Mediatype.cpp

/*
 * Copyright (C) 2008, OV2 <overfiend@sessionclan.de>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program.  If not, see <http://www.gnu.org/licenses/>.
 */

/* 
   most of the information used in this file comes from
   http://msdn.microsoft.com/en-us/library/ms787558(VS.85).aspx
 */

#include <string.h>
#include <objbase.h>
#include <stdio.h>
#include <streams.h>

#include "Anchor.h"
#include "Utils.h"
#include "RFS.h"
#include "Mediatype.h"
#include "File.h"


/* getNextToken extracts the current token from the string and 
   and sets the starting point of the next token */
static void getNextToken (wchar_t **strTok, wchar_t **nextTokenStart)
{
	while (**strTok == TEXT(' '))
		(*strTok) ++;

	for (int i = 0; i < lstrlen (*strTok); i ++)
	{
		if ((*strTok) [i] == TEXT(','))
		{
			(*strTok) [i] = TEXT('\0');
			*nextTokenStart = *strTok + i + 1;
			return;
		}
	}
	*nextTokenStart = NULL;
}

/* parseCheckBytes splits valueData and places the parsed values in the
   checkBytes array (allocated here) */
static int parseCheckBytes (wchar_t *valueData, CheckByteDetails **checkBytes)
{
	int tokenCount = 0, byteDetailsCount = 0, value;
	wchar_t *token, *nextTokenStart;

	for (int i = 0; i < lstrlen (valueData); i ++)
		if (valueData [i] == TEXT(','))
			tokenCount ++;

	if ((++ tokenCount) % 4)
		return 0;

	byteDetailsCount = tokenCount / 4;
	*checkBytes = new CheckByteDetails [byteDetailsCount];

	if (!*checkBytes)
	{
		ErrorMsg (0, L"Out of memory.");
		return -1;
	}

	nextTokenStart = valueData;

	for (int i = 0; i < byteDetailsCount; i ++)
	{
		token = nextTokenStart;
		getNextToken (&token, &nextTokenStart);
		swscanf_s (token, TEXT(" %lld"), &(*checkBytes) [i].offset);	// offset in file
		token = nextTokenStart;
		getNextToken (&token, &nextTokenStart);
		swscanf_s (token, TEXT(" %u"), &(*checkBytes) [i].byteCount);	// byte count of mask and value
		(*checkBytes) [i].mask = new BYTE [(*checkBytes) [i].byteCount];
		(*checkBytes) [i].value = new BYTE [(*checkBytes) [i].byteCount];

		if (!(*checkBytes) [i].mask || !(*checkBytes) [i].value)
		{
			ErrorMsg (0, L"Out of memory.");
			return -1;
		}

		//TODO: if the next token is smaller than byteCount and not empty we could pad with zeroes
		token = nextTokenStart;
		getNextToken (&token, &nextTokenStart);

		if (*token == TEXT('\0'))	// no mask means all ones
			for (unsigned int j = 0; j < (*checkBytes) [i].byteCount; j ++)
				(*checkBytes) [i].mask [j] = 0xFF;
		else					// otherwise parse mask
		{
			for (unsigned int j = 0; j < (*checkBytes) [i].byteCount && j * 2 < lstrlen (token); j ++)
			{
				swscanf_s (token + j * 2, TEXT(" %2x"), &value);
				(*checkBytes) [i].mask [j] = value;
			}
		}

		token = nextTokenStart;
		getNextToken (&token, &nextTokenStart);

		for (unsigned int j = 0; j < (*checkBytes) [i].byteCount && j * 2 < lstrlen (token); j ++)
		{
			swscanf_s (token + j * 2, TEXT(" %2x"), &value);	// parse value
			(*checkBytes) [i].value [j] = value;
		}
	}
	return byteDetailsCount;
}

/* parses the HKEY_CLASSES_ROOT\Media Type registry key and
   fills the mediaTypeList with all valid byte marks and their
   corresponding major/subtypes */
int getMediaTypeList (List<MediaType> *mediaTypeList)
{
//these values come from http://msdn.microsoft.com/en-us/library/ms724872(VS.85).aspx
#define MAX_VALUE_SIZE	16384
#define MAX_KEYNAME_SIZE 256

	wchar_t keyName [MAX_KEYNAME_SIZE];
	wchar_t subKeyName [MAX_KEYNAME_SIZE];
	wchar_t valueName [MAX_VALUE_SIZE];
	LONG ret, retSub, retVal;

	HKEY mTypeKey, majorTypeKey, subTypeKey;
	DWORD valueType, valueSize, valueNameSize;
	wchar_t *valueData;
	int mediaTypeCount = 0;
	MediaType *newType;
	CheckByteGroup *newGroup;

	ret = RegOpenKey (HKEY_CLASSES_ROOT, TEXT("Media Type"), &mTypeKey);
	if (ret != ERROR_SUCCESS)
		return -1;

	for (int i = 0; ret != ERROR_NO_MORE_ITEMS; i ++)
	{
		ret = RegEnumKey (mTypeKey, i, keyName, MAX_KEYNAME_SIZE);

		if (ret != ERROR_SUCCESS || !lstrcmp (keyName, TEXT("Extensions")))	// we don't want the Extensions subkey
			continue;

		retSub = RegOpenKey (mTypeKey, keyName, &majorTypeKey);

		if (retSub != ERROR_SUCCESS)
			continue;

		for (int j = 0; retSub != ERROR_NO_MORE_ITEMS; j ++)
		{
			retSub = RegEnumKey (majorTypeKey, j, subKeyName, MAX_KEYNAME_SIZE);
			if (retSub != ERROR_SUCCESS)
				continue;

			retVal = RegOpenKey (majorTypeKey, subKeyName, &subTypeKey);

			newType = new MediaType;
			if (!newType)
			{
				ErrorMsg (0, L"Out of memory.");
				return -1;
			}

			CLSIDFromString (keyName, &newType->majorType);
			CLSIDFromString (subKeyName, &newType->subType);

			for (int k = 0; retVal != ERROR_NO_MORE_ITEMS; k ++)
			{
				valueNameSize = MAX_VALUE_SIZE;
				retVal = RegEnumValue (subTypeKey, k, valueName, &valueNameSize, NULL, &valueType, NULL, &valueSize);

				if (retVal != ERROR_SUCCESS || valueType != REG_SZ || !lstrcmp (valueName, TEXT("Source Filter")))	// we don't need the source filter value
					continue;

				valueData = (wchar_t *) new BYTE [valueSize];

				valueNameSize = MAX_VALUE_SIZE;
				retVal = RegEnumValue (subTypeKey, k, valueName, &valueNameSize, NULL, &valueType,
					(LPBYTE) valueData, &valueSize);

				if (retVal != ERROR_SUCCESS)
				{
					delete [] valueData;
					continue;
				}

				newGroup = new CheckByteGroup;	// each value is one group

				if (!newGroup)
				{
					delete [] valueData;
					ErrorMsg (0, L"Out of memory.");
					return -1;
				}

				newGroup->checkByteCount = parseCheckBytes (valueData, &newGroup->checkBytes);

				if (newGroup->checkByteCount == -1)
				{
					delete [] valueData;
					delete newGroup;

					return -1;	// this means out of memory
				}

				if (!newGroup->checkByteCount)
				{
					// if we get here there was a parse error or an invalid value in the registry
					delete newGroup;
				}
				else
				{
					newType->checkByteGroups.InsertLast (newGroup);
					newType->checkByteGroupCount ++;
				}

				delete [] valueData;
			}

			if (!newType->checkByteGroupCount)
			{
				// if we get here we were unable to parse any of the values in the current subkey
				delete newType;
			}
			else
			{
				mediaTypeList->InsertLast (newType);
				mediaTypeCount ++;
			}
		}
	}
	return mediaTypeCount;
}

int checkFileForMediaType (File *file, List<MediaType> *mediaTypeList, MediaType **foundMediaType)
{
	MediaType *mt;
	CheckByteGroup *cbg;
	bool matches;
	BOOL ret;
	LONG lBytesRead;
	LONGLONG actOffset;
	BYTE *necessaryBytes;

	mt = mediaTypeList->First ();

	while (mt != NULL)
	{							// loop over all filetypes
		cbg = mt->checkByteGroups.First ();

		while (cbg != NULL)
		{						// loop over all groups of checkbytes
			matches = true;

			for (unsigned int i = 0; i < cbg->checkByteCount; i ++)
			{					// we need to match all fields in one group
				if (cbg->checkBytes [i].offset < 0)
					actOffset = file->size + cbg->checkBytes [i].offset;
				else
					actOffset = cbg->checkBytes [i].offset;

				if (actOffset > file->size || actOffset < 0)
				{
					matches = false;
					break;
				}

				necessaryBytes = new BYTE [cbg->checkBytes [i].byteCount];

				if (!necessaryBytes)
				{
					ErrorMsg (0, L"Out of memory.");
					return 0;
				}

				// read the necessary amount of bytes to compare to value (after masking)
				ret = file->SyncRead (actOffset, cbg->checkBytes [i].byteCount, necessaryBytes, &lBytesRead);

				if (ret != S_OK)
				{
					matches = false;
					delete [] necessaryBytes;
					break;
				}

				// mask and compare all bytes in this entry
				for (unsigned int j = 0; j < cbg->checkBytes [i].byteCount; j ++)
				{
					if ((necessaryBytes [j] & cbg->checkBytes [i].mask [j]) != cbg->checkBytes [i].value [j])
					{
						matches = false;
						break;
					}
				}
				delete [] necessaryBytes;

				if (!matches)
					break;
			}

			if (matches)		// one group is enought to match the filetype
			{
				*foundMediaType = mt;
				return 1;
			}
			cbg = mt->checkByteGroups.Next (cbg);
		}
		mt = mediaTypeList->Next (mt);
	}

	*foundMediaType = NULL;
	return 1;
}