const HParser *init_parser(void)
{
// CORE
H_RULE (digit, h_ch_range(0x30, 0x39));
H_RULE (alpha, h_choice(h_ch_range(0x41, 0x5a), h_ch_range(0x61, 0x7a), NULL));
H_RULE (space, h_in((uint8_t *)" \t\n\r\f\v", 6));
// AUX.
H_RULE (plus, h_ch('+'));
H_RULE (slash, h_ch('/'));
H_ARULE(equals, h_ch('='));
H_ARULE(bsfdig, h_choice(alpha, digit, plus, slash, NULL));
H_ARULE(bsfdig_4bit, h_in((uint8_t *)"AEIMQUYcgkosw048", 16));
H_ARULE(bsfdig_2bit, h_in((uint8_t *)"AQgw", 4));
H_ARULE(base64_3, h_repeat_n(bsfdig, 4));
H_ARULE(base64_2, h_sequence(bsfdig, bsfdig, bsfdig_4bit, equals, NULL));
H_ARULE(base64_1, h_sequence(bsfdig, bsfdig_2bit, equals, equals, NULL));
H_ARULE(base64, h_sequence(h_many(base64_3),
h_optional(h_choice(base64_2,
base64_1, NULL)),
NULL));
H_ARULE(ws, h_many(space));
H_ARULE(document, h_sequence(ws, base64, ws, h_end_p(), NULL));
// BUG sometimes inputs that should just don't parse.
// It *seemed* to happen mostly with things like "bbbbaaaaBA==".
// Using less actions seemed to make it less likely.
return document;
}
static void test_bug118(void) {
// https://github.com/UpstandingHackers/hammer/issues/118
// Adapted from https://gist.github.com/mrdomino/c6bc91a7cb3b9817edb5
HParseResult* p;
const uint8_t *input = (uint8_t*)"\x69\x5A\x6A\x7A\x8A\x9A";
#define MY_ENDIAN (BIT_BIG_ENDIAN | BYTE_LITTLE_ENDIAN)
H_RULE(nibble, h_with_endianness(MY_ENDIAN, h_bits(4, false)));
H_RULE(sample, h_with_endianness(MY_ENDIAN, h_bits(10, false)));
#undef MY_ENDIAN
H_RULE(samples, h_sequence(h_repeat_n(sample, 3), h_ignore(h_bits(2, false)), NULL));
H_RULE(header_ok, h_sequence(nibble, nibble, NULL));
H_RULE(header_weird, h_sequence(nibble, nibble, nibble, NULL));
H_RULE(parser_ok, h_sequence(header_ok, samples, NULL));
H_RULE(parser_weird, h_sequence(header_weird, samples, NULL));
p = h_parse(parser_weird, input, 6);
g_check_cmp_int32(p->bit_length, ==, 44);
h_parse_result_free(p);
p = h_parse(parser_ok, input, 6);
g_check_cmp_int32(p->bit_length, ==, 40);
h_parse_result_free(p);
}
void init_parser(void)
{
// CORE
HParser *digit = h_ch_range(0x30, 0x39);
HParser *alpha = h_choice(h_ch_range(0x41, 0x5a), h_ch_range(0x61, 0x7a), NULL);
// AUX.
HParser *plus = h_ch('+');
HParser *slash = h_ch('/');
HParser *equals = h_ch('=');
HParser *bsfdig = h_choice(alpha, digit, plus, slash, NULL);
HParser *bsfdig_4bit = h_in((uint8_t *)"AEIMQUYcgkosw048", 16);
HParser *bsfdig_2bit = h_in((uint8_t *)"AQgw", 4);
HParser *base64_3 = h_repeat_n(bsfdig, 4);
HParser *base64_2 = h_sequence(bsfdig, bsfdig, bsfdig_4bit, equals, NULL);
HParser *base64_1 = h_sequence(bsfdig, bsfdig_2bit, equals, equals, NULL);
HParser *base64 = h_sequence(h_many(base64_3),
h_optional(h_choice(base64_2,
base64_1, NULL)),
NULL);
document = h_sequence(h_whitespace(base64), h_whitespace(h_end_p()), NULL);
}
