void ParseFile()
{
for (;;) {
uint8_t *objectStart = cursor;
GUIDType type = ReadGUID();
uint64_t size = ReadQWord();
uint8_t *next = objectStart + size;
printf( "Object size: %llu\n", size );
switch (type) {
case GUIDHeader:
printf( "Header GUID\n" );
ParseHeaders();
break;
case GUIDData:
printf( "Found Data object\n" );
return;
case GUIDUnknown:
default:
printf( "Unknown GUID\n" );
break;
}
cursor = next;
}
}
ErrorCorrectionType ReadErrorCorrectionType()
{
GUIDType guid = ReadGUID();
switch (guid) {
case GUIDErrorCorrectionNone:
return ErrorCorrectionTypeNone;
case GUIDErrorCorrectionAudioSpread:
return ErrorCorrectionTypeAudioSpread;
default:
printf( "Found invalid error correction type\n" );
return ErrorCorrectionTypeNone;
}
}
StreamType ReadStreamType()
{
GUIDType guid = ReadGUID();
switch (guid) {
case GUIDMediaTypeAudio:
return StreamTypeAudio;
case GUIDMediaTypeVideo:
return StreamTypeVideo;
default:
printf( "Found invalid stream type\n" );
return StreamTypeError;
}
}
void KRegistryKey::ReadGUID(const KString& ValueName, GUID& RValue, const GUID& DefaultValue) const
{
DEBUG_VERIFY_ALLOCATION;
DEBUG_VERIFY(IsOpen());
try
{
ReadGUID(ValueName, RValue);
}
catch(...)
{
RValue = DefaultValue;
}
}
bool G2Packet::GetTo(QUuid& pGUID)
{
if ( m_bCompound == false ) return false;
if ( GetRemaining() < 4 + 16 ) return false;
char* pTest = m_oBuffer.data() + m_nPosition;
if ( pTest[0] != 0x48 ) return false;
if ( pTest[1] != 0x10 ) return false;
if ( pTest[2] != 'T' ) return false;
if ( pTest[3] != 'O' ) return false;
m_nPosition = 4;
pGUID = ReadGUID();
m_nPosition = 0;
return true;
}
void ParseHeaderExtension()
{
cursor += 16; // Skip reserved field 1
cursor += 2; // Skip reserved field 2
uint16_t dataSize = ReadDWord();
printf( "\tHeader extension data size: %d\n", dataSize );
uint8_t *end = cursor + dataSize;
while (cursor < end) {
uint8_t *objectStart = cursor;
GUIDType type = ReadGUID();
uint64_t size = ReadQWord();
uint8_t *next = objectStart + size;
switch (type) {
case GUIDUnknown:
default:
printf( "Unknown extended header object\n" );
break;
}
cursor = next;
}
}
void ParseHeaders()
{
printf( "Begin parsing headers\n" );
uint32_t headerCount = ReadDWord();
cursor += 2;
printf( "\t%d headers\n\n", headerCount );
for (uint32_t i = 0; i < headerCount; i++) {
uint8_t *objectStart = cursor;
GUIDType type = ReadGUID();
uint64_t size = ReadQWord();
uint8_t *next = objectStart + size;
printf( "\tOffset %ld, Object size: %llu\n", objectStart - buffer, size );
switch (type) {
case GUIDFileProperties:
printf( "\tFile properties found...\n" );
ParseFileProperties();
break;
case GUIDStreamProperties:
printf( "\tStream properties found...\n" );
ParseStreamProperties();
break;
case GUIDHeaderExtension:
printf( "\tHeader extension found...\n" );
ParseHeaderExtension();
break;
case GUIDUnknown:
default:
printf( "\tUnknown GUID\n" );
break;
}
cursor = next;
printf( "\n" );
}
printf( "end parsing headers\n" );
}
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 unsigned kNumHeaderSectionsMax = 5;
if (numHeaderSections != kNumHeaderSectionsMax)
return S_FALSE;
IsArc = true;
ReadUInt32(); // Len of post-header table
Byte g[16];
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();
UInt64 end = sectionOffsets[i] + sectionSizes[i];
database.UpdatePhySize(end);
}
// 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() != kSignature_CAOL)
return S_FALSE;
if (ReadUInt32() != 2) // (Most likely a version number)
return S_FALSE;
UInt32 caolLength = ReadUInt32(); // $50 (Len 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)
if (caolLength == 0x2C)
{
// fprintf(stdout, "\n !!!NewFormat\n");
// fflush(stdout);
database.ContentOffset = 0; // maybe we must add database.StartPosition here?
database.NewFormat = true;
}
else if (caolLength == 0x50)
{
ReadUInt32(); // 0 (Unknown)
if (ReadUInt32() != kSignature_ITSF)
return S_FALSE;
if (ReadUInt32() != 4) // $4 (Version number -- CHM uses 3)
return S_FALSE;
if (ReadUInt32() != 0x20) // $20 (length of ITSF)
return S_FALSE;
UInt32 unknown = ReadUInt32();
if (unknown != 0 && unknown != 1) // = 0 for some HxW files, 1 in other cases;
return S_FALSE;
database.ContentOffset = database.StartPosition + ReadUInt64();
/* UInt32 timeStamp = */ ReadUInt32();
// A timestamp of some sort.
// 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(); // BE?
}
else
return S_FALSE;
}
// Section 0
ReadChunk(inStream, database.StartPosition + sectionOffsets[0], sectionSizes[0]);
if (sectionSizes[0] < 0x18)
return S_FALSE;
if (ReadUInt32() != 0x01FE)
return S_FALSE;
ReadUInt32(); // unknown: 0
UInt64 fileSize = ReadUInt64();
database.UpdatePhySize(fileSize);
ReadUInt32(); // unknown: 0
ReadUInt32(); // unknown: 0
// Section 1: The Directory Listing
ReadChunk(inStream, database.StartPosition + sectionOffsets[1], sectionSizes[1]);
if (ReadUInt32() != kSignature_IFCM)
return S_FALSE;
if (ReadUInt32() != 1) // (probably a version number)
return S_FALSE;
UInt32 dirChunkSize = ReadUInt32(); // $2000
if (dirChunkSize < 64)
return S_FALSE;
ReadUInt32(); // $100000 (unknown)
ReadUInt32(); // -1 (unknown)
ReadUInt32(); // -1 (unknown)
UInt32 numDirChunks = ReadUInt32();
ReadUInt32(); // 0 (unknown, probably high word of above)
for (UInt32 ci = 0; ci < numDirChunks; ci++)
{
UInt64 chunkPos = _inBuffer.GetProcessedSize();
if (ReadUInt32() == kSignature_AOLL)
{
UInt32 quickrefLength = ReadUInt32(); // Len of quickref area at end of directory chunk
if (quickrefLength > dirChunkSize || quickrefLength < 2)
return S_FALSE;
ReadUInt64(); // Directory chunk number
// This must match physical position in file, that is
// the chunk size times the chunk number must be the
// offset from the end of the directory header.
ReadUInt64(); // Chunk number of previous listing chunk when reading
// directory in sequence (-1 if first listing chunk)
ReadUInt64(); // Chunk number of next listing chunk when reading
// directory in sequence (-1 if last listing chunk)
ReadUInt64(); // Number of first listing entry in this chunk
ReadUInt32(); // 1 (unknown -- other values have also been seen here)
ReadUInt32(); // 0 (unknown)
unsigned numItems = 0;
for (;;)
{
UInt64 offset = _inBuffer.GetProcessedSize() - chunkPos;
UInt32 offsetLimit = dirChunkSize - quickrefLength;
if (offset > offsetLimit)
return S_FALSE;
if (offset == offsetLimit)
break;
if (database.NewFormat)
{
UInt16 nameLen = ReadUInt16();
if (nameLen == 0)
return S_FALSE;
UString name;
ReadUString((unsigned)nameLen, name);
AString s;
ConvertUnicodeToUTF8(name, s);
Byte b = ReadByte();
s.Add_Space();
PrintByte(b, s);
s.Add_Space();
UInt64 len = ReadEncInt();
// then number of items ?
// then length ?
// then some data (binary encoding?)
while (len-- != 0)
{
b = ReadByte();
PrintByte(b, s);
}
database.NewFormatString += s;
database.NewFormatString += "\r\n";
}
else
{
RINOK(ReadDirEntry(database));
}
numItems++;
}
Skip(quickrefLength - 2);
if (ReadUInt16() != numItems)
return S_FALSE;
if (numItems > numDirEntries)
return S_FALSE;
numDirEntries -= numItems;
}
else
Skip(dirChunkSize - 4);
}
return numDirEntries == 0 ? S_OK : S_FALSE;
}
HRESULT CInArchive::OpenChm(IInStream *inStream, CDatabase &database)
{
UInt32 headerSize = ReadUInt32();
if (headerSize != 0x60)
return S_FALSE;
database.PhySize = headerSize;
UInt32 unknown1 = ReadUInt32();
if (unknown1 != 0 && unknown1 != 1) // it's 0 in one .sll file
return S_FALSE;
IsArc = true;
/* 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();
Byte g[16];
ReadGUID(g); // {7C01FD10-7BAA-11D0-9E0C-00A0-C922-E6EC}
ReadGUID(g); // {7C01FD11-7BAA-11D0-9E0C-00A0-C922-E6EC}
const unsigned kNumSections = 2;
UInt64 sectionOffsets[kNumSections];
UInt64 sectionSizes[kNumSections];
unsigned i;
for (i = 0; i < kNumSections; i++)
{
sectionOffsets[i] = ReadUInt64();
sectionSizes[i] = ReadUInt64();
UInt64 end = sectionOffsets[i] + sectionSizes[i];
database.UpdatePhySize(end);
}
// if (chmVersion == 3)
database.ContentOffset = ReadUInt64();
/*
else
database.ContentOffset = database.StartPosition + 0x58
*/
// Section 0
ReadChunk(inStream, sectionOffsets[0], sectionSizes[0]);
if (sectionSizes[0] < 0x18)
return S_FALSE;
if (ReadUInt32() != 0x01FE)
return S_FALSE;
ReadUInt32(); // unknown: 0
UInt64 fileSize = ReadUInt64();
database.UpdatePhySize(fileSize);
ReadUInt32(); // unknown: 0
ReadUInt32(); // unknown: 0
// Section 1: The Directory Listing
ReadChunk(inStream, sectionOffsets[1], sectionSizes[1]);
if (ReadUInt32() != kSignature_ITSP)
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() == kSignature_PMGL)
{
// 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(); // Len 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)
unsigned numItems = 0;
for (;;)
{
UInt64 offset = _inBuffer.GetProcessedSize() - chunkPos;
UInt32 offsetLimit = dirChunkSize - quickrefLength;
if (offset > offsetLimit)
return S_FALSE;
if (offset == offsetLimit)
break;
RINOK(ReadDirEntry(database));
numItems++;
}
Skip(quickrefLength - 2);
unsigned rrr = ReadUInt16();
if (rrr != numItems)
{
// Lazarus 9-26-2 chm contains 0 here.
if (rrr != 0)
return S_FALSE;
}
}
else
Skip(dirChunkSize - 4);
}
return S_OK;
}