mirror of
https://github.com/Cisco-Talos/clamav.git
synced 2025-10-19 18:33:16 +00:00

This patch adds experimental-quality CMake build tooling. The libmspack build required a modification to use "" instead of <> for header #includes. This will hopefully be included in the libmspack upstream project when adding CMake build tooling to libmspack. Removed use of libltdl when using CMake. Flex & Bison are now required to build. If -DMAINTAINER_MODE, then GPERF is also required, though it currently doesn't actually do anything. TODO! I found that the autotools build system was generating the lexer output but not actually compiling it, instead using previously generated (and manually renamed) lexer c source. As a consequence, changes to the .l and .y files weren't making it into the build. To resolve this, I removed generated flex/bison files and fixed the tooling to use the freshly generated files. Flex and bison are now required build tools. On Windows, this adds a dependency on the winflexbison package, which can be obtained using Chocolatey or may be manually installed. CMake tooling only has partial support for building with external LLVM library, and no support for the internal LLVM (to be removed in the future). I.e. The CMake build currently only supports the bytecode interpreter. Many files used include paths relative to the top source directory or relative to the current project, rather than relative to each build target. Modern CMake support requires including internal dependency headers the same way you would external dependency headers (albeit with "" instead of <>). This meant correcting all header includes to be relative to the build targets and not relative to the workspace. For example, ... ```c include "../libclamav/clamav.h" include "clamd/clamd_others.h" ``` ... becomes: ```c // libclamav include "clamav.h" // clamd include "clamd_others.h" ``` Fixes header name conflicts by renaming a few of the files. Converted the "shared" code into a static library, which depends on libclamav. The ironically named "shared" static library provides features common to the ClamAV apps which are not required in libclamav itself and are not intended for use by downstream projects. This change was required for correct modern CMake practices but was also required to use the automake "subdir-objects" option. This eliminates warnings when running autoreconf which, in the next version of autoconf & automake are likely to break the build. libclamav used to build in multiple stages where an earlier stage is a static library containing utils required by the "shared" code. Linking clamdscan and clamdtop with this libclamav utils static lib allowed these two apps to function without libclamav. While this is nice in theory, the practical gains are minimal and it complicates the build system. As such, the autotools and CMake tooling was simplified for improved maintainability and this feature was thrown out. clamdtop and clamdscan now require libclamav to function. Removed the nopthreads version of the autotools libclamav_internal_utils static library and added pthread linking to a couple apps that may have issues building on some platforms without it, with the intention of removing needless complexity from the source. Kept the regular version of libclamav_internal_utils.la though it is no longer used anywhere but in libclamav. Added an experimental doxygen build option which attempts to build clamav.h and libfreshclam doxygen html docs. The CMake build tooling also may build the example program(s), which isn't a feature in the Autotools build system. Changed C standard to C90+ due to inline linking issues with socket.h when linking libfreshclam.so on Linux. Generate common.rc for win32. Fix tabs/spaces in shared Makefile.am, and remove vestigial ifndef from misc.c. Add CMake files to the automake dist, so users can try the new CMake tooling w/out having to build from a git clone. clamonacc changes: - Renamed FANOTIFY macro to HAVE_SYS_FANOTIFY_H to better match other similar macros. - Added a new clamav-clamonacc.service systemd unit file, based on the work of ChadDevOps & Aaron Brighton. - Added missing clamonacc man page. Updates to clamdscan man page, add missing options. Remove vestigial CL_NOLIBCLAMAV definitions (all apps now use libclamav). Rename Windows mspack.dll to libmspack.dll so all ClamAV-built libraries have the lib-prefix with Visual Studio as with CMake.
709 lines
25 KiB
C
709 lines
25 KiB
C
/*
|
|
* Unit tests for regular expression processing.
|
|
*
|
|
* Copyright (C) 2013-2020 Cisco Systems, Inc. and/or its affiliates. All rights reserved.
|
|
* Copyright (C) 2008-2013 Sourcefire, Inc.
|
|
*
|
|
* Authors: Török Edvin
|
|
*
|
|
* This program is free software; you can redistribute it and/or modify
|
|
* it under the terms of the GNU General Public License version 2 as
|
|
* published by the Free Software Foundation.
|
|
*
|
|
* 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, write to the Free Software
|
|
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
|
|
* MA 02110-1301, USA.
|
|
*/
|
|
#if HAVE_CONFIG_H
|
|
#include "clamav-config.h"
|
|
#endif
|
|
|
|
#include <stdio.h>
|
|
#include <stdlib.h>
|
|
#include <limits.h>
|
|
#include <string.h>
|
|
#include <check.h>
|
|
|
|
// libclamav
|
|
#include "clamav.h"
|
|
#include "others.h"
|
|
#include "mbox.h"
|
|
#include "message.h"
|
|
#include "htmlnorm.h"
|
|
#include "phishcheck.h"
|
|
#include "regex_suffix.h"
|
|
#include "regex_list.h"
|
|
#include "phish_domaincheck_db.h"
|
|
#include "phish_whitelist.h"
|
|
|
|
#include "checks.h"
|
|
|
|
static size_t cb_called = 0;
|
|
|
|
static cl_error_t cb_fail(void *cbdata, const char *suffix, size_t len, const struct regex_list *regex)
|
|
{
|
|
UNUSEDPARAM(cbdata);
|
|
UNUSEDPARAM(suffix);
|
|
UNUSEDPARAM(len);
|
|
UNUSEDPARAM(regex);
|
|
|
|
ck_abort_msg("this pattern is not supposed to have a suffix");
|
|
return CL_EMEM;
|
|
}
|
|
|
|
static cl_error_t cb_expect_single(void *cbdata, const char *suffix, size_t len, const struct regex_list *regex)
|
|
{
|
|
const char *expected = cbdata;
|
|
|
|
UNUSEDPARAM(len);
|
|
UNUSEDPARAM(regex);
|
|
|
|
cb_called++;
|
|
ck_assert_msg(suffix && strcmp(suffix, expected) == 0,
|
|
"suffix mismatch, was: %s, expected: %s\n", suffix, expected);
|
|
return CL_SUCCESS;
|
|
}
|
|
|
|
static struct regex_list regex;
|
|
START_TEST(empty)
|
|
{
|
|
const char pattern[] = "";
|
|
cl_error_t rc;
|
|
regex_t *preg;
|
|
|
|
errmsg_expected();
|
|
preg = malloc(sizeof(*regex.preg));
|
|
ck_assert_msg(!!preg, "malloc");
|
|
rc = cli_regex2suffix(pattern, preg, cb_fail, NULL);
|
|
free(preg);
|
|
ck_assert_msg(rc == REG_EMPTY, "empty pattern");
|
|
ck_assert_msg(cb_called == 0, "callback shouldn't be called");
|
|
}
|
|
END_TEST
|
|
|
|
START_TEST(one)
|
|
{
|
|
char pattern[] = "a";
|
|
cl_error_t rc;
|
|
regex_t *preg;
|
|
|
|
preg = malloc(sizeof(*regex.preg));
|
|
ck_assert_msg(!!preg, "malloc");
|
|
rc = cli_regex2suffix(pattern, preg, cb_expect_single, pattern);
|
|
ck_assert_msg(rc == CL_SUCCESS, "single character pattern");
|
|
cli_regfree(preg);
|
|
free(preg);
|
|
ck_assert_msg(cb_called == 1, "callback should be called once");
|
|
}
|
|
END_TEST
|
|
|
|
static const char *ex1[] =
|
|
{"com|de", "moc", "ed", NULL};
|
|
static const char *ex2[] =
|
|
{"xd|(a|e)bc", "dx", "cba", "cbe", NULL};
|
|
|
|
static const char **tests[] = {
|
|
ex1,
|
|
ex2};
|
|
|
|
static cl_error_t cb_expect_multi(void *cbdata, const char *suffix, size_t len, const struct regex_list *r)
|
|
{
|
|
const char **exp = cbdata;
|
|
|
|
UNUSEDPARAM(r);
|
|
|
|
ck_assert_msg(!!exp, "expected data");
|
|
exp++;
|
|
ck_assert_msg(!!*exp, "expected no suffix, got: %s\n", suffix);
|
|
ck_assert_msg(!!exp[cb_called], "expected less suffixes, but already got: %d\n", cb_called);
|
|
ck_assert_msg(strcmp(exp[cb_called], suffix) == 0,
|
|
"suffix mismatch, was: %s, expected: %s\n", suffix, exp[cb_called]);
|
|
ck_assert_msg(strlen(suffix) == len, "incorrect suffix len, expected: %d, got: %d\n", strlen(suffix), len);
|
|
cb_called++;
|
|
return CL_SUCCESS;
|
|
}
|
|
|
|
START_TEST(test_suffix)
|
|
{
|
|
cl_error_t rc;
|
|
regex_t *preg;
|
|
const char *pattern = tests[_i][0];
|
|
size_t n = 0;
|
|
const char **p = tests[_i];
|
|
|
|
ck_assert_msg(!!pattern, "test pattern");
|
|
preg = malloc(sizeof(*regex.preg));
|
|
ck_assert_msg(!!preg, "malloc");
|
|
rc = cli_regex2suffix(pattern, preg, cb_expect_multi, tests[_i]);
|
|
ck_assert_msg(rc == CL_SUCCESS, "single character pattern");
|
|
cli_regfree(preg);
|
|
free(preg);
|
|
p++;
|
|
while (*p++) n++;
|
|
ck_assert_msg(cb_called == n,
|
|
"suffix number mismatch, expected: %d, was: %d\n", n, cb_called);
|
|
}
|
|
END_TEST
|
|
|
|
static void setup(void)
|
|
{
|
|
cb_called = 0;
|
|
}
|
|
|
|
static void teardown(void)
|
|
{
|
|
}
|
|
|
|
static struct regex_matcher matcher;
|
|
|
|
static void rsetup(void)
|
|
{
|
|
cl_error_t rc;
|
|
#ifdef USE_MPOOL
|
|
matcher.mempool = mpool_create();
|
|
#endif
|
|
rc = init_regex_list(&matcher, 1);
|
|
ck_assert_msg(rc == CL_SUCCESS, "init_regex_list");
|
|
}
|
|
|
|
static void rteardown(void)
|
|
{
|
|
regex_list_done(&matcher);
|
|
#ifdef USE_MPOOL
|
|
mpool_destroy(matcher.mempool);
|
|
#endif
|
|
}
|
|
|
|
typedef enum rtest_result {
|
|
RTR_PHISH,
|
|
RTR_WHITELISTED,
|
|
RTR_CLEAN,
|
|
RTR_BLACKLISTED, // if 2nd db is loaded
|
|
RTR_INVALID_REGEX
|
|
} rtr_t;
|
|
|
|
static const struct rtest {
|
|
const char *pattern; /* NULL if not meant for whitelist testing */
|
|
const char *realurl;
|
|
const char *displayurl;
|
|
rtr_t result;
|
|
} rtests[] = {
|
|
{NULL, "http://fake.example.com", "http://foo@key.com/", RTR_PHISH},
|
|
{NULL, "http://fake.example.com", "foo.example.com@key.com", RTR_PHISH},
|
|
{NULL, "http://fake.example.com", "foo@key.com", RTR_CLEAN},
|
|
{NULL, "http://fake.example.com", "=====key.com", RTR_PHISH},
|
|
{NULL, "http://key.com", "=====key.com", RTR_CLEAN},
|
|
{NULL, " http://key.com", "=====key.com", RTR_CLEAN},
|
|
{NULL, "http://key.com@fake.example.com", "key.com", RTR_PHISH},
|
|
{NULL, " http://key.com@fake.example.com", "key.com", RTR_PHISH},
|
|
{NULL, " http://key.com@fake.example.com ", "key.com", RTR_PHISH},
|
|
/* entry taken from .wdb with a / appended */
|
|
{".+\\.ebayrtm\\.com([/?].*)?:.+\\.ebay\\.(de|com|co\\.uk)([/?].*)?/",
|
|
"http://srx.main.ebayrtm.com",
|
|
"pages.ebay.de",
|
|
RTR_WHITELISTED /* should be whitelisted */},
|
|
{".+\\.ebayrtm\\.com([/?].*)?:.+\\.ebay\\.(de|com|co\\.uk)([/?].*)?/",
|
|
"http://srx.main.ebayrtm.com.evil.example.com",
|
|
"pages.ebay.de",
|
|
RTR_PHISH},
|
|
{".+\\.ebayrtm\\.com([/?].*)?:.+\\.ebay\\.(de|com|co\\.uk)([/?].*)?/",
|
|
"www.www.ebayrtm.com?somecgi",
|
|
"www.ebay.com/something", RTR_WHITELISTED},
|
|
{NULL,
|
|
"http://key.com", "go to key.com", RTR_CLEAN},
|
|
{":.+\\.paypal\\.(com|de|fr|it)([/?].*)?:.+\\.ebay\\.(at|be|ca|ch|co\\.uk|de|es|fr|ie|in|it|nl|ph|pl|com(\\.(au|cn|hk|my|sg))?)([/?].*)?/",
|
|
"http://www.paypal.com", "pics.ebay.com", RTR_WHITELISTED},
|
|
{NULL, "http://somefakeurl.example.com", "someotherdomain-key.com", RTR_CLEAN},
|
|
{NULL, "http://somefakeurl.example.com", "someotherdomain.key.com", RTR_PHISH},
|
|
{NULL, "http://malware-test.example.com/something", "test", RTR_BLACKLISTED},
|
|
{NULL, "http://phishing-test.example.com/something", "test", RTR_BLACKLISTED},
|
|
{NULL, "http://sub.malware-test.example.com/2", "test", RTR_BLACKLISTED},
|
|
{NULL, "http://sub.phishing-test.example.com/2", "test", RTR_BLACKLISTED},
|
|
{NULL, "http://user@malware-test.example.com/2", "test", RTR_BLACKLISTED},
|
|
{NULL, "http://user@phishing-test.example.com/2", "test", RTR_BLACKLISTED},
|
|
{NULL, "http://user@malware-test.example.com/2/test", "test", RTR_BLACKLISTED},
|
|
{NULL, "http://user@phishing-test.example.com/2/test", "test", RTR_BLACKLISTED},
|
|
{NULL, "http://user@malware-test.example.com/", "test", RTR_BLACKLISTED},
|
|
{NULL, "http://user@phishing-test.example.com/", "test", RTR_BLACKLISTED},
|
|
{NULL, "http://x.exe", "http:///x.exe", RTR_CLEAN},
|
|
{".+\\.ebayrtm\\.com([/?].*)?:[^.]+\\.ebay\\.(de|com|co\\.uk)/",
|
|
"http://srx.main.ebayrtm.com",
|
|
"pages.ebay.de",
|
|
RTR_WHITELISTED /* should be whitelisted */},
|
|
{".+\\.ebayrtm\\.com([/?].*)?:.+[r-t]\\.ebay\\.(de|com|co\\.uk)/",
|
|
"http://srx.main.ebayrtm.com",
|
|
"pages.ebay.de",
|
|
RTR_WHITELISTED /* should be whitelisted */},
|
|
{".+\\.ebayrtm\\.com([/?].*)?:.+[r-t]\\.ebay\\.(de|com|co\\.uk)/",
|
|
"http://srx.main.ebayrtm.com",
|
|
"pages.ebay.de",
|
|
RTR_WHITELISTED /* should be whitelisted */},
|
|
{"[t-", "", "", RTR_INVALID_REGEX},
|
|
{NULL, "http://co.uk", "http:// co.uk", RTR_CLEAN},
|
|
{NULL, "http://co.uk", " ", RTR_CLEAN},
|
|
{NULL, "127.0.0.1", "pages.ebay.de", RTR_CLEAN},
|
|
{".+\\.ebayrtm\\.com([/?].*)?:.+\\.ebay\\.(de|com|co\\.uk)([/?].*)?/",
|
|
"http://pages.ebay.de@fake.example.com", "pages.ebay.de", RTR_PHISH},
|
|
{NULL, "http://key.com", "https://key.com", RTR_PHISH},
|
|
{NULL, "http://key.com%00fake.example.com", "https://key.com", RTR_PHISH},
|
|
{NULL, "http://key.com.example.com", "key.com.invalid", RTR_PHISH}};
|
|
|
|
START_TEST(regex_list_match_test)
|
|
{
|
|
const char *info;
|
|
const struct rtest *rtest = &rtests[_i];
|
|
char *pattern, *realurl;
|
|
cl_error_t rc;
|
|
|
|
if (!rtest->pattern) {
|
|
ck_assert_msg(rtest->result != RTR_WHITELISTED,
|
|
"whitelist test must have pattern set");
|
|
/* this test entry is not meant for whitelist testing */
|
|
return;
|
|
}
|
|
|
|
ck_assert_msg(rtest->result == RTR_PHISH || rtest->result == RTR_WHITELISTED || rtest->result == RTR_INVALID_REGEX,
|
|
"whitelist test result must be either RTR_PHISH or RTR_WHITELISTED or RTR_INVALID_REGEX");
|
|
pattern = cli_strdup(rtest->pattern);
|
|
ck_assert_msg(!!pattern, "cli_strdup");
|
|
|
|
rc = regex_list_add_pattern(&matcher, pattern);
|
|
if (rtest->result == RTR_INVALID_REGEX) {
|
|
ck_assert_msg(rc, "regex_list_add_pattern should return error");
|
|
free(pattern);
|
|
return;
|
|
} else
|
|
ck_assert_msg(rc == CL_SUCCESS, "regex_list_add_pattern");
|
|
free(pattern);
|
|
|
|
matcher.list_loaded = 1;
|
|
|
|
rc = cli_build_regex_list(&matcher);
|
|
ck_assert_msg(rc == CL_SUCCESS, "cli_build_regex_list");
|
|
|
|
ck_assert_msg(is_regex_ok(&matcher), "is_regex_ok");
|
|
|
|
realurl = cli_strdup(rtest->realurl);
|
|
rc = regex_list_match(&matcher, realurl, rtest->displayurl, NULL, 1, &info, 1);
|
|
ck_assert_msg(rc == rtest->result, "regex_list_match");
|
|
/* regex_list_match is not supposed to modify realurl in this case */
|
|
ck_assert_msg(!strcmp(realurl, rtest->realurl), "realurl altered");
|
|
free(realurl);
|
|
}
|
|
END_TEST
|
|
|
|
static struct cl_engine *engine;
|
|
static int loaded_2 = 0;
|
|
|
|
static void psetup_impl(int load2)
|
|
{
|
|
FILE *f;
|
|
cl_error_t rc;
|
|
unsigned signo = 0;
|
|
|
|
engine = cl_engine_new();
|
|
ck_assert_msg(!!engine, "cl_engine_new");
|
|
|
|
phishing_init(engine);
|
|
ck_assert_msg(!!engine->phishcheck, "phishing_init");
|
|
|
|
rc = init_domainlist(engine);
|
|
ck_assert_msg(rc == CL_SUCCESS, "init_domainlist");
|
|
|
|
f = fdopen(open_testfile("input/daily.pdb"), "r");
|
|
ck_assert_msg(!!f, "fopen daily.pdb");
|
|
|
|
rc = load_regex_matcher(engine, engine->domainlist_matcher, f, &signo, 0, 0, NULL, 1);
|
|
ck_assert_msg(rc == CL_SUCCESS, "load_regex_matcher");
|
|
fclose(f);
|
|
|
|
ck_assert_msg(signo == 201, "Incorrect number of signatures: %u, expected %u", signo, 201);
|
|
|
|
if (load2) {
|
|
f = fdopen(open_testfile("input/daily.gdb"), "r");
|
|
ck_assert_msg(!!f, "fopen daily.gdb");
|
|
|
|
signo = 0;
|
|
rc = load_regex_matcher(engine, engine->domainlist_matcher, f, &signo, 0, 0, NULL, 1);
|
|
ck_assert_msg(rc == CL_SUCCESS, "load_regex_matcher");
|
|
fclose(f);
|
|
|
|
ck_assert_msg(signo == 4, "Incorrect number of signatures: %u, expected %u", signo, 4);
|
|
}
|
|
loaded_2 = load2;
|
|
|
|
rc = init_whitelist(engine);
|
|
ck_assert_msg(rc == CL_SUCCESS, "init_whitelist");
|
|
|
|
f = fdopen(open_testfile("input/daily.wdb"), "r");
|
|
signo = 0;
|
|
rc = load_regex_matcher(engine, engine->whitelist_matcher, f, &signo, 0, 1, NULL, 1);
|
|
ck_assert_msg(rc == CL_SUCCESS, "load_regex_matcher");
|
|
fclose(f);
|
|
|
|
ck_assert_msg(signo == 31, "Incorrect number of signatures: %u, expected %u", signo, 31);
|
|
|
|
rc = cli_build_regex_list(engine->whitelist_matcher);
|
|
ck_assert_msg(rc == CL_SUCCESS, "cli_build_regex_list");
|
|
|
|
rc = cli_build_regex_list(engine->domainlist_matcher);
|
|
ck_assert_msg(rc == CL_SUCCESS, "cli_build_regex_list");
|
|
|
|
ck_assert_msg(is_regex_ok(engine->whitelist_matcher), "is_regex_ok");
|
|
ck_assert_msg(is_regex_ok(engine->domainlist_matcher), "is_regex_ok");
|
|
}
|
|
|
|
static void psetup(void)
|
|
{
|
|
psetup_impl(0);
|
|
}
|
|
|
|
static void psetup2(void)
|
|
{
|
|
psetup_impl(1);
|
|
}
|
|
|
|
static void pteardown(void)
|
|
{
|
|
if (engine) {
|
|
cl_engine_free(engine);
|
|
}
|
|
engine = NULL;
|
|
}
|
|
|
|
static void do_phishing_test(const struct rtest *rtest)
|
|
{
|
|
char *realurl;
|
|
cli_ctx ctx;
|
|
struct cl_scan_options options;
|
|
const char *virname = NULL;
|
|
tag_arguments_t hrefs;
|
|
cl_error_t rc;
|
|
|
|
memset(&ctx, 0, sizeof(ctx));
|
|
memset(&options, 0, sizeof(struct cl_scan_options));
|
|
ctx.options = &options;
|
|
|
|
realurl = cli_strdup(rtest->realurl);
|
|
ck_assert_msg(!!realurl, "cli_strdup");
|
|
|
|
hrefs.count = 1;
|
|
hrefs.value = cli_malloc(sizeof(*hrefs.value));
|
|
ck_assert_msg(!!hrefs.value, "cli_malloc");
|
|
hrefs.value[0] = (unsigned char *)realurl;
|
|
hrefs.contents = cli_malloc(sizeof(*hrefs.contents));
|
|
ck_assert_msg(!!hrefs.contents, "cli_malloc");
|
|
hrefs.tag = cli_malloc(sizeof(*hrefs.tag));
|
|
ck_assert_msg(!!hrefs.tag, "cli_malloc");
|
|
hrefs.tag[0] = (unsigned char *)cli_strdup("href");
|
|
hrefs.contents[0] = (unsigned char *)cli_strdup(rtest->displayurl);
|
|
|
|
ctx.engine = engine;
|
|
ctx.virname = &virname;
|
|
|
|
rc = phishingScan(&ctx, &hrefs);
|
|
|
|
html_tag_arg_free(&hrefs);
|
|
ck_assert_msg(rc == CL_CLEAN, "phishingScan");
|
|
switch (rtest->result) {
|
|
case RTR_PHISH:
|
|
ck_assert_msg(ctx.found_possibly_unwanted,
|
|
"this should be phishing, realURL: %s, displayURL: %s",
|
|
rtest->realurl, rtest->displayurl);
|
|
break;
|
|
case RTR_WHITELISTED:
|
|
ck_assert_msg(!ctx.found_possibly_unwanted,
|
|
"this should be whitelisted, realURL: %s, displayURL: %s",
|
|
rtest->realurl, rtest->displayurl);
|
|
break;
|
|
case RTR_CLEAN:
|
|
ck_assert_msg(!ctx.found_possibly_unwanted,
|
|
"this should be clean, realURL: %s, displayURL: %s",
|
|
rtest->realurl, rtest->displayurl);
|
|
break;
|
|
case RTR_BLACKLISTED:
|
|
if (!loaded_2)
|
|
ck_assert_msg(!ctx.found_possibly_unwanted,
|
|
"this should be clean, realURL: %s, displayURL: %s",
|
|
rtest->realurl, rtest->displayurl);
|
|
else {
|
|
ck_assert_msg(ctx.found_possibly_unwanted,
|
|
"this should be blacklisted, realURL: %s, displayURL: %s",
|
|
rtest->realurl, rtest->displayurl);
|
|
if (*ctx.virname) {
|
|
char *phishingFound = NULL;
|
|
char *detectionName = NULL;
|
|
if (strstr(rtest->realurl, "malware-test")) {
|
|
detectionName = "Heuristics.Safebrowsing.Suspected-malware_safebrowsing.clamav.net";
|
|
|
|
} else if (strstr(rtest->realurl, "phishing-test")) {
|
|
detectionName = "Heuristics.Safebrowsing.Suspected-phishing_safebrowsing.clamav.net";
|
|
}
|
|
ck_assert_msg(detectionName != NULL, "\n\t Blacklist test case error - malware-test or phishing-test not found in: %s\n", rtest->realurl);
|
|
phishingFound = strstr((const char *)*ctx.virname, detectionName);
|
|
ck_assert_msg(phishingFound != NULL, "\n\t should be: %s,\n\t but is: %s\n", detectionName, *ctx.virname);
|
|
}
|
|
}
|
|
break;
|
|
case RTR_INVALID_REGEX:
|
|
/* don't worry about it, this was tested in regex_list_match_test() */
|
|
break;
|
|
}
|
|
}
|
|
|
|
static void do_phishing_test_allscan(const struct rtest *rtest)
|
|
{
|
|
char *realurl;
|
|
cli_ctx ctx;
|
|
const char *virname = NULL;
|
|
tag_arguments_t hrefs;
|
|
cl_error_t rc;
|
|
struct cl_scan_options options;
|
|
|
|
memset(&ctx, 0, sizeof(ctx));
|
|
memset(&options, 0, sizeof(struct cl_scan_options));
|
|
ctx.options = &options;
|
|
|
|
realurl = cli_strdup(rtest->realurl);
|
|
ck_assert_msg(!!realurl, "cli_strdup");
|
|
|
|
hrefs.count = 1;
|
|
hrefs.value = cli_malloc(sizeof(*hrefs.value));
|
|
ck_assert_msg(!!hrefs.value, "cli_malloc");
|
|
hrefs.value[0] = (unsigned char *)realurl;
|
|
hrefs.contents = cli_malloc(sizeof(*hrefs.contents));
|
|
ck_assert_msg(!!hrefs.contents, "cli_malloc");
|
|
hrefs.tag = cli_malloc(sizeof(*hrefs.tag));
|
|
ck_assert_msg(!!hrefs.tag, "cli_malloc");
|
|
hrefs.tag[0] = (unsigned char *)cli_strdup("href");
|
|
hrefs.contents[0] = (unsigned char *)cli_strdup(rtest->displayurl);
|
|
|
|
ctx.engine = engine;
|
|
ctx.virname = &virname;
|
|
ctx.options->general |= CL_SCAN_GENERAL_ALLMATCHES;
|
|
|
|
rc = phishingScan(&ctx, &hrefs);
|
|
|
|
html_tag_arg_free(&hrefs);
|
|
if (rtest->result == RTR_PHISH || (loaded_2 != 0 && rtest->result == RTR_BLACKLISTED)) {
|
|
ck_assert_msg(rc == CL_VIRUS, "phishingScan returned \"%s\", expected \"%s\". \n\trealURL: %s \n\tdisplayURL: %s",
|
|
cl_strerror(rc),
|
|
cl_strerror(CL_VIRUS),
|
|
rtest->realurl, rtest->displayurl);
|
|
} else {
|
|
ck_assert_msg(rc == CL_CLEAN, "phishingScan returned \"%s\", expected \"%s\". \n\trealURL: %s \n\tdisplayURL: %s",
|
|
cl_strerror(rc),
|
|
cl_strerror(CL_CLEAN),
|
|
rtest->realurl, rtest->displayurl);
|
|
}
|
|
switch (rtest->result) {
|
|
case RTR_PHISH:
|
|
ck_assert_msg(ctx.num_viruses,
|
|
"this should be phishing, realURL: %s, displayURL: %s",
|
|
rtest->realurl, rtest->displayurl);
|
|
break;
|
|
case RTR_WHITELISTED:
|
|
ck_assert_msg(!ctx.num_viruses,
|
|
"this should be whitelisted, realURL: %s, displayURL: %s",
|
|
rtest->realurl, rtest->displayurl);
|
|
break;
|
|
case RTR_CLEAN:
|
|
ck_assert_msg(!ctx.num_viruses,
|
|
"this should be clean, realURL: %s, displayURL: %s",
|
|
rtest->realurl, rtest->displayurl);
|
|
break;
|
|
case RTR_BLACKLISTED:
|
|
if (!loaded_2)
|
|
ck_assert_msg(!ctx.num_viruses,
|
|
"this should be clean, realURL: %s, displayURL: %s",
|
|
rtest->realurl, rtest->displayurl);
|
|
else {
|
|
ck_assert_msg(ctx.num_viruses,
|
|
"this should be blacklisted, realURL: %s, displayURL: %s",
|
|
rtest->realurl, rtest->displayurl);
|
|
if (*ctx.virname) {
|
|
char *phishingFound = NULL;
|
|
char *detectionName = NULL;
|
|
if (strstr(rtest->realurl, "malware-test")) {
|
|
detectionName = "Heuristics.Safebrowsing.Suspected-malware_safebrowsing.clamav.net";
|
|
|
|
} else if (strstr(rtest->realurl, "phishing-test")) {
|
|
detectionName = "Heuristics.Safebrowsing.Suspected-phishing_safebrowsing.clamav.net";
|
|
}
|
|
ck_assert_msg(detectionName != NULL, "\n\t Blacklist test case error - malware-test or phishing-test not found in: %s\n", rtest->realurl);
|
|
phishingFound = strstr((const char *)*ctx.virname, detectionName);
|
|
ck_assert_msg(phishingFound != NULL, "\n\t should be: %s,\n\t but is: %s\n", detectionName, *ctx.virname);
|
|
}
|
|
}
|
|
break;
|
|
case RTR_INVALID_REGEX:
|
|
/* don't worry about it, this was tested in regex_list_match_test() */
|
|
break;
|
|
}
|
|
}
|
|
|
|
START_TEST(phishingScan_test)
|
|
{
|
|
do_phishing_test(&rtests[_i]);
|
|
}
|
|
END_TEST
|
|
|
|
START_TEST(phishingScan_test_allscan)
|
|
{
|
|
do_phishing_test_allscan(&rtests[_i]);
|
|
}
|
|
END_TEST
|
|
|
|
static struct uc {
|
|
const char *in;
|
|
const char *host;
|
|
const char *path;
|
|
} uc[] = {
|
|
{":example/%25%32%35", "example/", "%25"},
|
|
{":example/%25%32%35%25%32%35", "example/", "%25%25"},
|
|
{":example/abc%25%32%35asd", "example/", "abc%25asd"},
|
|
{":www.example.com/", "www.example.com/", ""},
|
|
{":%31%32%37%2e%30%2e%30%2e%31/%2E%73%65%63%75%72%65/%77%77%77%2e%65%78%61%6d%70%6c%65%2e%63%6f%6d/",
|
|
"127.0.0.1/", ".secure/www.example.com/"},
|
|
{":127.0.0.1/uploads/%20%20%20%20/.verify/.blah=abcd-ef=gh/",
|
|
"127.0.0.1/", "uploads/%20%20%20%20/.verify/.blah=abcd-ef=gh/"},
|
|
{"http://example%23.com/%61%40%62%252B",
|
|
"example%23.com/", "a@b+"},
|
|
{"http://example.com/blah/..", "example.com/", ""},
|
|
{"http://example.com/blah/../x", "example.com/", "x"},
|
|
{"http://example.com/./a", "example.com/", "a"}};
|
|
|
|
START_TEST(test_url_canon)
|
|
{
|
|
char urlbuff[1024 + 3];
|
|
char *host = NULL;
|
|
const char *path = NULL;
|
|
size_t host_len, path_len;
|
|
struct uc *u = &uc[_i];
|
|
|
|
cli_url_canon(u->in, strlen(u->in), urlbuff, sizeof(urlbuff), &host, &host_len, &path, &path_len);
|
|
ck_assert_msg(!!host && !!path, "null results\n");
|
|
ck_assert_msg(!strcmp(u->host, host), "host incorrect: %s\n", host);
|
|
ck_assert_msg(!strcmp(u->path, path), "path incorrect: %s\n", path);
|
|
}
|
|
END_TEST
|
|
|
|
static struct regex_test {
|
|
const char *regex;
|
|
const char *text;
|
|
int match;
|
|
} rg[] = {
|
|
{"\\.exe$", "test.exe", 1},
|
|
{"\\.exe$", "test.eXe", 0},
|
|
{"(?i)\\.exe$", "test.exe", 1},
|
|
{"(?i)\\.exe$", "test.eXe", 1}};
|
|
|
|
START_TEST(test_regexes)
|
|
{
|
|
regex_t reg;
|
|
struct regex_test *tst = &rg[_i];
|
|
int match;
|
|
|
|
ck_assert_msg(cli_regcomp(®, tst->regex, REG_EXTENDED | REG_NOSUB) == 0, "cli_regcomp");
|
|
match = (cli_regexec(®, tst->text, 0, NULL, 0) == REG_NOMATCH) ? 0 : 1;
|
|
ck_assert_msg(match == tst->match, "cli_regexec failed for %s and %s\n", tst->regex, tst->text);
|
|
cli_regfree(®);
|
|
}
|
|
END_TEST
|
|
|
|
START_TEST(phishing_fake_test)
|
|
{
|
|
char buf[4096];
|
|
FILE *f = fdopen(open_testfile("input/daily.pdb"), "r");
|
|
ck_assert_msg(!!f, "fopen daily.pdb");
|
|
while (fgets(buf, sizeof(buf), f)) {
|
|
struct rtest rtest;
|
|
const char *pdb = strchr(buf, ':');
|
|
ck_assert_msg(!!pdb, "missing : in pdb");
|
|
rtest.realurl = pdb;
|
|
rtest.displayurl = pdb;
|
|
rtest.result = RTR_CLEAN;
|
|
do_phishing_test(&rtest);
|
|
rtest.realurl = "http://fake.example.com";
|
|
rtest.result = 0;
|
|
do_phishing_test(&rtest);
|
|
}
|
|
fclose(f);
|
|
}
|
|
END_TEST
|
|
|
|
START_TEST(phishing_fake_test_allscan)
|
|
{
|
|
char buf[4096];
|
|
FILE *f = fdopen(open_testfile("input/daily.pdb"), "r");
|
|
ck_assert_msg(!!f, "fopen daily.pdb");
|
|
while (fgets(buf, sizeof(buf), f)) {
|
|
struct rtest rtest;
|
|
const char *pdb = strchr(buf, ':');
|
|
ck_assert_msg(!!pdb, "missing : in pdb");
|
|
rtest.realurl = pdb;
|
|
rtest.displayurl = pdb;
|
|
rtest.result = RTR_CLEAN;
|
|
do_phishing_test_allscan(&rtest);
|
|
rtest.realurl = "http://fake.example.com";
|
|
rtest.result = 0;
|
|
do_phishing_test_allscan(&rtest);
|
|
}
|
|
fclose(f);
|
|
}
|
|
END_TEST
|
|
|
|
Suite *test_regex_suite(void)
|
|
{
|
|
Suite *s = suite_create("regex");
|
|
TCase *tc_api, *tc_matching, *tc_phish, *tc_phish2, *tc_regex;
|
|
|
|
tc_api = tcase_create("cli_regex2suffix");
|
|
suite_add_tcase(s, tc_api);
|
|
tcase_add_checked_fixture(tc_api, setup, teardown);
|
|
tcase_add_test(tc_api, empty);
|
|
tcase_add_test(tc_api, one);
|
|
|
|
tcase_add_loop_test(tc_api, test_suffix, 0, sizeof(tests) / sizeof(tests[0]));
|
|
|
|
tc_matching = tcase_create("regex_list");
|
|
suite_add_tcase(s, tc_matching);
|
|
tcase_add_checked_fixture(tc_matching, rsetup, rteardown);
|
|
|
|
tcase_add_loop_test(tc_matching, regex_list_match_test, 0, sizeof(rtests) / sizeof(rtests[0]));
|
|
|
|
tc_phish = tcase_create("phishingScan");
|
|
suite_add_tcase(s, tc_phish);
|
|
tcase_add_unchecked_fixture(tc_phish, psetup, pteardown);
|
|
|
|
tcase_add_loop_test(tc_phish, phishingScan_test, 0, sizeof(rtests) / sizeof(rtests[0]));
|
|
tcase_add_loop_test(tc_phish, phishingScan_test_allscan, 0, sizeof(rtests) / sizeof(rtests[0]));
|
|
|
|
tcase_add_test(tc_phish, phishing_fake_test);
|
|
tcase_add_test(tc_phish, phishing_fake_test_allscan);
|
|
|
|
tc_phish2 = tcase_create("phishingScan with 2 dbs");
|
|
suite_add_tcase(s, tc_phish2);
|
|
tcase_add_unchecked_fixture(tc_phish2, psetup2, pteardown);
|
|
|
|
tcase_add_loop_test(tc_phish2, phishingScan_test, 0, sizeof(rtests) / sizeof(rtests[0]));
|
|
tcase_add_loop_test(tc_phish2, phishingScan_test_allscan, 0, sizeof(rtests) / sizeof(rtests[0]));
|
|
|
|
tcase_add_test(tc_phish2, phishing_fake_test);
|
|
tcase_add_test(tc_phish2, phishing_fake_test_allscan);
|
|
|
|
tcase_add_loop_test(tc_phish, test_url_canon, 0, sizeof(uc) / sizeof(uc[0]));
|
|
|
|
tc_regex = tcase_create("cli_regcomp/execute");
|
|
suite_add_tcase(s, tc_regex);
|
|
|
|
tcase_add_loop_test(tc_regex, test_regexes, 0, sizeof(rg) / sizeof(rg[0]));
|
|
|
|
return s;
|
|
}
|