
      // is calculated as 1 + (1 << quickref density). So for density = 2, n = 5. 

      UInt32 quickrefLength = ReadUInt32(); // Length of free space and/or quickref area at end of directory chunk 
      if (quickrefLength > dirChunkSize || quickrefLength < 2)
        return S_FALSE;
      ReadUInt32(); // Always 0
      ReadUInt32(); // Chunk number of previous listing chunk when reading
                    // directory in sequence (-1 if this is the first listing chunk)
      ReadUInt32(); // Chunk number of next  listing chunk when reading
                    // directory in sequence (-1 if this is the last listing chunk)
      int numItems = 0;
      while (true)
      {
        UInt64 offset = _inBuffer.GetProcessedSize() - chunkPos;
        UInt32 offsetLimit = dirChunkSize - quickrefLength;
        if (offset > offsetLimit)
          return S_FALSE;
        if (offset == offsetLimit)
          break;
        RINOK(ReadDirEntry(database));
        numItems++;
      }
      Skeep(quickrefLength - 2);
      if (ReadUInt16() != numItems)
        return S_FALSE;
    }
    else
      Skeep(dirChunkSize - 4);
  }
  return S_OK;
}

HRESULT CInArchive::OpenHelp2(IInStream *inStream, CDatabase &database)
{
  if (ReadUInt32() != 1) // version
    return S_FALSE;
  if (ReadUInt32() != 0x28) // Location of header section table
    return S_FALSE;
  UInt32 numHeaderSections = ReadUInt32();
  const int kNumHeaderSectionsMax = 5;
  if (numHeaderSections != kNumHeaderSectionsMax)
    return S_FALSE;
  ReadUInt32(); // Length of post-header table
  GUID g; 
  ReadGUID(g);  // {0A9007C1-4076-11D3-8789-0000F8105754}

  // header section table 
  UInt64 sectionOffsets[kNumHeaderSectionsMax];
  UInt64 sectionSizes[kNumHeaderSectionsMax];
  UInt32 i;
  for (i = 0; i < numHeaderSections; i++)
  {
    sectionOffsets[i] = ReadUInt64();
    sectionSizes[i] = ReadUInt64();
  }
  
  // Post-Header
  ReadUInt32(); // 2
  ReadUInt32(); // 0x98: offset to CAOL from beginning of post-header)
  // ----- Directory information
  ReadUInt64(); // Chunk number of top-level AOLI chunk in directory, or -1
  ReadUInt64(); // Chunk number of first AOLL chunk in directory
  ReadUInt64(); // Chunk number of last AOLL chunk in directory
  ReadUInt64(); // 0 (unknown)
  ReadUInt32(); // $2000 (Directory chunk size of directory)
  ReadUInt32(); // Quickref density for main directory, usually 2
  ReadUInt32(); // 0 (unknown)
  ReadUInt32(); // Depth of main directory index tree
                // 1 there is no index, 2 if there is one level of AOLI chunks.
  ReadUInt64(); // 0 (unknown)
  UInt64 numDirEntries = ReadUInt64(); // Number of directory entries
  // ----- Directory Index Information
  ReadUInt64(); // -1 (unknown, probably chunk number of top-level AOLI in directory index)
  ReadUInt64(); // Chunk number of first AOLL chunk in directory index
  ReadUInt64(); // Chunk number of last AOLL chunk in directory index
  ReadUInt64(); // 0 (unknown)
  ReadUInt32(); // $200 (Directory chunk size of directory index)
  ReadUInt32(); // Quickref density for directory index, usually 2
  ReadUInt32(); // 0 (unknown)
  ReadUInt32(); // Depth of directory index index tree.
  ReadUInt64(); // Possibly flags -- sometimes 1, sometimes 0.
  ReadUInt64(); // Number of directory index entries (same as number of AOLL
               // chunks in main directory)
  
  // (The obvious guess for the following two fields, which recur in a number 
  // of places, is they are maximum sizes for the directory and directory index. 
  // However, I have seen no direct evidence that this is the case.) 

  ReadUInt32(); // $100000 (Same as field following chunk size in directory)
  ReadUInt32(); // $20000 (Same as field following chunk size in directory index)

  ReadUInt64(); // 0 (unknown)
  if (ReadUInt32() != NHeader::kCaolSignature)
    return S_FALSE;
  if (ReadUInt32() != 2) // (Most likely a version number)
    return S_FALSE;
  UInt32 caolLength = ReadUInt32(); // $50 (Length of the CAOL section, which includes the ITSF section)
  if (caolLength >= 0x2C)
  {
    UInt32 c7 = ReadUInt16(); // Unknown.  Remains the same when identical files are built.
              // Does not appear to be a checksum.  Many files have
              // 'HH' (HTML Help?) here, indicating this may be a compiler ID
              //  field.  But at least one ITOL/ITLS compiler does not set this
              // field to a constant value.
    ReadUInt16(); // 0 (Unknown.  Possibly part of 00A4 field)
    ReadUInt32(); // Unknown.  Two values have been seen -- $43ED, and 0.
    ReadUInt32(); // $2000 (Directory chunk size of directory)
    ReadUInt32(); // $200 (Directory chunk size of directory index)
    ReadUInt32(); // $100000 (Same as field following chunk size in directory)
    ReadUInt32(); // $20000 (Same as field following chunk size in directory index)
    ReadUInt32(); // 0 (unknown)
    ReadUInt32(); // 0 (Unknown)