mirror of
https://github.com/AFLplusplus/AFLplusplus.git
synced 2025-06-11 01:31:37 +00:00
remove ALTERNATIVE_TOKENIZE
This commit is contained in:
vanhauser-thc
parent
ca063c92d2
commit
7eaef449a1
@ -22,7 +22,6 @@ extern "C" {
|
||||
|
||||
#define AUTOTOKENS_DEBUG 0
|
||||
#define AUTOTOKENS_ONLY_FAV 0
|
||||
#define AUTOTOKENS_ALTERNATIVE_TOKENIZE 0
|
||||
#define AUTOTOKENS_CHANGE_MIN 8
|
||||
#define AUTOTOKENS_CHANGE_MAX 64
|
||||
#define AUTOTOKENS_WHITESPACE " "
|
||||
@ -60,7 +59,6 @@ typedef struct my_mutator {
|
||||
static afl_state *afl_ptr;
|
||||
static int debug = AUTOTOKENS_DEBUG;
|
||||
static int only_fav = AUTOTOKENS_ONLY_FAV;
|
||||
static int alternative_tokenize = AUTOTOKENS_ALTERNATIVE_TOKENIZE;
|
||||
static int learn_dictionary_tokens = AUTOTOKENS_LEARN_DICT;
|
||||
static int fuzz_count_shift = AUTOTOKENS_FUZZ_COUNT_SHIFT;
|
||||
static int create_from_thin_air = AUTOTOKENS_CREATE_FROM_THIN_AIR;
|
||||
@ -142,7 +140,7 @@ extern "C" size_t afl_custom_fuzz(my_mutator_t *data, u8 *buf, size_t buf_size,
|
||||
|
||||
(void)(data);
|
||||
|
||||
if (s == NULL) {
|
||||
if (unlikely(s == NULL)) {
|
||||
|
||||
*out_buf = NULL;
|
||||
return 0;
|
||||
@ -183,9 +181,8 @@ extern "C" size_t afl_custom_fuzz(my_mutator_t *data, u8 *buf, size_t buf_size,
|
||||
} while (unlikely(
|
||||
|
||||
new_item == cur_item ||
|
||||
(!alternative_tokenize &&
|
||||
((whitespace_ids < new_item && whitespace_ids >= cur_item) ||
|
||||
(whitespace_ids >= new_item && whitespace_ids < cur_item)))));
|
||||
(whitespace_ids >= new_item && whitespace_ids < cur_item))));
|
||||
|
||||
DEBUGF(stderr, "MUT: %u -> %u\n", cur_item, new_item);
|
||||
m[pos] = new_item;
|
||||
@ -200,15 +197,13 @@ extern "C" size_t afl_custom_fuzz(my_mutator_t *data, u8 *buf, size_t buf_size,
|
||||
|
||||
new_item = rand_below(afl_ptr, current_id);
|
||||
|
||||
} while (unlikely(!alternative_tokenize && new_item >= whitespace_ids));
|
||||
} while (unlikely(new_item >= whitespace_ids));
|
||||
|
||||
u32 pos = rand_below(afl_ptr, m_size + 1);
|
||||
m.insert(m.begin() + pos, new_item);
|
||||
++m_size;
|
||||
DEBUGF(stderr, "INS: %u at %u\n", new_item, pos);
|
||||
|
||||
if (likely(!alternative_tokenize)) {
|
||||
|
||||
// if we insert an identifier or string we might need whitespace
|
||||
if (id_to_token[new_item].size() > 1) {
|
||||
|
||||
@ -234,8 +229,6 @@ extern "C" size_t afl_custom_fuzz(my_mutator_t *data, u8 *buf, size_t buf_size,
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
break;
|
||||
|
||||
}
|
||||
@ -290,8 +283,6 @@ extern "C" size_t afl_custom_fuzz(my_mutator_t *data, u8 *buf, size_t buf_size,
|
||||
|
||||
}
|
||||
|
||||
if (likely(!alternative_tokenize)) {
|
||||
|
||||
// do we need a whitespace/token at the beginning?
|
||||
if (dst_off && id_to_token[m[dst_off - 1]].size() > 1 &&
|
||||
id_to_token[m[dst_off]].size() > 1) {
|
||||
@ -311,8 +302,6 @@ extern "C" size_t afl_custom_fuzz(my_mutator_t *data, u8 *buf, size_t buf_size,
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
break;
|
||||
|
||||
}
|
||||
@ -332,8 +321,7 @@ extern "C" size_t afl_custom_fuzz(my_mutator_t *data, u8 *buf, size_t buf_size,
|
||||
|
||||
// if what we delete will result in a missing whitespace/token,
|
||||
// instead of deleting we switch the item to a whitespace or token.
|
||||
if (likely(!alternative_tokenize) && pos && pos + 1 < m_size &&
|
||||
id_to_token[m[pos - 1]].size() > 1 &&
|
||||
if (pos && pos + 1 < m_size && id_to_token[m[pos - 1]].size() > 1 &&
|
||||
id_to_token[m[pos + 1]].size() > 1) {
|
||||
|
||||
m[pos] = good_whitespace_or_singleval();
|
||||
@ -362,17 +350,11 @@ extern "C" size_t afl_custom_fuzz(my_mutator_t *data, u8 *buf, size_t buf_size,
|
||||
|
||||
}
|
||||
|
||||
u32 m_size_1 = m_size - 1;
|
||||
output = "";
|
||||
|
||||
for (i = 0; i < m_size; ++i) {
|
||||
|
||||
output += id_to_token[m[i]];
|
||||
if (unlikely(alternative_tokenize && i < m_size_1)) {
|
||||
|
||||
output += whitespace;
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -725,8 +707,6 @@ extern "C" unsigned char afl_custom_queue_get(void *data,
|
||||
|
||||
DEBUGF(stderr, "START!\n");
|
||||
|
||||
if (likely(!alternative_tokenize)) {
|
||||
|
||||
while (my_search_string(cur, ende, &match_begin, &match_end)) {
|
||||
|
||||
prev = cur;
|
||||
@ -835,245 +815,12 @@ extern "C" unsigned char afl_custom_queue_get(void *data,
|
||||
|
||||
}
|
||||
|
||||
} else {
|
||||
|
||||
// alternative tokenize
|
||||
while (my_search_string(cur, ende, &match_begin, &match_end)) {
|
||||
|
||||
prev = cur;
|
||||
found = match_begin;
|
||||
cur = match_end;
|
||||
IFDEBUG {
|
||||
|
||||
string foo(match_begin, match_end);
|
||||
DEBUGF(stderr,
|
||||
"string %s found at start %lu offset %lu continue at %lu\n",
|
||||
foo.c_str(), prev - input.begin(), found - prev,
|
||||
cur - input.begin());
|
||||
|
||||
}
|
||||
|
||||
if (prev < found) { // there are items between search start and find
|
||||
|
||||
sregex_token_iterator it{prev, found, regex_whitespace, -1};
|
||||
vector<std::string> tokenized{it, {}};
|
||||
tokenized.erase(std::remove_if(tokenized.begin(), tokenized.end(),
|
||||
[](std::string const &s) {
|
||||
|
||||
return s.size() == 0;
|
||||
|
||||
}),
|
||||
|
||||
tokenized.end());
|
||||
tokens.reserve(tokens.size() + tokenized.size() * 2 + 1);
|
||||
|
||||
IFDEBUG {
|
||||
|
||||
DEBUGF(stderr, "tokens1: %lu input size: %lu\n", tokenized.size(),
|
||||
input.size());
|
||||
for (auto x : tokenized) {
|
||||
|
||||
cerr << x << endl;
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
for (auto token : tokenized) {
|
||||
|
||||
string::const_iterator c = token.begin(), e = token.end(), f, p;
|
||||
smatch m;
|
||||
|
||||
while (regex_search(c, e, m, regex_word)) {
|
||||
|
||||
p = c;
|
||||
f = m[0].first;
|
||||
c = m[0].second;
|
||||
if (p < f) {
|
||||
|
||||
// there are items between search start and find
|
||||
while (p < f) {
|
||||
|
||||
IFDEBUG {
|
||||
|
||||
string foo(p, p + 1);
|
||||
DEBUGF(stderr, "before string: \"%s\"\n", foo.c_str());
|
||||
|
||||
}
|
||||
|
||||
tokens.push_back(std::string(p, p + 1));
|
||||
++p;
|
||||
|
||||
}
|
||||
|
||||
IFDEBUG {
|
||||
|
||||
string foo(p, f);
|
||||
DEBUGF(stderr, "before string: \"%s\"\n", foo.c_str());
|
||||
tokens.push_back(std::string(p, f));
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
DEBUGF(stderr,
|
||||
"SUBstring \"%s\" found at start %lu offset %lu continue "
|
||||
"at %lu\n",
|
||||
m[0].str().c_str(), p - input.begin(), m.position(),
|
||||
c - token.begin());
|
||||
tokens.push_back(m[0].str());
|
||||
|
||||
}
|
||||
|
||||
if (c < e) {
|
||||
|
||||
while (c < e) {
|
||||
|
||||
IFDEBUG {
|
||||
|
||||
string foo(c, c + 1);
|
||||
DEBUGF(stderr, "after string: \"%s\"\n", foo.c_str());
|
||||
|
||||
}
|
||||
|
||||
tokens.push_back(std::string(c, c + 1));
|
||||
++c;
|
||||
|
||||
}
|
||||
|
||||
IFDEBUG {
|
||||
|
||||
string foo(c, e);
|
||||
DEBUGF(stderr, "after string: \"%s\"\n", foo.c_str());
|
||||
|
||||
}
|
||||
|
||||
tokens.push_back(std::string(c, e));
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
tokens.push_back(string(match_begin, match_end));
|
||||
|
||||
}
|
||||
|
||||
if (cur < ende) {
|
||||
|
||||
sregex_token_iterator it{cur, ende, regex_whitespace, -1};
|
||||
vector<std::string> tokenized{it, {}};
|
||||
tokenized.erase(
|
||||
std::remove_if(tokenized.begin(), tokenized.end(),
|
||||
[](std::string const &s) { return s.size() == 0; }),
|
||||
tokenized.end());
|
||||
tokens.reserve(tokens.size() + tokenized.size() * 2 + 1);
|
||||
|
||||
IFDEBUG {
|
||||
|
||||
DEBUGF(stderr, "tokens2: %lu input size: %lu\n", tokenized.size(),
|
||||
input.size());
|
||||
for (auto x : tokenized) {
|
||||
|
||||
cerr << x << endl;
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
for (auto token : tokenized) {
|
||||
|
||||
string::const_iterator c = token.begin(), e = token.end(), f, p;
|
||||
smatch m;
|
||||
|
||||
while (regex_search(c, e, m, regex_word)) {
|
||||
|
||||
p = c;
|
||||
f = m[0].first;
|
||||
c = m[0].second;
|
||||
if (p < f) {
|
||||
|
||||
// there are items between search start and find
|
||||
while (p < f) {
|
||||
|
||||
IFDEBUG {
|
||||
|
||||
string foo(p, p + 1);
|
||||
DEBUGF(stderr, "before string: \"%s\"\n", foo.c_str());
|
||||
|
||||
}
|
||||
|
||||
tokens.push_back(std::string(p, p + 1));
|
||||
++p;
|
||||
|
||||
}
|
||||
|
||||
IFDEBUG {
|
||||
|
||||
string foo(p, f);
|
||||
DEBUGF(stderr, "before string: \"%s\"\n", foo.c_str());
|
||||
|
||||
}
|
||||
|
||||
tokens.push_back(std::string(p, f));
|
||||
|
||||
}
|
||||
|
||||
DEBUGF(stderr,
|
||||
"SUB2string \"%s\" found at start %lu offset %lu continue "
|
||||
"at %lu\n",
|
||||
m[0].str().c_str(), p - input.begin(), m.position(),
|
||||
c - token.begin());
|
||||
tokens.push_back(m[0].str());
|
||||
|
||||
}
|
||||
|
||||
if (c < e) {
|
||||
|
||||
while (c < e) {
|
||||
|
||||
IFDEBUG {
|
||||
|
||||
string foo(c, c + 1);
|
||||
DEBUGF(stderr, "after string: \"%s\"\n", foo.c_str());
|
||||
|
||||
}
|
||||
|
||||
tokens.push_back(std::string(c, c + 1));
|
||||
++c;
|
||||
|
||||
}
|
||||
|
||||
IFDEBUG {
|
||||
|
||||
string foo(c, e);
|
||||
DEBUGF(stderr, "after string: \"%s\"\n", foo.c_str());
|
||||
|
||||
}
|
||||
|
||||
tokens.push_back(std::string(c, e));
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
IFDEBUG {
|
||||
|
||||
DEBUGF(stderr, "DUMPING TOKENS:\n");
|
||||
u32 size_1 = tokens.size() - 1;
|
||||
for (u32 i = 0; i < tokens.size(); ++i) {
|
||||
|
||||
DEBUGF(stderr, "%s", tokens[i].c_str());
|
||||
if (unlikely(alternative_tokenize && i < size_1)) {
|
||||
|
||||
DEBUGF(stderr, "%s", whitespace.c_str());
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -1157,7 +904,6 @@ extern "C" my_mutator_t *afl_custom_init(afl_state *afl, unsigned int seed) {
|
||||
if (getenv("AUTOTOKENS_DEBUG")) { debug = 1; }
|
||||
if (getenv("AUTOTOKENS_ONLY_FAV")) { only_fav = 1; }
|
||||
if (getenv("AUTOTOKENS_CREATE_FROM_THIN_AIR")) { create_from_thin_air = 1; }
|
||||
if (getenv("AUTOTOKENS_ALTERNATIVE_TOKENIZE")) { alternative_tokenize = 1; }
|
||||
|
||||
if (getenv("AUTOTOKENS_LEARN_DICT")) {
|
||||
|
||||
@ -1180,14 +926,22 @@ extern "C" my_mutator_t *afl_custom_init(afl_state *afl, unsigned int seed) {
|
||||
if (getenv("AUTOTOKENS_CHANGE_MIN")) {
|
||||
|
||||
change_min = atoi(getenv("AUTOTOKENS_CHANGE_MIN"));
|
||||
if (change_min < 1 || change_min > 256) { change_min = AUTOTOKENS_CHANGE_MIN; }
|
||||
if (change_min < 1 || change_min > 256) {
|
||||
|
||||
change_min = AUTOTOKENS_CHANGE_MIN;
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
if (getenv("AUTOTOKENS_CHANGE_MAX")) {
|
||||
|
||||
change_max = atoi(getenv("AUTOTOKENS_CHANGE_MAX"));
|
||||
if (change_max < 1 || change_max > 4096) { change_max = AUTOTOKENS_CHANGE_MAX; }
|
||||
if (change_max < 1 || change_max > 4096) {
|
||||
|
||||
change_max = AUTOTOKENS_CHANGE_MAX;
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -1212,8 +966,6 @@ extern "C" my_mutator_t *afl_custom_init(afl_state *afl, unsigned int seed) {
|
||||
// set common whitespace tokens
|
||||
// we deliberately do not put uncommon ones here to these will count as
|
||||
// identifier tokens.
|
||||
if (!alternative_tokenize) {
|
||||
|
||||
token_to_id[" "] = current_id;
|
||||
id_to_token[current_id] = " ";
|
||||
++current_id;
|
||||
@ -1258,8 +1010,6 @@ extern "C" my_mutator_t *afl_custom_init(afl_state *afl, unsigned int seed) {
|
||||
id_to_token[current_id] = "'";
|
||||
++current_id;
|
||||
|
||||
}
|
||||
|
||||
return data;
|
||||
|
||||
}
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user