cdo/regex-engine
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Move regex_t into parse.h and rename to parse_tree_t

Browse Source
This commit is contained in:
Camden Dixie O'Brien
2024-11-03 12:20:17 +00:00
parent 1f248ad4cd
commit 656726a8c1
10 changed files with 315 additions and 328 deletions
Download Patch File Download Diff File Expand all files Collapse all files

249
tests/construct_tests.c
View File

@@ -34,13 +34,13 @@ static bool accepts(const fsa_t *nfa, const char *input)
static void test_empty_expression(void)
{
regex_term_t *terms = malloc(1 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_NONE;
terms[0].type = REGEX_TERM_EMPTY;
regex_sequence_t *alternatives = malloc(1 * sizeof(regex_sequence_t));
parse_term_t *terms = malloc(1 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_NONE;
terms[0].type = PARSE_TERM_EMPTY;
parse_sequence_t *alternatives = malloc(1 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 1;
alternatives[0].contents = terms;
const regex_t regex
const parse_tree_t regex
= { .count = 1, .capacity = 1, .contents = alternatives };
fsa_t fsa;
@@ -49,19 +49,19 @@ static void test_empty_expression(void)
ASSERT_TRUE(accepts(&fsa, ""));
ASSERT_FALSE(accepts(&fsa, "a"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
static void test_wildcard(void)
{
regex_term_t *terms = malloc(1 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_NONE;
terms[0].type = REGEX_TERM_WILDCARD;
regex_sequence_t *alternatives = malloc(1 * sizeof(regex_sequence_t));
parse_term_t *terms = malloc(1 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_NONE;
terms[0].type = PARSE_TERM_WILDCARD;
parse_sequence_t *alternatives = malloc(1 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 1;
alternatives[0].contents = terms;
const regex_t regex
const parse_tree_t regex
= { .count = 1, .capacity = 1, .contents = alternatives };
fsa_t fsa;
@@ -74,20 +74,20 @@ static void test_wildcard(void)
ASSERT_FALSE(accepts(&fsa, ""));
ASSERT_FALSE(accepts(&fsa, "aa"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
static void test_literal_expression(void)
{
regex_term_t *terms = malloc(1 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_NONE;
terms[0].type = REGEX_TERM_LITERAL;
parse_term_t *terms = malloc(1 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_NONE;
terms[0].type = PARSE_TERM_LITERAL;
terms[0].literal = 'a';
regex_sequence_t *alternatives = malloc(1 * sizeof(regex_sequence_t));
parse_sequence_t *alternatives = malloc(1 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 1;
alternatives[0].contents = terms;
const regex_t regex
const parse_tree_t regex
= { .count = 1, .capacity = 1, .contents = alternatives };
fsa_t fsa;
@@ -96,26 +96,27 @@ static void test_literal_expression(void)
ASSERT_TRUE(accepts(&fsa, "a"));
ASSERT_FALSE(accepts(&fsa, "b"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
static void test_sequence(void)
{
regex_term_t *terms = malloc(3 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_NONE;
terms[0].type = REGEX_TERM_LITERAL;
parse_term_t *terms = malloc(3 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_NONE;
terms[0].type = PARSE_TERM_LITERAL;
terms[0].literal = 'a';
terms[1].quantifier = REGEX_QUANTIFIER_NONE;
terms[1].type = REGEX_TERM_LITERAL;
terms[1].quantifier = PARSE_QUANTIFIER_NONE;
terms[1].type = PARSE_TERM_LITERAL;
terms[1].literal = 'b';
terms[2].quantifier = REGEX_QUANTIFIER_NONE;
terms[2].type = REGEX_TERM_LITERAL;
terms[2].quantifier = PARSE_QUANTIFIER_NONE;
terms[2].type = PARSE_TERM_LITERAL;
terms[2].literal = 'c';
regex_sequence_t *alternatives = malloc(1 * sizeof(regex_sequence_t));
parse_sequence_t *alternatives = malloc(1 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 3;
alternatives[0].contents = terms;
regex_t regex = { .count = 1, .capacity = 1, .contents = alternatives };
parse_tree_t regex
= { .count = 1, .capacity = 1, .contents = alternatives };
fsa_t fsa;
construct_nfa(&regex, &fsa);
@@ -126,24 +127,25 @@ static void test_sequence(void)
ASSERT_FALSE(accepts(&fsa, "d"));
ASSERT_FALSE(accepts(&fsa, "abcd"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
static void test_union(void)
{
const char *literals = "abc";
regex_sequence_t *alternatives = malloc(3 * sizeof(regex_sequence_t));
parse_sequence_t *alternatives = malloc(3 * sizeof(parse_sequence_t));
for (int i = 0; i < 3; ++i) {
regex_term_t *terms = malloc(1 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_NONE;
terms[0].type = REGEX_TERM_LITERAL;
parse_term_t *terms = malloc(1 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_NONE;
terms[0].type = PARSE_TERM_LITERAL;
terms[0].literal = literals[i];
alternatives[i].count = alternatives[i].capacity = 1;
alternatives[i].contents = terms;
}
regex_t regex = { .count = 3, .capacity = 3, .contents = alternatives };
parse_tree_t regex
= { .count = 3, .capacity = 3, .contents = alternatives };
fsa_t fsa;
construct_nfa(&regex, &fsa);
@@ -154,20 +156,21 @@ static void test_union(void)
ASSERT_FALSE(accepts(&fsa, "d"));
ASSERT_FALSE(accepts(&fsa, "aa"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
static void test_star(void)
{
regex_term_t *terms = malloc(1 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_STAR;
terms[0].type = REGEX_TERM_LITERAL;
parse_term_t *terms = malloc(1 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_STAR;
terms[0].type = PARSE_TERM_LITERAL;
terms[0].literal = 'a';
regex_sequence_t *alternatives = malloc(1 * sizeof(regex_sequence_t));
parse_sequence_t *alternatives = malloc(1 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 1;
alternatives[0].contents = terms;
regex_t regex = { .count = 1, .capacity = 1, .contents = alternatives };
parse_tree_t regex
= { .count = 1, .capacity = 1, .contents = alternatives };
fsa_t fsa;
construct_nfa(&regex, &fsa);
@@ -177,20 +180,21 @@ static void test_star(void)
ASSERT_TRUE(accepts(&fsa, "aaaaaa"));
ASSERT_FALSE(accepts(&fsa, "b"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
static void test_plus(void)
{
regex_term_t *terms = malloc(1 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_PLUS;
terms[0].type = REGEX_TERM_LITERAL;
parse_term_t *terms = malloc(1 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_PLUS;
terms[0].type = PARSE_TERM_LITERAL;
terms[0].literal = 'a';
regex_sequence_t *alternatives = malloc(1 * sizeof(regex_sequence_t));
parse_sequence_t *alternatives = malloc(1 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 1;
alternatives[0].contents = terms;
regex_t regex = { .count = 1, .capacity = 1, .contents = alternatives };
parse_tree_t regex
= { .count = 1, .capacity = 1, .contents = alternatives };
fsa_t fsa;
construct_nfa(&regex, &fsa);
@@ -200,20 +204,21 @@ static void test_plus(void)
ASSERT_FALSE(accepts(&fsa, ""));
ASSERT_FALSE(accepts(&fsa, "b"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
static void test_qmark(void)
{
regex_term_t *terms = malloc(1 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_QMARK;
terms[0].type = REGEX_TERM_LITERAL;
parse_term_t *terms = malloc(1 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_QMARK;
terms[0].type = PARSE_TERM_LITERAL;
terms[0].literal = 'a';
regex_sequence_t *alternatives = malloc(1 * sizeof(regex_sequence_t));
parse_sequence_t *alternatives = malloc(1 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 1;
alternatives[0].contents = terms;
regex_t regex = { .count = 1, .capacity = 1, .contents = alternatives };
parse_tree_t regex
= { .count = 1, .capacity = 1, .contents = alternatives };
fsa_t fsa;
construct_nfa(&regex, &fsa);
@@ -223,29 +228,30 @@ static void test_qmark(void)
ASSERT_FALSE(accepts(&fsa, "aa"));
ASSERT_FALSE(accepts(&fsa, "b"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
static void test_subexpression(void)
{
regex_term_t *inner_terms = malloc(1 * sizeof(regex_term_t));
inner_terms[0].quantifier = REGEX_QUANTIFIER_NONE;
inner_terms[0].type = REGEX_TERM_LITERAL;
parse_term_t *inner_terms = malloc(1 * sizeof(parse_term_t));
inner_terms[0].quantifier = PARSE_QUANTIFIER_NONE;
inner_terms[0].type = PARSE_TERM_LITERAL;
inner_terms[0].literal = 'a';
regex_sequence_t *inner_alternatives
= malloc(1 * sizeof(regex_sequence_t));
parse_sequence_t *inner_alternatives
= malloc(1 * sizeof(parse_sequence_t));
inner_alternatives[0].count = inner_alternatives[0].capacity = 1;
inner_alternatives[0].contents = inner_terms;
regex_term_t *terms = malloc(1 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_NONE;
terms[0].type = REGEX_TERM_SUBEXPR;
parse_term_t *terms = malloc(1 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_NONE;
terms[0].type = PARSE_TERM_SUBEXPR;
terms[0].subexpr.count = terms[0].subexpr.capacity = 1;
terms[0].subexpr.contents = inner_alternatives;
regex_sequence_t *alternatives = malloc(1 * sizeof(regex_sequence_t));
parse_sequence_t *alternatives = malloc(1 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 1;
alternatives[0].contents = terms;
regex_t regex = { .count = 1, .capacity = 1, .contents = alternatives };
parse_tree_t regex
= { .count = 1, .capacity = 1, .contents = alternatives };
fsa_t fsa;
construct_nfa(&regex, &fsa);
@@ -253,7 +259,7 @@ static void test_subexpression(void)
ASSERT_TRUE(accepts(&fsa, "a"));
ASSERT_FALSE(accepts(&fsa, "b"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
@@ -263,16 +269,16 @@ static void test_class(void)
class_contents[0] = 'a';
class_contents[1] = 'b';
class_contents[2] = 'c';
regex_term_t *terms = malloc(1 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_NONE;
terms[0].type = REGEX_TERM_CLASS;
parse_term_t *terms = malloc(1 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_NONE;
terms[0].type = PARSE_TERM_CLASS;
terms[0].class.negated = false;
terms[0].class.count = terms[0].class.capacity = 3;
terms[0].class.contents = class_contents;
regex_sequence_t *alternatives = malloc(1 * sizeof(regex_sequence_t));
parse_sequence_t *alternatives = malloc(1 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 1;
alternatives[0].contents = terms;
const regex_t regex
const parse_tree_t regex
= { .count = 1, .capacity = 1, .contents = alternatives };
fsa_t fsa;
@@ -285,7 +291,7 @@ static void test_class(void)
ASSERT_FALSE(accepts(&fsa, "aa"));
ASSERT_FALSE(accepts(&fsa, "d"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
@@ -295,16 +301,16 @@ static void test_negated_class(void)
class_contents[0] = 'a';
class_contents[1] = 'b';
class_contents[2] = 'c';
regex_term_t *terms = malloc(1 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_NONE;
terms[0].type = REGEX_TERM_CLASS;
parse_term_t *terms = malloc(1 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_NONE;
terms[0].type = PARSE_TERM_CLASS;
terms[0].class.negated = true;
terms[0].class.count = terms[0].class.capacity = 3;
terms[0].class.contents = class_contents;
regex_sequence_t *alternatives = malloc(1 * sizeof(regex_sequence_t));
parse_sequence_t *alternatives = malloc(1 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 1;
alternatives[0].contents = terms;
const regex_t regex
const parse_tree_t regex
= { .count = 1, .capacity = 1, .contents = alternatives };
fsa_t fsa;
@@ -318,42 +324,43 @@ static void test_negated_class(void)
ASSERT_FALSE(accepts(&fsa, ""));
ASSERT_FALSE(accepts(&fsa, "aa"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
static void test_sequence_containing_starred_union(void)
{
// ab(c|d)*
regex_term_t *inner_terms0 = malloc(1 * sizeof(regex_term_t));
inner_terms0[0].quantifier = REGEX_QUANTIFIER_NONE;
inner_terms0[0].type = REGEX_TERM_LITERAL;
parse_term_t *inner_terms0 = malloc(1 * sizeof(parse_term_t));
inner_terms0[0].quantifier = PARSE_QUANTIFIER_NONE;
inner_terms0[0].type = PARSE_TERM_LITERAL;
inner_terms0[0].literal = 'c';
regex_term_t *inner_terms1 = malloc(1 * sizeof(regex_term_t));
inner_terms1[0].quantifier = REGEX_QUANTIFIER_NONE;
inner_terms1[0].type = REGEX_TERM_LITERAL;
parse_term_t *inner_terms1 = malloc(1 * sizeof(parse_term_t));
inner_terms1[0].quantifier = PARSE_QUANTIFIER_NONE;
inner_terms1[0].type = PARSE_TERM_LITERAL;
inner_terms1[0].literal = 'd';
regex_sequence_t *inner_alternatives
= malloc(2 * sizeof(regex_sequence_t));
parse_sequence_t *inner_alternatives
= malloc(2 * sizeof(parse_sequence_t));
inner_alternatives[0].count = inner_alternatives[0].capacity = 1;
inner_alternatives[0].contents = inner_terms0;
inner_alternatives[1].count = inner_alternatives[1].capacity = 1;
inner_alternatives[1].contents = inner_terms1;
regex_term_t *terms = malloc(3 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_NONE;
terms[0].type = REGEX_TERM_LITERAL;
parse_term_t *terms = malloc(3 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_NONE;
terms[0].type = PARSE_TERM_LITERAL;
terms[0].literal = 'a';
terms[1].quantifier = REGEX_QUANTIFIER_NONE;
terms[1].type = REGEX_TERM_LITERAL;
terms[1].quantifier = PARSE_QUANTIFIER_NONE;
terms[1].type = PARSE_TERM_LITERAL;
terms[1].literal = 'b';
terms[2].quantifier = REGEX_QUANTIFIER_STAR;
terms[2].type = REGEX_TERM_SUBEXPR;
terms[2].quantifier = PARSE_QUANTIFIER_STAR;
terms[2].type = PARSE_TERM_SUBEXPR;
terms[2].subexpr.count = terms[2].subexpr.capacity = 2;
terms[2].subexpr.contents = inner_alternatives;
regex_sequence_t *alternatives = malloc(1 * sizeof(regex_sequence_t));
parse_sequence_t *alternatives = malloc(1 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 3;
alternatives[0].contents = terms;
regex_t regex = { .count = 1, .capacity = 1, .contents = alternatives };
parse_tree_t regex
= { .count = 1, .capacity = 1, .contents = alternatives };
fsa_t fsa;
construct_nfa(&regex, &fsa);
@@ -369,7 +376,7 @@ static void test_sequence_containing_starred_union(void)
ASSERT_FALSE(accepts(&fsa, "d"));
ASSERT_FALSE(accepts(&fsa, "foo"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
@@ -377,23 +384,24 @@ static void
test_union_of_single_term_and_sequence_containing_starred_term(void)
{
// a|b*c
regex_term_t *terms0 = malloc(1 * sizeof(regex_term_t));
terms0[0].quantifier = REGEX_QUANTIFIER_NONE;
terms0[0].type = REGEX_TERM_LITERAL;
parse_term_t *terms0 = malloc(1 * sizeof(parse_term_t));
terms0[0].quantifier = PARSE_QUANTIFIER_NONE;
terms0[0].type = PARSE_TERM_LITERAL;
terms0[0].literal = 'a';
regex_term_t *terms1 = malloc(2 * sizeof(regex_term_t));
terms1[0].quantifier = REGEX_QUANTIFIER_STAR;
terms1[0].type = REGEX_TERM_LITERAL;
parse_term_t *terms1 = malloc(2 * sizeof(parse_term_t));
terms1[0].quantifier = PARSE_QUANTIFIER_STAR;
terms1[0].type = PARSE_TERM_LITERAL;
terms1[0].literal = 'b';
terms1[1].quantifier = REGEX_QUANTIFIER_NONE;
terms1[1].type = REGEX_TERM_LITERAL;
terms1[1].quantifier = PARSE_QUANTIFIER_NONE;
terms1[1].type = PARSE_TERM_LITERAL;
terms1[1].literal = 'c';
regex_sequence_t *alternatives = malloc(2 * sizeof(regex_sequence_t));
parse_sequence_t *alternatives = malloc(2 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 1;
alternatives[0].contents = terms0;
alternatives[1].count = alternatives[1].capacity = 2;
alternatives[1].contents = terms1;
regex_t regex = { .count = 2, .capacity = 2, .contents = alternatives };
parse_tree_t regex
= { .count = 2, .capacity = 2, .contents = alternatives };
fsa_t fsa;
construct_nfa(&regex, &fsa);
@@ -405,38 +413,39 @@ test_union_of_single_term_and_sequence_containing_starred_term(void)
ASSERT_FALSE(accepts(&fsa, "foo"));
ASSERT_FALSE(accepts(&fsa, "ba"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}
static void test_sequence_of_subexpr_a_or_empty_and_b(void)
{
// (a|ε)b
regex_term_t *inner_terms0 = malloc(1 * sizeof(regex_term_t));
inner_terms0[0].quantifier = REGEX_QUANTIFIER_NONE;
inner_terms0[0].type = REGEX_TERM_LITERAL;
parse_term_t *inner_terms0 = malloc(1 * sizeof(parse_term_t));
inner_terms0[0].quantifier = PARSE_QUANTIFIER_NONE;
inner_terms0[0].type = PARSE_TERM_LITERAL;
inner_terms0[0].literal = 'a';
regex_term_t *inner_terms1 = malloc(1 * sizeof(regex_term_t));
inner_terms1[0].quantifier = REGEX_QUANTIFIER_NONE;
inner_terms1[0].type = REGEX_TERM_EMPTY;
regex_sequence_t *inner_alternatives
= malloc(2 * sizeof(regex_sequence_t));
parse_term_t *inner_terms1 = malloc(1 * sizeof(parse_term_t));
inner_terms1[0].quantifier = PARSE_QUANTIFIER_NONE;
inner_terms1[0].type = PARSE_TERM_EMPTY;
parse_sequence_t *inner_alternatives
= malloc(2 * sizeof(parse_sequence_t));
inner_alternatives[0].count = inner_alternatives[0].capacity = 1;
inner_alternatives[0].contents = inner_terms0;
inner_alternatives[1].count = inner_alternatives[1].capacity = 1;
inner_alternatives[1].contents = inner_terms1;
regex_term_t *terms = malloc(2 * sizeof(regex_term_t));
terms[0].quantifier = REGEX_QUANTIFIER_NONE;
terms[0].type = REGEX_TERM_SUBEXPR;
parse_term_t *terms = malloc(2 * sizeof(parse_term_t));
terms[0].quantifier = PARSE_QUANTIFIER_NONE;
terms[0].type = PARSE_TERM_SUBEXPR;
terms[0].subexpr.count = terms[0].subexpr.capacity = 2;
terms[0].subexpr.contents = inner_alternatives;
terms[1].quantifier = REGEX_QUANTIFIER_NONE;
terms[1].type = REGEX_TERM_LITERAL;
terms[1].quantifier = PARSE_QUANTIFIER_NONE;
terms[1].type = PARSE_TERM_LITERAL;
terms[1].literal = 'b';
regex_sequence_t *alternatives = malloc(1 * sizeof(regex_sequence_t));
parse_sequence_t *alternatives = malloc(1 * sizeof(parse_sequence_t));
alternatives[0].count = alternatives[0].capacity = 2;
alternatives[0].contents = terms;
regex_t regex = { .count = 1, .capacity = 1, .contents = alternatives };
parse_tree_t regex
= { .count = 1, .capacity = 1, .contents = alternatives };
fsa_t fsa;
construct_nfa(&regex, &fsa);
@@ -446,7 +455,7 @@ static void test_sequence_of_subexpr_a_or_empty_and_b(void)
ASSERT_FALSE(accepts(&fsa, ""));
ASSERT_FALSE(accepts(&fsa, "a"));
regex_free(&regex);
parse_tree_free(&regex);
fsa_free(&fsa);
}