clamav/libclamav/swf.c

568 lines
21 KiB
C
Raw Normal View History

/*
2020-01-03 15:44:07 -05:00
* Copyright (C) 2013-2020 Cisco Systems, Inc. and/or its affiliates. All rights reserved.
* Copyright (C) 2011-2013 Sourcefire, Inc.
*
* The code is based on Flasm, command line assembler & disassembler of Flash
* ActionScript bytecode Copyright (c) 2001 Opaque Industries, (c) 2002-2007
* Igor Kogan, (c) 2005 Wang Zhen. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without modification,
* are permitted provided that the following conditions are met:
*
* - Redistributions of source code must retain the above copyright notice, this list
* of conditions and the following disclaimer.
* - Redistributions in binary form must reproduce the above copyright notice, this
* list of conditions and the following disclaimer in the documentation and/or other
* materials provided with the distribution.
* - Neither the name of the Opaque Industries nor the names of its contributors may
* be used to endorse or promote products derived from this software without specific
* prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY
* EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT
* SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
* TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
* BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY
* WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#if HAVE_CONFIG_H
#include "clamav-config.h"
#endif
#include <stdio.h>
#include <string.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <sys/stat.h>
#ifdef HAVE_UNISTD_H
#include <unistd.h>
#endif
#include <time.h>
#include <zlib.h>
#include "swf.h"
#include "clamav.h"
2011-04-06 15:53:28 +02:00
#include "scanners.h"
#include "lzma_iface.h"
#define EC16(v) le16_to_host(v)
#define EC32(v) le32_to_host(v)
#define INITBITS \
{ \
if (fmap_readn(map, &get_c, offset, sizeof(get_c)) == sizeof(get_c)) { \
bitpos = 8; \
bitbuf = (unsigned int)get_c; \
offset += sizeof(get_c); \
} else { \
cli_warnmsg("cli_scanswf: INITBITS: Can't read file or file truncated\n"); \
return CL_EFORMAT; \
} \
}
#define GETBITS(v, n) \
{ \
getbits_n = n; \
bits = 0; \
while (getbits_n > bitpos) { \
getbits_n -= bitpos; \
bits |= bitbuf << getbits_n; \
if (fmap_readn(map, &get_c, offset, sizeof(get_c)) == sizeof(get_c)) { \
bitbuf = (unsigned int)get_c; \
bitpos = 8; \
offset += sizeof(get_c); \
} else { \
cli_warnmsg("cli_scanswf: GETBITS: Can't read file or file truncated\n"); \
return CL_EFORMAT; \
} \
} \
bitpos -= getbits_n; \
bits |= bitbuf >> bitpos; \
bitbuf &= 0xff >> (8 - bitpos); \
v = bits & 0xffff; \
}
#define GETWORD(v) \
{ \
if (fmap_readn(map, &get_c, offset, sizeof(get_c)) == sizeof(get_c)) { \
getword_1 = (unsigned int)get_c; \
offset += sizeof(get_c); \
} else { \
cli_warnmsg("cli_scanswf: GETWORD: Can't read file or file truncated\n"); \
return CL_EFORMAT; \
} \
if (fmap_readn(map, &get_c, offset, sizeof(get_c)) == sizeof(get_c)) { \
getword_2 = (unsigned int)get_c; \
offset += sizeof(get_c); \
} else { \
cli_warnmsg("cli_scanswf: GETWORD: Can't read file or file truncated\n"); \
return CL_EFORMAT; \
} \
v = (uint16_t)(getword_1 & 0xff) | ((getword_2 & 0xff) << 8); \
}
#define GETDWORD(v) \
{ \
GETWORD(getdword_1); \
GETWORD(getdword_2); \
v = (uint32_t)(getdword_1 | (getdword_2 << 16)); \
}
struct swf_file_hdr {
char signature[3];
uint8_t version;
uint32_t filesize;
};
static int scanzws(cli_ctx *ctx, struct swf_file_hdr *hdr)
{
struct CLI_LZMA lz;
unsigned char inbuff[FILEBUFF], outbuff[FILEBUFF];
fmap_t *map = *ctx->fmap;
/* strip off header */
off_t offset = 8;
uint32_t d_insize;
size_t outsize = 8;
int ret, lret;
size_t count;
char *tmpname;
int fd;
Record names of extracted files A way is needed to record scanned file names for two purposes: 1. File names (and extensions) must be stored in the json metadata properties recorded when using the --gen-json clamscan option. Future work may use this to compare file extensions with detected file types. 2. File names are useful when interpretting tmp directory output when using the --leave-temps option. This commit enables file name retention for later use by storing file names in the fmap header structure, if a file name exists. To store the names in fmaps, an optional name argument has been added to any internal scan API's that create fmaps and every call to these APIs has been modified to pass a file name or NULL if a file name is not required. The zip and gpt parsers required some modification to record file names. The NSIS and XAR parsers fail to collect file names at all and will require future work to support file name extraction. Also: - Added recursive extraction to the tmp directory when the --leave-temps option is enabled. When not enabled, the tmp directory structure remains flat so as to prevent the likelihood of exceeding MAX_PATH. The current tmp directory is stored in the scan context. - Made the cli_scanfile() internal API non-static and added it to scanners.h so it would be accessible outside of scanners.c in order to remove code duplication within libmspack.c. - Added function comments to scanners.h and matcher.h - Converted a TDB-type macros and LSIG-type macros to enums for improved type safey. - Converted more return status variables from `int` to `cl_error_t` for improved type safety, and corrected ooxml file typing functions so they use `cli_file_t` exclusively rather than mixing types with `cl_error_t`. - Restructured the magic_scandesc() function to use goto's for error handling and removed the early_ret_from_magicscan() macro and magic_scandesc_cleanup() function. This makes the code easier to read and made it easier to add the recursive tmp directory cleanup to magic_scandesc(). - Corrected zip, egg, rar filename extraction issues. - Removed use of extra sub-directory layer for zip, egg, and rar file extraction. For Zip, this also involved changing the extracted filenames to be randomly generated rather than using the "zip.###" file name scheme.
2020-03-19 21:23:54 -04:00
if ((ret = cli_gentempfd(ctx->sub_tmpdir, &tmpname, &fd)) != CL_SUCCESS) {
cli_errmsg("scanzws: Can't generate temporary file\n");
return ret;
}
hdr->signature[0] = 'F';
if (cli_writen(fd, hdr, sizeof(struct swf_file_hdr)) != sizeof(struct swf_file_hdr)) {
cli_errmsg("scanzws: Can't write to file %s\n", tmpname);
close(fd);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EWRITE;
}
/* read 4 bytes (for compressed 32-bit filesize) [not used for LZMA] */
if (fmap_readn(map, &d_insize, offset, sizeof(d_insize)) != sizeof(d_insize)) {
cli_errmsg("scanzws: Error reading SWF file\n");
close(fd);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EREAD;
}
offset += sizeof(d_insize);
/* check if declared input size matches actual output size */
/* map->len = header (8 bytes) + d_insize (4 bytes) + flags (5 bytes) + compressed stream */
if (d_insize != (map->len - 17)) {
cli_warnmsg("SWF: declared input length != compressed stream size, %u != %llu\n",
d_insize, (long long unsigned)(map->len - 17));
} else {
cli_dbgmsg("SWF: declared input length == compressed stream size, %u == %llu\n",
d_insize, (long long unsigned)(map->len - 17));
}
/* first buffer required for initializing LZMA */
ret = fmap_readn(map, inbuff, offset, FILEBUFF);
if (ret < 0) {
cli_errmsg("scanzws: Error reading SWF file\n");
close(fd);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EUNPACK;
}
/* nothing written, likely truncated */
if (!ret) {
cli_errmsg("scanzws: possibly truncated file\n");
close(fd);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EFORMAT;
}
offset += ret;
memset(&lz, 0, sizeof(lz));
lz.next_in = inbuff;
lz.next_out = outbuff;
lz.avail_in = ret;
lz.avail_out = FILEBUFF;
lret = cli_LzmaInit(&lz, hdr->filesize);
if (lret != LZMA_RESULT_OK) {
cli_errmsg("scanzws: LzmaInit() failed\n");
close(fd);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EUNPACK;
}
while (lret == LZMA_RESULT_OK) {
if (lz.avail_in == 0) {
lz.next_in = inbuff;
ret = fmap_readn(map, inbuff, offset, FILEBUFF);
if (ret < 0) {
cli_errmsg("scanzws: Error reading SWF file\n");
cli_LzmaShutdown(&lz);
close(fd);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EUNPACK;
}
if (!ret)
break;
lz.avail_in = ret;
offset += ret;
}
lret = cli_LzmaDecode(&lz);
count = FILEBUFF - lz.avail_out;
if (count) {
if (cli_checklimits("SWF", ctx, outsize + count, 0, 0) != CL_SUCCESS)
break;
if (cli_writen(fd, outbuff, count) != count) {
cli_errmsg("scanzws: Can't write to file %s\n", tmpname);
cli_LzmaShutdown(&lz);
close(fd);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EWRITE;
}
outsize += count;
}
lz.next_out = outbuff;
lz.avail_out = FILEBUFF;
}
cli_LzmaShutdown(&lz);
if (lret != LZMA_STREAM_END && lret != LZMA_RESULT_OK) {
/* outsize starts at 8, therefore, if its still 8, nothing was decompressed */
if (outsize == 8) {
cli_infomsg(ctx, "scanzws: Error decompressing SWF file. No data decompressed.\n");
close(fd);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EUNPACK;
}
cli_infomsg(ctx, "scanzws: Error decompressing SWF file. Scanning what was decompressed.\n");
}
cli_dbgmsg("SWF: Decompressed[LZMA] to %s, size %llu\n", tmpname, (long long unsigned)outsize);
/* check if declared output size matches actual output size */
if (hdr->filesize != outsize) {
cli_warnmsg("SWF: declared output length != inflated stream size, %u != %llu\n",
hdr->filesize, (long long unsigned)outsize);
} else {
cli_dbgmsg("SWF: declared output length == inflated stream size, %u == %llu\n",
hdr->filesize, (long long unsigned)outsize);
}
ret = cli_magic_scan_desc(fd, tmpname, ctx, NULL);
close(fd);
if (!(ctx->engine->keeptmp)) {
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
}
free(tmpname);
return ret;
}
static int scancws(cli_ctx *ctx, struct swf_file_hdr *hdr)
{
z_stream stream;
char inbuff[FILEBUFF], outbuff[FILEBUFF];
fmap_t *map = *ctx->fmap;
int offset = 8, ret, zret, zend;
size_t outsize = 8;
size_t count;
char *tmpname;
int fd;
Record names of extracted files A way is needed to record scanned file names for two purposes: 1. File names (and extensions) must be stored in the json metadata properties recorded when using the --gen-json clamscan option. Future work may use this to compare file extensions with detected file types. 2. File names are useful when interpretting tmp directory output when using the --leave-temps option. This commit enables file name retention for later use by storing file names in the fmap header structure, if a file name exists. To store the names in fmaps, an optional name argument has been added to any internal scan API's that create fmaps and every call to these APIs has been modified to pass a file name or NULL if a file name is not required. The zip and gpt parsers required some modification to record file names. The NSIS and XAR parsers fail to collect file names at all and will require future work to support file name extraction. Also: - Added recursive extraction to the tmp directory when the --leave-temps option is enabled. When not enabled, the tmp directory structure remains flat so as to prevent the likelihood of exceeding MAX_PATH. The current tmp directory is stored in the scan context. - Made the cli_scanfile() internal API non-static and added it to scanners.h so it would be accessible outside of scanners.c in order to remove code duplication within libmspack.c. - Added function comments to scanners.h and matcher.h - Converted a TDB-type macros and LSIG-type macros to enums for improved type safey. - Converted more return status variables from `int` to `cl_error_t` for improved type safety, and corrected ooxml file typing functions so they use `cli_file_t` exclusively rather than mixing types with `cl_error_t`. - Restructured the magic_scandesc() function to use goto's for error handling and removed the early_ret_from_magicscan() macro and magic_scandesc_cleanup() function. This makes the code easier to read and made it easier to add the recursive tmp directory cleanup to magic_scandesc(). - Corrected zip, egg, rar filename extraction issues. - Removed use of extra sub-directory layer for zip, egg, and rar file extraction. For Zip, this also involved changing the extracted filenames to be randomly generated rather than using the "zip.###" file name scheme.
2020-03-19 21:23:54 -04:00
if ((ret = cli_gentempfd(ctx->sub_tmpdir, &tmpname, &fd)) != CL_SUCCESS) {
cli_errmsg("scancws: Can't generate temporary file\n");
return ret;
}
hdr->signature[0] = 'F';
if (cli_writen(fd, hdr, sizeof(struct swf_file_hdr)) != sizeof(struct swf_file_hdr)) {
cli_errmsg("scancws: Can't write to file %s\n", tmpname);
2011-04-15 12:39:45 +02:00
close(fd);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EWRITE;
}
stream.avail_in = 0;
stream.next_in = (Bytef *)inbuff;
stream.next_out = (Bytef *)outbuff;
stream.zalloc = (alloc_func)NULL;
stream.zfree = (free_func)NULL;
stream.opaque = (voidpf)0;
stream.avail_out = FILEBUFF;
zret = inflateInit(&stream);
if (zret != Z_OK) {
cli_errmsg("scancws: inflateInit() failed\n");
close(fd);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EUNPACK;
}
do {
if (stream.avail_in == 0) {
stream.next_in = (Bytef *)inbuff;
ret = fmap_readn(map, inbuff, offset, FILEBUFF);
if (ret < 0) {
cli_errmsg("scancws: Error reading SWF file\n");
close(fd);
inflateEnd(&stream);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EUNPACK;
}
if (!ret)
break;
stream.avail_in = ret;
offset += ret;
}
zret = inflate(&stream, Z_SYNC_FLUSH);
count = FILEBUFF - stream.avail_out;
if (count) {
if (cli_checklimits("SWF", ctx, outsize + count, 0, 0) != CL_SUCCESS)
break;
if (cli_writen(fd, outbuff, count) != count) {
cli_errmsg("scancws: Can't write to file %s\n", tmpname);
inflateEnd(&stream);
close(fd);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EWRITE;
}
outsize += count;
}
stream.next_out = (Bytef *)outbuff;
stream.avail_out = FILEBUFF;
} while (zret == Z_OK);
2014-11-04 18:38:34 -05:00
zend = inflateEnd(&stream);
if ((zret != Z_STREAM_END && zret != Z_OK) || zend != Z_OK) {
/*
* outsize is initialized to 8, it being 8 here means that we couldn't even read a single byte.
* If outsize > 8, then we have data. Let's scan what we have.
*/
if (outsize == 8) {
cli_infomsg(ctx, "scancws: Error decompressing SWF file. No data decompressed.\n");
close(fd);
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
free(tmpname);
return CL_EUNPACK;
}
cli_infomsg(ctx, "scancws: Error decompressing SWF file. Scanning what was decompressed.\n");
}
cli_dbgmsg("SWF: Decompressed[zlib] to %s, size %zu\n", tmpname, outsize);
/* check if declared output size matches actual output size */
if (hdr->filesize != outsize) {
cli_warnmsg("SWF: declared output length != inflated stream size, %u != %zu\n",
hdr->filesize, outsize);
} else {
cli_dbgmsg("SWF: declared output length == inflated stream size, %u == %zu\n",
hdr->filesize, outsize);
}
ret = cli_magic_scan_desc(fd, tmpname, ctx, NULL);
close(fd);
if (!ctx->engine->keeptmp) {
if (cli_unlink(tmpname)) {
free(tmpname);
return CL_EUNLINK;
}
}
free(tmpname);
return ret;
}
static const char *tagname(tag_id id)
{
unsigned int i;
for (i = 0; tag_names[i].name; i++)
if (tag_names[i].id == id)
return tag_names[i].name;
return NULL;
}
int cli_scanswf(cli_ctx *ctx)
{
struct swf_file_hdr file_hdr;
fmap_t *map = *ctx->fmap;
unsigned int bitpos, bitbuf, getbits_n, nbits, getword_1, getword_2, getdword_1, getdword_2;
const char *pt;
unsigned char get_c;
size_t offset = 0;
unsigned int val, foo, tag_hdr, tag_type, tag_len;
unsigned long int bits;
cli_dbgmsg("in cli_scanswf()\n");
if (fmap_readn(map, &file_hdr, offset, sizeof(file_hdr)) != sizeof(file_hdr)) {
cli_dbgmsg("SWF: Can't read file header\n");
return CL_CLEAN;
}
offset += sizeof(file_hdr);
/*
** SWF stores the integer bytes with the least significate byte first
*/
file_hdr.filesize = le32_to_host(file_hdr.filesize);
cli_dbgmsg("SWF: Version: %u\n", file_hdr.version);
cli_dbgmsg("SWF: File size: %u\n", file_hdr.filesize);
if (!strncmp(file_hdr.signature, "CWS", 3)) {
cli_dbgmsg("SWF: zlib compressed file\n");
return scancws(ctx, &file_hdr);
} else if (!strncmp(file_hdr.signature, "ZWS", 3)) {
cli_dbgmsg("SWF: LZMA compressed file\n");
return scanzws(ctx, &file_hdr);
} else if (!strncmp(file_hdr.signature, "FWS", 3)) {
cli_dbgmsg("SWF: Uncompressed file\n");
} else {
cli_dbgmsg("SWF: Not a SWF file\n");
return CL_CLEAN;
}
INITBITS;
GETBITS(nbits, 5);
cli_dbgmsg("SWF: FrameSize RECT size bits: %u\n", nbits);
{
uint32_t xMin = 0, xMax = 0, yMin = 0, yMax = 0;
GETBITS(xMin, nbits); /* Should be zero */
GETBITS(xMax, nbits);
GETBITS(yMin, nbits); /* Should be zero */
GETBITS(yMax, nbits);
cli_dbgmsg("SWF: FrameSize xMin %u xMax %u yMin %u yMax %u\n", xMin, xMax, yMin, yMax);
}
GETWORD(foo);
GETWORD(val);
cli_dbgmsg("SWF: Frames total: %d\n", val);
2014-05-23 11:54:02 -04:00
/* Skip Flash tag walk unless debug mode */
if (!cli_debug_flag) {
2014-05-23 11:54:02 -04:00
return CL_CLEAN;
}
while (offset < map->len) {
GETWORD(tag_hdr);
tag_type = tag_hdr >> 6;
if (tag_type == 0)
break;
tag_len = tag_hdr & 0x3f;
if (tag_len == 0x3f)
GETDWORD(tag_len);
pt = tagname(tag_type);
cli_dbgmsg("SWF: %s\n", pt ? pt : "UNKNOWN TAG");
cli_dbgmsg("SWF: Tag length: %u\n", tag_len);
if (tag_len > map->len) {
cli_dbgmsg("SWF: Invalid tag length.\n");
return CL_EFORMAT;
}
if ((offset + tag_len) < offset) {
cli_warnmsg("SWF: Tag length too large.\n");
break;
}
if (!pt) {
offset += tag_len;
continue;
}
switch (tag_type) {
case TAG_SCRIPTLIMITS: {
unsigned int recursion, timeout;
GETWORD(recursion);
GETWORD(timeout);
cli_dbgmsg("SWF: scriptLimits recursion %u timeout %u\n", recursion, timeout);
break;
}
case TAG_FILEATTRIBUTES:
GETDWORD(val);
cli_dbgmsg("SWF: File attributes:\n");
if (val & SWF_ATTR_USENETWORK)
cli_dbgmsg(" * Use network\n");
if (val & SWF_ATTR_RELATIVEURLS)
cli_dbgmsg(" * Relative URLs\n");
if (val & SWF_ATTR_SUPPRESSCROSSDOMAINCACHE)
cli_dbgmsg(" * Suppress cross domain cache\n");
if (val & SWF_ATTR_ACTIONSCRIPT3)
cli_dbgmsg(" * ActionScript 3.0\n");
if (val & SWF_ATTR_HASMETADATA)
cli_dbgmsg(" * Has metadata\n");
if (val & SWF_ATTR_USEDIRECTBLIT)
cli_dbgmsg(" * Use hardware acceleration\n");
if (val & SWF_ATTR_USEGPU)
cli_dbgmsg(" * Use GPU\n");
break;
default:
offset += tag_len;
continue;
}
}
return CL_CLEAN;
}