git-svn: trunk@2992
Tomasz Kojm authored on 2007/03/29 06:38:07... | ... |
@@ -1,3 +1,7 @@ |
1 |
+Wed Mar 28 21:45:12 CEST 2007 (tk) |
|
2 |
+---------------------------------- |
|
3 |
+ * libclamav: make some cleanups and add support for nibble matching |
|
4 |
+ |
|
1 | 5 |
Tue Mar 27 22:05:28 BST 2007 (njh) |
2 | 6 |
---------------------------------- |
3 | 7 |
* clamav-milter.c: Added IPv6 support, based on a patch by |
... | ... |
@@ -109,32 +109,6 @@ extern "C" |
109 | 109 |
#define cl_perror cl_strerror |
110 | 110 |
|
111 | 111 |
/* internal structures */ |
112 |
-struct cli_bm_patt { |
|
113 |
- unsigned char *pattern; |
|
114 |
- char *virname, *offset; |
|
115 |
- const char *viralias; |
|
116 |
- unsigned int length; |
|
117 |
- unsigned short target; |
|
118 |
- struct cli_bm_patt *next; |
|
119 |
-}; |
|
120 |
- |
|
121 |
-struct cli_ac_patt { |
|
122 |
- short int *pattern, *prefix; |
|
123 |
- unsigned int length, mindist, maxdist, prefix_length; |
|
124 |
- char *virname, *offset; |
|
125 |
- const char *viralias; |
|
126 |
- unsigned short int sigid, parts, partno, alt, *altn, alt_pattern; |
|
127 |
- unsigned short type, target; |
|
128 |
- unsigned char **altc; |
|
129 |
- struct cli_ac_patt *next; |
|
130 |
-}; |
|
131 |
- |
|
132 |
-struct cli_ac_node { |
|
133 |
- unsigned char islast; |
|
134 |
- struct cli_ac_patt *list; |
|
135 |
- struct cli_ac_node *trans[256], *fail; |
|
136 |
-}; |
|
137 |
- |
|
138 | 112 |
struct cli_md5_node { |
139 | 113 |
char *virname, *viralias; |
140 | 114 |
unsigned char *md5; |
... | ... |
@@ -150,20 +124,6 @@ struct cli_meta_node { |
150 | 150 |
struct cli_meta_node *next; |
151 | 151 |
}; |
152 | 152 |
|
153 |
-struct cli_matcher { |
|
154 |
- unsigned int maxpatlen; /* maximal length of pattern in db */ |
|
155 |
- unsigned short ac_only; |
|
156 |
- |
|
157 |
- /* Extended Boyer-Moore */ |
|
158 |
- int *bm_shift; |
|
159 |
- struct cli_bm_patt **bm_suffix; |
|
160 |
- |
|
161 |
- /* Extended Aho-Corasick */ |
|
162 |
- unsigned int ac_depth; |
|
163 |
- struct cli_ac_node *ac_root, **ac_nodetable; |
|
164 |
- unsigned int ac_partsigs, ac_nodes; |
|
165 |
-}; |
|
166 |
- |
|
167 | 153 |
struct cl_engine { |
168 | 154 |
unsigned int refcount; /* reference counter */ |
169 | 155 |
unsigned short ncore; |
... | ... |
@@ -171,7 +131,7 @@ struct cl_engine { |
171 | 171 |
unsigned int dboptions; |
172 | 172 |
|
173 | 173 |
/* Roots table */ |
174 |
- struct cli_matcher **root; |
|
174 |
+ void **root; |
|
175 | 175 |
|
176 | 176 |
/* MD5 */ |
177 | 177 |
struct cli_md5_node **md5_hlist; |
178 | 178 |
deleted file mode 100644 |
... | ... |
@@ -1,25 +0,0 @@ |
1 |
-/* |
|
2 |
- * Copyright (C) 2002 - 2005 Tomasz Kojm <tkojm@clamav.net> |
|
3 |
- * |
|
4 |
- * This program is free software; you can redistribute it and/or modify |
|
5 |
- * it under the terms of the GNU General Public License as published by |
|
6 |
- * the Free Software Foundation; either version 2 of the License, or |
|
7 |
- * (at your option) any later version. |
|
8 |
- * |
|
9 |
- * This program is distributed in the hope that it will be useful, |
|
10 |
- * but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
11 |
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
12 |
- * GNU General Public License for more details. |
|
13 |
- * |
|
14 |
- * You should have received a copy of the GNU General Public License |
|
15 |
- * along with this program; if not, write to the Free Software |
|
16 |
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, |
|
17 |
- * MA 02110-1301, USA. |
|
18 |
- */ |
|
19 |
- |
|
20 |
-#ifndef DATADIR |
|
21 |
-# define DATADIR "/usr/local/share/clamav" |
|
22 |
-#endif |
|
23 |
- |
|
24 |
-#define CLI_IGN -200 |
|
25 |
-#define CLI_ALT -201 |
... | ... |
@@ -382,13 +382,6 @@ int cli_addtypesigs(struct cl_engine *engine) |
382 | 382 |
return CL_EMEM; |
383 | 383 |
} |
384 | 384 |
|
385 |
- if(engine->ncore) { |
|
386 |
- /* |
|
387 |
- cli_dbgmsg("cli_addtypesigs: AC depth 10 (ncore mode)\n"); |
|
388 |
- cli_ac_setdepth(10); |
|
389 |
- */ |
|
390 |
- } |
|
391 |
- |
|
392 | 385 |
root->ac_root = (struct cli_ac_node *) cli_calloc(1, sizeof(struct cli_ac_node)); |
393 | 386 |
if(!root->ac_root) { |
394 | 387 |
cli_errmsg("cli_addtypesigs: Can't initialise AC pattern matcher\n"); |
... | ... |
@@ -4,7 +4,7 @@ |
4 | 4 |
* http://www-sr.informatik.uni-tuebingen.de/~buehler/AC/AC.html |
5 | 5 |
* Thanks to Kurt Huwig for pointing me to this page. |
6 | 6 |
* |
7 |
- * Copyright (C) 2002 - 2006 Tomasz Kojm <tkojm@clamav.net> |
|
7 |
+ * Copyright (C) 2002 - 2007 Tomasz Kojm <tkojm@clamav.net> |
|
8 | 8 |
* |
9 | 9 |
* This program is free software; you can redistribute it and/or modify |
10 | 10 |
* it under the terms of the GNU General Public License as published by |
... | ... |
@@ -37,7 +37,6 @@ |
37 | 37 |
#include "others.h" |
38 | 38 |
#include "matcher.h" |
39 | 39 |
#include "matcher-ac.h" |
40 |
-#include "defaults.h" |
|
41 | 40 |
#include "filetypes.h" |
42 | 41 |
#include "cltypes.h" |
43 | 42 |
|
... | ... |
@@ -46,12 +45,12 @@ struct nodelist { |
46 | 46 |
struct nodelist *next; |
47 | 47 |
}; |
48 | 48 |
|
49 |
-unsigned short ac_depth = AC_DEFAULT_DEPTH; |
|
49 |
+static uint8_t ac_depth = AC_DEFAULT_DEPTH; |
|
50 | 50 |
|
51 | 51 |
int cli_ac_addpatt(struct cli_matcher *root, struct cli_ac_patt *pattern) |
52 | 52 |
{ |
53 | 53 |
struct cli_ac_node *pos, *next; |
54 |
- int i; |
|
54 |
+ uint8_t i; |
|
55 | 55 |
|
56 | 56 |
if(pattern->length < ac_depth) |
57 | 57 |
return CL_EPATSHORT; |
... | ... |
@@ -59,7 +58,7 @@ int cli_ac_addpatt(struct cli_matcher *root, struct cli_ac_patt *pattern) |
59 | 59 |
pos = root->ac_root; |
60 | 60 |
|
61 | 61 |
for(i = 0; i < ac_depth; i++) { |
62 |
- next = pos->trans[((unsigned char) pattern->pattern[i]) & 0xff]; |
|
62 |
+ next = pos->trans[(unsigned char) (pattern->pattern[i] & 0xff)]; |
|
63 | 63 |
|
64 | 64 |
if(!next) { |
65 | 65 |
next = (struct cli_ac_node *) cli_calloc(1, sizeof(struct cli_ac_node)); |
... | ... |
@@ -248,7 +247,7 @@ inline static int cli_findpos(const unsigned char *buffer, unsigned int depth, u |
248 | 248 |
if(bufferpos == postfixend) |
249 | 249 |
return 0; |
250 | 250 |
|
251 |
- if(pattern->pattern[i] == CLI_ALT) { |
|
251 |
+ if((pattern->pattern[i] & CLI_MATCH_WILDCARD) == CLI_MATCH_ALTERNATIVE) { |
|
252 | 252 |
found = 0; |
253 | 253 |
for(j = 0; j < pattern->altn[alt]; j++) { |
254 | 254 |
if(pattern->altc[alt][j] == buffer[bufferpos]) { |
... | ... |
@@ -261,7 +260,15 @@ inline static int cli_findpos(const unsigned char *buffer, unsigned int depth, u |
261 | 261 |
return 0; |
262 | 262 |
alt++; |
263 | 263 |
|
264 |
- } else if(pattern->pattern[i] != CLI_IGN && (unsigned char) pattern->pattern[i] != buffer[bufferpos]) |
|
264 |
+ } else if((pattern->pattern[i] & CLI_MATCH_WILDCARD) == CLI_MATCH_NIBBLE_HIGH) { |
|
265 |
+ if((unsigned char) (pattern->pattern[i] & 0x00f0) != (buffer[bufferpos] & 0xf0)) |
|
266 |
+ return 0; |
|
267 |
+ |
|
268 |
+ } else if((pattern->pattern[i] & CLI_MATCH_WILDCARD) == CLI_MATCH_NIBBLE_LOW) { |
|
269 |
+ if((unsigned char) (pattern->pattern[i] & 0x000f) != (buffer[bufferpos] & 0x0f)) |
|
270 |
+ return 0; |
|
271 |
+ |
|
272 |
+ } else if((pattern->pattern[i] & CLI_MATCH_WILDCARD) != CLI_MATCH_IGNORE && (unsigned char) pattern->pattern[i] != buffer[bufferpos]) |
|
265 | 273 |
return 0; |
266 | 274 |
|
267 | 275 |
bufferpos++; |
... | ... |
@@ -276,7 +283,7 @@ inline static int cli_findpos(const unsigned char *buffer, unsigned int depth, u |
276 | 276 |
|
277 | 277 |
for(i = 0; i < pattern->prefix_length; i++) { |
278 | 278 |
|
279 |
- if(pattern->prefix[i] == CLI_ALT) { |
|
279 |
+ if((pattern->prefix[i] & CLI_MATCH_WILDCARD) == CLI_MATCH_ALTERNATIVE) { |
|
280 | 280 |
found = 0; |
281 | 281 |
for(j = 0; j < pattern->altn[alt]; j++) { |
282 | 282 |
if(pattern->altc[alt][j] == buffer[bufferpos]) { |
... | ... |
@@ -289,7 +296,15 @@ inline static int cli_findpos(const unsigned char *buffer, unsigned int depth, u |
289 | 289 |
return 0; |
290 | 290 |
alt++; |
291 | 291 |
|
292 |
- } else if(pattern->prefix[i] != CLI_IGN && (unsigned char) pattern->prefix[i] != buffer[bufferpos]) |
|
292 |
+ } else if((pattern->prefix[i] & CLI_MATCH_WILDCARD) == CLI_MATCH_NIBBLE_HIGH) { |
|
293 |
+ if((unsigned char) (pattern->prefix[i] & 0x00f0) != (buffer[bufferpos] & 0xf0)) |
|
294 |
+ return 0; |
|
295 |
+ |
|
296 |
+ } else if((pattern->prefix[i] & CLI_MATCH_WILDCARD) == CLI_MATCH_NIBBLE_LOW) { |
|
297 |
+ if((unsigned char) (pattern->prefix[i] & 0x000f) != (buffer[bufferpos] & 0x0f)) |
|
298 |
+ return 0; |
|
299 |
+ |
|
300 |
+ } else if(!(pattern->prefix[i] & CLI_MATCH_IGNORE) && (unsigned char) pattern->prefix[i] != buffer[bufferpos]) |
|
293 | 301 |
return 0; |
294 | 302 |
|
295 | 303 |
bufferpos++; |
... | ... |
@@ -299,7 +314,7 @@ inline static int cli_findpos(const unsigned char *buffer, unsigned int depth, u |
299 | 299 |
return 1; |
300 | 300 |
} |
301 | 301 |
|
302 |
-int cli_ac_initdata(struct cli_ac_data *data, unsigned int partsigs, unsigned int tracklen) |
|
302 |
+int cli_ac_initdata(struct cli_ac_data *data, uint32_t partsigs, uint8_t tracklen) |
|
303 | 303 |
{ |
304 | 304 |
unsigned int i, j; |
305 | 305 |
|
... | ... |
@@ -321,7 +336,7 @@ int cli_ac_initdata(struct cli_ac_data *data, unsigned int partsigs, unsigned in |
321 | 321 |
} |
322 | 322 |
memset(data->inioff, -1, partsigs * sizeof(off_t)); |
323 | 323 |
|
324 |
- data->partcnt = (unsigned int *) cli_calloc(partsigs, sizeof(unsigned int)); |
|
324 |
+ data->partcnt = (uint16_t *) cli_calloc(partsigs, sizeof(uint16_t)); |
|
325 | 325 |
|
326 | 326 |
if(!data->partcnt) { |
327 | 327 |
cli_errmsg("cli_ac_init(): unable to cli_calloc(%u, %u)\n", partsigs, sizeof(unsigned int)); |
... | ... |
@@ -348,7 +363,7 @@ int cli_ac_initdata(struct cli_ac_data *data, unsigned int partsigs, unsigned in |
348 | 348 |
return CL_EMEM; |
349 | 349 |
} |
350 | 350 |
|
351 |
- data->maxshift = (int *) cli_malloc(partsigs * sizeof(int)); |
|
351 |
+ data->maxshift = (int32_t *) cli_malloc(partsigs * sizeof(int32_t)); |
|
352 | 352 |
|
353 | 353 |
if(!data->maxshift) { |
354 | 354 |
cli_errmsg("cli_ac_init(): unable to cli_malloc(%u)\n", partsigs * sizeof(int)); |
... | ... |
@@ -359,9 +374,9 @@ int cli_ac_initdata(struct cli_ac_data *data, unsigned int partsigs, unsigned in |
359 | 359 |
return CL_EMEM; |
360 | 360 |
} |
361 | 361 |
|
362 |
- memset(data->maxshift, -1, partsigs * sizeof(int)); |
|
362 |
+ memset(data->maxshift, -1, partsigs * sizeof(int32_t)); |
|
363 | 363 |
|
364 |
- data->partoff = (unsigned int **) cli_calloc(partsigs, sizeof(unsigned int *)); |
|
364 |
+ data->partoff = (uint32_t **) cli_calloc(partsigs, sizeof(uint32_t *)); |
|
365 | 365 |
|
366 | 366 |
if(!data->partoff) { |
367 | 367 |
cli_errmsg("cli_ac_init(): unable to cli_calloc(%u, %u)\n", partsigs, sizeof(unsigned int)); |
... | ... |
@@ -379,7 +394,7 @@ int cli_ac_initdata(struct cli_ac_data *data, unsigned int partsigs, unsigned in |
379 | 379 |
*/ |
380 | 380 |
|
381 | 381 |
for(i = 0; i < partsigs; i++) { |
382 |
- data->partoff[i] = (unsigned int *) cli_calloc(tracklen, sizeof(unsigned int)); |
|
382 |
+ data->partoff[i] = (uint32_t *) cli_calloc(tracklen, sizeof(uint32_t)); |
|
383 | 383 |
|
384 | 384 |
if(!data->partoff[i]) { |
385 | 385 |
for(j = 0; j < i; j++) |
... | ... |
@@ -418,12 +433,12 @@ void cli_ac_freedata(struct cli_ac_data *data) |
418 | 418 |
} |
419 | 419 |
} |
420 | 420 |
|
421 |
-int cli_ac_scanbuff(const unsigned char *buffer, unsigned int length, const char **virname, const struct cli_matcher *root, struct cli_ac_data *mdata, unsigned short otfrec, unsigned long int offset, cli_file_t ftype, int fd, struct cli_matched_type **ftoffset) |
|
421 |
+int cli_ac_scanbuff(const unsigned char *buffer, uint32_t length, const char **virname, const struct cli_matcher *root, struct cli_ac_data *mdata, uint8_t otfrec, uint32_t offset, cli_file_t ftype, int fd, struct cli_matched_type **ftoffset) |
|
422 | 422 |
{ |
423 | 423 |
struct cli_ac_node *current; |
424 | 424 |
struct cli_ac_patt *pt; |
425 | 425 |
int type = CL_CLEAN, j; |
426 |
- unsigned int i, position, curroff; |
|
426 |
+ uint32_t i, position, curroff; |
|
427 | 427 |
uint8_t offnum, found; |
428 | 428 |
struct cli_matched_type *tnode, *tnode_last = NULL; |
429 | 429 |
struct cli_target_info info; |
... | ... |
@@ -614,8 +629,3 @@ int cli_ac_scanbuff(const unsigned char *buffer, unsigned int length, const char |
614 | 614 |
|
615 | 615 |
return otfrec ? type : CL_CLEAN; |
616 | 616 |
} |
617 |
- |
|
618 |
-void cli_ac_setdepth(unsigned int depth) |
|
619 |
-{ |
|
620 |
- ac_depth = depth; |
|
621 |
-} |
... | ... |
@@ -1,5 +1,5 @@ |
1 | 1 |
/* |
2 |
- * Copyright (C) 2002 - 2005 Tomasz Kojm <tkojm@clamav.net> |
|
2 |
+ * Copyright (C) 2002 - 2007 Tomasz Kojm <tkojm@clamav.net> |
|
3 | 3 |
* |
4 | 4 |
* This program is free software; you can redistribute it and/or modify |
5 | 5 |
* it under the terms of the GNU General Public License as published by |
... | ... |
@@ -22,8 +22,6 @@ |
22 | 22 |
|
23 | 23 |
#include <sys/types.h> |
24 | 24 |
|
25 |
-#include "clamav.h" |
|
26 |
-#include "matcher.h" |
|
27 | 25 |
#include "filetypes.h" |
28 | 26 |
#include "cltypes.h" |
29 | 27 |
|
... | ... |
@@ -31,21 +29,41 @@ |
31 | 31 |
#define AC_DEFAULT_TRACKLEN 8 |
32 | 32 |
|
33 | 33 |
struct cli_ac_data { |
34 |
- unsigned int partsigs; |
|
34 |
+ uint32_t partsigs; |
|
35 | 35 |
off_t *inioff; |
36 |
- unsigned int *partcnt; |
|
37 |
- unsigned int **partoff; |
|
36 |
+ uint16_t *partcnt; |
|
37 |
+ uint32_t **partoff; |
|
38 | 38 |
uint8_t *offcnt; |
39 | 39 |
uint8_t *offidx; |
40 |
- int *maxshift; |
|
40 |
+ int32_t *maxshift; |
|
41 |
+}; |
|
42 |
+ |
|
43 |
+struct cli_ac_patt { |
|
44 |
+ uint16_t *pattern, *prefix, length, prefix_length; |
|
45 |
+ uint32_t mindist, maxdist; |
|
46 |
+ char *virname, *offset; |
|
47 |
+ const char *viralias; |
|
48 |
+ uint32_t sigid; |
|
49 |
+ uint16_t parts, partno, alt, *altn, alt_pattern; |
|
50 |
+ uint8_t target; |
|
51 |
+ uint16_t type; |
|
52 |
+ unsigned char **altc; |
|
53 |
+ struct cli_ac_patt *next; |
|
41 | 54 |
}; |
42 | 55 |
|
56 |
+struct cli_ac_node { |
|
57 |
+ uint8_t islast; |
|
58 |
+ struct cli_ac_patt *list; |
|
59 |
+ struct cli_ac_node *trans[256], *fail; |
|
60 |
+}; |
|
61 |
+ |
|
62 |
+#include "matcher.h" |
|
63 |
+ |
|
43 | 64 |
int cli_ac_addpatt(struct cli_matcher *root, struct cli_ac_patt *pattern); |
44 |
-int cli_ac_initdata(struct cli_ac_data *data, unsigned int partsigs, unsigned int histlen); |
|
65 |
+int cli_ac_initdata(struct cli_ac_data *data, uint32_t partsigs, uint8_t tracklen); |
|
45 | 66 |
void cli_ac_freedata(struct cli_ac_data *data); |
46 |
-int cli_ac_scanbuff(const unsigned char *buffer, unsigned int length, const char **virname, const struct cli_matcher *root, struct cli_ac_data *mdata, unsigned short otfrec, unsigned long int offset, cli_file_t ftype, int fd, struct cli_matched_type **ftoffset); |
|
67 |
+int cli_ac_scanbuff(const unsigned char *buffer, uint32_t length, const char **virname, const struct cli_matcher *root, struct cli_ac_data *mdata, uint8_t otfrec, uint32_t offset, cli_file_t ftype, int fd, struct cli_matched_type **ftoffset); |
|
47 | 68 |
int cli_ac_buildtrie(struct cli_matcher *root); |
48 | 69 |
void cli_ac_free(struct cli_matcher *root); |
49 |
-void cli_ac_setdepth(unsigned int depth); |
|
50 | 70 |
|
51 | 71 |
#endif |
... | ... |
@@ -17,6 +17,12 @@ |
17 | 17 |
* MA 02110-1301, USA. |
18 | 18 |
*/ |
19 | 19 |
|
20 |
+#if HAVE_CONFIG_H |
|
21 |
+#include "clamav-config.h" |
|
22 |
+#endif |
|
23 |
+ |
|
24 |
+#include <stdio.h> |
|
25 |
+ |
|
20 | 26 |
#include "clamav.h" |
21 | 27 |
#include "memory.h" |
22 | 28 |
#include "others.h" |
... | ... |
@@ -129,7 +135,7 @@ void cli_bm_free(struct cli_matcher *root) |
129 | 129 |
} |
130 | 130 |
} |
131 | 131 |
|
132 |
-int cli_bm_scanbuff(const unsigned char *buffer, unsigned int length, const char **virname, const struct cli_matcher *root, unsigned long int offset, cli_file_t ftype, int fd) |
|
132 |
+int cli_bm_scanbuff(const unsigned char *buffer, uint32_t length, const char **virname, const struct cli_matcher *root, uint32_t offset, cli_file_t ftype, int fd) |
|
133 | 133 |
{ |
134 | 134 |
unsigned int i, j, shift, off, found = 0; |
135 | 135 |
int idxtest; |
... | ... |
@@ -20,14 +20,22 @@ |
20 | 20 |
#ifndef __MATCHER_BM_H |
21 | 21 |
#define __MATCHER_BM_H |
22 | 22 |
|
23 |
-#include "clamav.h" |
|
24 | 23 |
#include "matcher.h" |
25 |
-#include "matcher-bm.h" |
|
26 | 24 |
#include "filetypes.h" |
25 |
+#include "cltypes.h" |
|
26 |
+ |
|
27 |
+struct cli_bm_patt { |
|
28 |
+ unsigned char *pattern; |
|
29 |
+ uint32_t length; |
|
30 |
+ char *virname, *offset; |
|
31 |
+ const char *viralias; |
|
32 |
+ uint8_t target; |
|
33 |
+ struct cli_bm_patt *next; |
|
34 |
+}; |
|
27 | 35 |
|
28 | 36 |
int cli_bm_addpatt(struct cli_matcher *root, struct cli_bm_patt *pattern); |
29 | 37 |
int cli_bm_init(struct cli_matcher *root); |
30 |
-int cli_bm_scanbuff(const unsigned char *buffer, unsigned int length, const char **virname, const struct cli_matcher *root, unsigned long int offset, cli_file_t ftype, int fd); |
|
38 |
+int cli_bm_scanbuff(const unsigned char *buffer, uint32_t length, const char **virname, const struct cli_matcher *root, uint32_t offset, cli_file_t ftype, int fd); |
|
31 | 39 |
void cli_bm_free(struct cli_matcher *root); |
32 | 40 |
|
33 | 41 |
#endif |
... | ... |
@@ -41,6 +41,7 @@ |
41 | 41 |
#include "execs.h" |
42 | 42 |
#include "special.h" |
43 | 43 |
#include "str.h" |
44 |
+#include "cltypes.h" |
|
44 | 45 |
|
45 | 46 |
#ifdef HAVE_NCORE |
46 | 47 |
#include "matcher-ncore.h" |
... | ... |
@@ -50,9 +51,10 @@ static cli_file_t targettab[CL_TARGET_TABLE_SIZE] = { 0, CL_TYPE_MSEXE, CL_TYPE_ |
50 | 50 |
|
51 | 51 |
extern short cli_debug_flag; |
52 | 52 |
|
53 |
-int cli_scanbuff(const unsigned char *buffer, unsigned int length, const char **virname, const struct cl_engine *engine, cli_file_t ftype) |
|
53 |
+int cli_scanbuff(const unsigned char *buffer, uint32_t length, const char **virname, const struct cl_engine *engine, cli_file_t ftype) |
|
54 | 54 |
{ |
55 |
- int ret = CL_CLEAN, i; |
|
55 |
+ int ret = CL_CLEAN; |
|
56 |
+ unsigned int i; |
|
56 | 57 |
struct cli_ac_data mdata; |
57 | 58 |
struct cli_matcher *groot, *troot = NULL; |
58 | 59 |
|
... | ... |
@@ -285,12 +287,11 @@ int cli_validatesig(cli_file_t ftype, const char *offstr, off_t fileoff, struct |
285 | 285 |
return 1; |
286 | 286 |
} |
287 | 287 |
|
288 |
-int cli_scandesc(int desc, cli_ctx *ctx, unsigned short otfrec, cli_file_t ftype, unsigned short ftonly, struct cli_matched_type **ftoffset) |
|
288 |
+int cli_scandesc(int desc, cli_ctx *ctx, uint8_t otfrec, cli_file_t ftype, uint8_t ftonly, struct cli_matched_type **ftoffset) |
|
289 | 289 |
{ |
290 | 290 |
unsigned char *buffer, *buff, *endbl, *upt; |
291 | 291 |
int ret = CL_CLEAN, type = CL_CLEAN, i, bytes; |
292 |
- unsigned int buffersize, length, maxpatlen, shift = 0; |
|
293 |
- unsigned long int offset = 0; |
|
292 |
+ uint32_t buffersize, length, maxpatlen, shift = 0, offset = 0; |
|
294 | 293 |
struct cli_ac_data gdata, tdata; |
295 | 294 |
cli_md5_ctx md5ctx; |
296 | 295 |
unsigned char digest[16]; |
... | ... |
@@ -29,6 +29,29 @@ |
29 | 29 |
#include "cltypes.h" |
30 | 30 |
#include "md5.h" |
31 | 31 |
|
32 |
+#include "matcher-ac.h" |
|
33 |
+#include "matcher-bm.h" |
|
34 |
+ |
|
35 |
+#define CLI_MATCH_WILDCARD 0xff00 |
|
36 |
+#define CLI_MATCH_IGNORE 0x0100 |
|
37 |
+#define CLI_MATCH_ALTERNATIVE 0x0200 |
|
38 |
+#define CLI_MATCH_NIBBLE_HIGH 0x0300 |
|
39 |
+#define CLI_MATCH_NIBBLE_LOW 0x0400 |
|
40 |
+ |
|
41 |
+struct cli_matcher { |
|
42 |
+ uint16_t maxpatlen; |
|
43 |
+ uint8_t ac_only; |
|
44 |
+ |
|
45 |
+ /* Extended Boyer-Moore */ |
|
46 |
+ int32_t *bm_shift; |
|
47 |
+ struct cli_bm_patt **bm_suffix; |
|
48 |
+ |
|
49 |
+ /* Extended Aho-Corasick */ |
|
50 |
+ uint8_t ac_depth; |
|
51 |
+ struct cli_ac_node *ac_root, **ac_nodetable; |
|
52 |
+ uint32_t ac_partsigs, ac_nodes; |
|
53 |
+}; |
|
54 |
+ |
|
32 | 55 |
#define CL_TARGET_TABLE_SIZE 7 |
33 | 56 |
|
34 | 57 |
struct cli_target_info { |
... | ... |
@@ -37,9 +60,9 @@ struct cli_target_info { |
37 | 37 |
int8_t status; /* 0 == not initialised, 1 == initialised OK, -1 == error */ |
38 | 38 |
}; |
39 | 39 |
|
40 |
-int cli_scandesc(int desc, cli_ctx *ctx, unsigned short otfrec, cli_file_t ftype, unsigned short ftonly, struct cli_matched_type **ftoffset); |
|
40 |
+int cli_scanbuff(const unsigned char *buffer, uint32_t length, const char **virname, const struct cl_engine *engine, cli_file_t ftype); |
|
41 | 41 |
|
42 |
-int cli_scanbuff(const unsigned char *buffer, unsigned int length, const char **virname, const struct cl_engine *engine, cli_file_t ftype); |
|
42 |
+int cli_scandesc(int desc, cli_ctx *ctx, uint8_t otfrec, cli_file_t ftype, uint8_t ftonly, struct cli_matched_type **ftoffset); |
|
43 | 43 |
|
44 | 44 |
int cli_validatesig(cli_file_t ftype, const char *offstr, off_t fileoff, struct cli_target_info *info, int desc, const char *virname); |
45 | 45 |
|
... | ... |
@@ -81,7 +81,7 @@ static pthread_mutex_t cli_gentempname_mutex = PTHREAD_MUTEX_INITIALIZER; |
81 | 81 |
#define P_tmpdir "C:\\WINDOWS\\TEMP" |
82 | 82 |
#endif |
83 | 83 |
|
84 |
-#define CL_FLEVEL 14 /* don't touch it */ |
|
84 |
+#define CL_FLEVEL 15 /* don't touch it */ |
|
85 | 85 |
|
86 | 86 |
short cli_debug_flag = 0, cli_leavetemps_flag = 0; |
87 | 87 |
|
... | ... |
@@ -286,7 +286,7 @@ char *cli_md5file(const char *filename) |
286 | 286 |
return md5str; |
287 | 287 |
} |
288 | 288 |
|
289 |
-static char *cli_md5buff(const char *buffer, unsigned int len, unsigned char *dig) |
|
289 |
+static char *cli_md5buff(const unsigned char *buffer, unsigned int len, unsigned char *dig) |
|
290 | 290 |
{ |
291 | 291 |
unsigned char digest[16]; |
292 | 292 |
char *md5str, *pt; |
... | ... |
@@ -295,7 +295,7 @@ static char *cli_md5buff(const char *buffer, unsigned int len, unsigned char *di |
295 | 295 |
|
296 | 296 |
|
297 | 297 |
cli_md5_init(&ctx); |
298 |
- cli_md5_update(&ctx, (const unsigned char *) buffer, len); |
|
298 |
+ cli_md5_update(&ctx, buffer, len); |
|
299 | 299 |
cli_md5_final(digest, &ctx); |
300 | 300 |
|
301 | 301 |
if(dig) |
... | ... |
@@ -465,7 +465,7 @@ static char *cli_gentempname(const char *dir) |
465 | 465 |
for(i = 16; i < 48; i++) |
466 | 466 |
salt[i] = cli_rndnum(256); |
467 | 467 |
|
468 |
- tmp = cli_md5buff((char *) salt, 48, name_salt); |
|
468 |
+ tmp = cli_md5buff(salt, 48, name_salt); |
|
469 | 469 |
|
470 | 470 |
#ifdef CL_THREAD_SAFE |
471 | 471 |
pthread_mutex_unlock(&cli_gentempname_mutex); |
... | ... |
@@ -49,9 +49,9 @@ |
49 | 49 |
#endif |
50 | 50 |
#include "matcher-ac.h" |
51 | 51 |
#include "matcher-bm.h" |
52 |
+#include "matcher.h" |
|
52 | 53 |
#include "others.h" |
53 | 54 |
#include "str.h" |
54 |
-#include "defaults.h" |
|
55 | 55 |
#include "dconf.h" |
56 | 56 |
#include "lockdb.h" |
57 | 57 |
#include "readdb.h" |
... | ... |
@@ -101,6 +101,9 @@ static int cli_ac_addsig(struct cli_matcher *root, const char *virname, const ch |
101 | 101 |
free(hex); \ |
102 | 102 |
} |
103 | 103 |
|
104 |
+ if(strlen(hexsig) / 2 < AC_DEFAULT_DEPTH) |
|
105 |
+ return CL_EPATSHORT; |
|
106 |
+ |
|
104 | 107 |
if((new = (struct cli_ac_patt *) cli_calloc(1, sizeof(struct cli_ac_patt))) == NULL) |
105 | 108 |
return CL_EMEM; |
106 | 109 |
|
... | ... |
@@ -210,7 +213,7 @@ static int cli_ac_addsig(struct cli_matcher *root, const char *virname, const ch |
210 | 210 |
} |
211 | 211 |
} |
212 | 212 |
|
213 |
- if((new->pattern = cli_hex2si(new->alt ? hex : hexsig)) == NULL) { |
|
213 |
+ if((new->pattern = cli_hex2ui(new->alt ? hex : hexsig)) == NULL) { |
|
214 | 214 |
FREE_ALT; |
215 | 215 |
if(new->offset) |
216 | 216 |
free(new->offset); |
... | ... |
@@ -221,17 +224,17 @@ static int cli_ac_addsig(struct cli_matcher *root, const char *virname, const ch |
221 | 221 |
new->length = strlen(new->alt ? hex : hexsig) / 2; |
222 | 222 |
|
223 | 223 |
for(i = 0; i < AC_DEFAULT_DEPTH; i++) { |
224 |
- if(new->pattern[i] == CLI_IGN || new->pattern[i] == CLI_ALT) { |
|
224 |
+ if(new->pattern[i] & CLI_MATCH_WILDCARD) { |
|
225 | 225 |
wprefix = 1; |
226 | 226 |
break; |
227 | 227 |
} |
228 | 228 |
} |
229 | 229 |
|
230 | 230 |
if(wprefix) { |
231 |
- for(; i < new->length - AC_DEFAULT_DEPTH + 1; i++) { |
|
231 |
+ for(; i < (uint16_t) (new->length - AC_DEFAULT_DEPTH + 1); i++) { |
|
232 | 232 |
wprefix = 0; |
233 | 233 |
for(j = i; j < i + AC_DEFAULT_DEPTH; j++) { |
234 |
- if(new->pattern[j] == CLI_IGN || new->pattern[j] == CLI_ALT) { |
|
234 |
+ if(new->pattern[j] & CLI_MATCH_WILDCARD) { |
|
235 | 235 |
wprefix = 1; |
236 | 236 |
break; |
237 | 237 |
} |
... | ... |
@@ -255,7 +258,7 @@ static int cli_ac_addsig(struct cli_matcher *root, const char *virname, const ch |
255 | 255 |
new->length -= i; |
256 | 256 |
|
257 | 257 |
for(i = 0; i < new->prefix_length; i++) |
258 |
- if(new->prefix[i] == CLI_ALT) |
|
258 |
+ if((new->prefix[i] & CLI_MATCH_WILDCARD) == CLI_MATCH_ALTERNATIVE) |
|
259 | 259 |
new->alt_pattern++; |
260 | 260 |
} |
261 | 261 |
|
... | ... |
@@ -524,7 +527,7 @@ int cli_initengine(struct cl_engine **engine, unsigned int options) |
524 | 524 |
|
525 | 525 |
(*engine)->refcount = 1; |
526 | 526 |
|
527 |
- (*engine)->root = (struct cli_matcher **) cli_calloc(CL_TARGET_TABLE_SIZE, sizeof(struct cli_matcher *)); |
|
527 |
+ (*engine)->root = cli_calloc(CL_TARGET_TABLE_SIZE, sizeof(struct cli_matcher *)); |
|
528 | 528 |
if(!(*engine)->root) { |
529 | 529 |
/* no need to free previously allocated memory here */ |
530 | 530 |
cli_errmsg("Can't allocate memory for roots!\n"); |
... | ... |
@@ -1674,9 +1677,9 @@ void cl_free(struct cl_engine *engine) |
1674 | 1674 |
if(engine->root) { |
1675 | 1675 |
for(i = 0; i < CL_TARGET_TABLE_SIZE; i++) { |
1676 | 1676 |
if((root = engine->root[i])) { |
1677 |
- cli_ac_free(root); |
|
1678 |
- if(!engine->root[i]->ac_only) |
|
1677 |
+ if(!root->ac_only) |
|
1679 | 1678 |
cli_bm_free(root); |
1679 |
+ cli_ac_free(root); |
|
1680 | 1680 |
free(root); |
1681 | 1681 |
} |
1682 | 1682 |
} |
... | ... |
@@ -1748,7 +1748,7 @@ static int cli_scanembpe(int desc, cli_ctx *ctx) |
1748 | 1748 |
static int cli_scanraw(int desc, cli_ctx *ctx, cli_file_t type) |
1749 | 1749 |
{ |
1750 | 1750 |
int ret = CL_CLEAN, nret = CL_CLEAN; |
1751 |
- unsigned short ftrec, break_loop = 0; |
|
1751 |
+ uint8_t ftrec, break_loop = 0; |
|
1752 | 1752 |
struct cli_matched_type *ftoffset = NULL, *fpt; |
1753 | 1753 |
uint32_t lastzip, lastrar; |
1754 | 1754 |
struct cli_exe_info peinfo; |
... | ... |
@@ -33,7 +33,8 @@ |
33 | 33 |
|
34 | 34 |
#include "clamav.h" |
35 | 35 |
#include "others.h" |
36 |
-#include "defaults.h" |
|
36 |
+#include "matcher.h" |
|
37 |
+#include "cltypes.h" |
|
37 | 38 |
|
38 | 39 |
static int cli_hex2int(int c) |
39 | 40 |
{ |
... | ... |
@@ -53,30 +54,53 @@ static int cli_hex2int(int c) |
53 | 53 |
return -1; |
54 | 54 |
} |
55 | 55 |
|
56 |
-short int *cli_hex2si(const char *hex) |
|
56 |
+uint16_t *cli_hex2ui(const char *hex) |
|
57 | 57 |
{ |
58 |
- short int *str, *ptr, val, c; |
|
59 |
- int i, len; |
|
58 |
+ uint16_t *str, *ptr, val; |
|
59 |
+ unsigned int i, len; |
|
60 |
+ int c; |
|
60 | 61 |
|
61 | 62 |
|
62 | 63 |
len = strlen(hex); |
63 | 64 |
|
64 | 65 |
if(len % 2 != 0) { |
65 |
- cli_errmsg("cli_hex2si(): Malformed hexstring: %s (length: %d)\n", hex, len); |
|
66 |
+ cli_errmsg("cli_hex2si(): Malformed hexstring: %s (length: %u)\n", hex, len); |
|
66 | 67 |
return NULL; |
67 | 68 |
} |
68 | 69 |
|
69 |
- str = cli_calloc((len / 2) + 1, sizeof(short int)); |
|
70 |
+ str = cli_calloc((len / 2) + 1, sizeof(uint16_t)); |
|
70 | 71 |
if(!str) |
71 | 72 |
return NULL; |
72 | 73 |
|
73 | 74 |
ptr = str; |
74 | 75 |
|
75 | 76 |
for(i = 0; i < len; i += 2) { |
76 |
- if(hex[i] == '?') { |
|
77 |
- val = CLI_IGN; |
|
77 |
+ val = 0; |
|
78 |
+ |
|
79 |
+ if(hex[i] == '?' && hex[i + 1] == '?') { |
|
80 |
+ val |= CLI_MATCH_IGNORE; |
|
81 |
+ |
|
82 |
+ } else if(hex[i + 1] == '?') { |
|
83 |
+ if((c = cli_hex2int(hex[i])) >= 0) { |
|
84 |
+ val = c << 4; |
|
85 |
+ } else { |
|
86 |
+ free(str); |
|
87 |
+ return NULL; |
|
88 |
+ } |
|
89 |
+ val |= CLI_MATCH_NIBBLE_HIGH; |
|
90 |
+ |
|
91 |
+ } else if(hex[i] == '?') { |
|
92 |
+ if((c = cli_hex2int(hex[i + 1])) >= 0) { |
|
93 |
+ val = c; |
|
94 |
+ } else { |
|
95 |
+ free(str); |
|
96 |
+ return NULL; |
|
97 |
+ } |
|
98 |
+ val |= CLI_MATCH_NIBBLE_LOW; |
|
99 |
+ |
|
78 | 100 |
} else if(hex[i] == '@') { |
79 |
- val = CLI_ALT; |
|
101 |
+ val |= CLI_MATCH_ALTERNATIVE; |
|
102 |
+ |
|
80 | 103 |
} else { |
81 | 104 |
if((c = cli_hex2int(hex[i])) >= 0) { |
82 | 105 |
val = c; |
... | ... |
@@ -91,6 +115,7 @@ short int *cli_hex2si(const char *hex) |
91 | 91 |
return NULL; |
92 | 92 |
} |
93 | 93 |
} |
94 |
+ |
|
94 | 95 |
*ptr++ = val; |
95 | 96 |
} |
96 | 97 |
|
... | ... |
@@ -22,10 +22,12 @@ |
22 | 22 |
|
23 | 23 |
#include <sys/types.h> |
24 | 24 |
|
25 |
+#include "cltypes.h" |
|
26 |
+ |
|
25 | 27 |
int cli_strbcasestr(const char *haystack, const char *needle); |
26 | 28 |
int cli_chomp(char *string); |
27 | 29 |
char *cli_strtok(const char *line, int field, const char *delim); |
28 |
-short int *cli_hex2si(const char *hex); |
|
30 |
+uint16_t *cli_hex2ui(const char *hex); |
|
29 | 31 |
char *cli_hex2str(const char *hex); |
30 | 32 |
int cli_hex2num(const char *hex); |
31 | 33 |
char *cli_str2hex(const char *string, unsigned int len); |