add more tests

lapla-cogito · Nov 30, 2024 · 425f7cb · 425f7cb
1 parent 6392092
commit 425f7cb
Show file tree

Hide file tree

Showing 4 changed files with 59 additions and 34 deletions.
diff --git a/README.md b/README.md
@@ -3,13 +3,14 @@
 A hobby regular expression engine in Rust.
 
 - supports Unicode characters
-- engines
+- implements 3 engines:
     - DFA-based engine
         - convert regex to NFA
         - convert NFA to DFA
     - VM-based engine
-        - caching
+        - cache
     - Derivative-based engine
+        - use Brzozowski's derivatives
 
 ## example
 
@@ -123,4 +124,4 @@ $ cargo bench
 
 ## license
 
-MIT
+MIT
diff --git a/src/automaton/dfa.rs b/src/automaton/dfa.rs
@@ -103,6 +103,7 @@ impl Dfa {
                 return false;
             }
         }
+
         self.accept().contains(&state)
     }
 }
@@ -111,31 +112,6 @@ impl Dfa {
 mod tests {
     use super::*;
 
-    #[test]
-    fn e_closure() {
-        let mut lexer = crate::lexer::Lexer::new("a|b*");
-        let mut parser = crate::parser::Parser::new(&mut lexer);
-        let nfa = crate::automaton::nfa::Nfa::new_from_node(
-            parser.parse().unwrap(),
-            &mut crate::automaton::nfa::NfaState::new(),
-        )
-        .unwrap();
-
-        let closure = nfa.epsilon_closure([nfa.start()].iter().cloned().collect());
-        assert_eq!(closure, [0, 2, 4, 5].iter().cloned().collect());
-
-        let mut lexer = crate::lexer::Lexer::new("a|b|c");
-        let mut parser = crate::parser::Parser::new(&mut lexer);
-        let nfa = crate::automaton::nfa::Nfa::new_from_node(
-            parser.parse().unwrap(),
-            &mut crate::automaton::nfa::NfaState::new(),
-        )
-        .unwrap();
-
-        let closure = nfa.epsilon_closure([nfa.start()].iter().cloned().collect());
-        assert_eq!(closure, [0, 2, 4, 6, 7].iter().cloned().collect());
-    }
-
     #[test]
     fn test_dfa_from_nfa() {
         let nfa = crate::automaton::nfa::Nfa::new_from_node(

diff --git a/src/automaton/nfa.rs b/src/automaton/nfa.rs
@@ -491,4 +491,25 @@ mod tests {
             .collect()
         );
     }
+
+    #[test]
+    fn e_closure() {
+        let mut lexer = crate::lexer::Lexer::new("a|b*");
+        let mut parser = crate::parser::Parser::new(&mut lexer);
+        let nfa = Nfa::new_from_node(
+            parser.parse().unwrap(),
+            &mut crate::automaton::nfa::NfaState::new(),
+        )
+        .unwrap();
+
+        let closure = nfa.epsilon_closure([nfa.start()].iter().cloned().collect());
+        assert_eq!(closure, [0, 2, 4, 5].iter().cloned().collect());
+
+        let mut lexer = crate::lexer::Lexer::new("a|b|c");
+        let mut parser = crate::parser::Parser::new(&mut lexer);
+        let nfa = Nfa::new_from_node(parser.parse().unwrap(), &mut NfaState::new()).unwrap();
+
+        let closure = nfa.epsilon_closure([nfa.start()].iter().cloned().collect());
+        assert_eq!(closure, [0, 2, 4, 6, 7].iter().cloned().collect());
+    }
 }
diff --git a/src/lib.rs b/src/lib.rs
@@ -139,7 +139,7 @@ mod tests {
     }
 
     #[test]
-    fn multibyte_dfa() {
+    fn nonascii_dfa() {
         let regex = RustRegex::new("あ|い*", "dfa").unwrap();
         assert!(regex.is_match("あ"));
         assert!(regex.is_match("い"));
@@ -193,11 +193,20 @@ mod tests {
         assert!(regex.is_match("♥"));
         assert!(!regex.is_match("♡"));
         assert!(!regex.is_match("👎️"));
+
+        let regex = RustRegex::new("ගවයා|ng'ombe", "dfa").unwrap();
+        assert!(regex.is_match("ගවයා"));
+        assert!(regex.is_match("ng'ombe"));
+        assert!(!regex.is_match("ගවයාng'ombe"));
+
+        let regex = RustRegex::new("(පරිගණකය)*", "dfa").unwrap();
+        assert!(regex.is_match("පරිගණකය"));
+        assert!(regex.is_match(""));
     }
 
     #[test]
     fn invalid_dfa() {
-        for test in ["a(b", "*", ")c", "*", "+"] {
+        for test in ["a(b", "*", ")c", "+"] {
             let regex = RustRegex::new(test, "dfa");
             assert!(regex.is_err());
         }
@@ -273,7 +282,7 @@ mod tests {
     }
 
     #[test]
-    fn multibyte_vm() {
+    fn nonascii_vm() {
         let regex = RustRegex::new("あ|い*", "vm").unwrap();
         assert!(regex.is_match("あ"));
         assert!(regex.is_match("い"));
@@ -327,11 +336,20 @@ mod tests {
         assert!(regex.is_match("♥"));
         assert!(!regex.is_match("♡"));
         assert!(!regex.is_match("👎️"));
+
+        let regex = RustRegex::new("ගවයා|ng'ombe", "vm").unwrap();
+        assert!(regex.is_match("ගවයා"));
+        assert!(regex.is_match("ng'ombe"));
+        assert!(!regex.is_match("ගවයාng'ombe"));
+
+        let regex = RustRegex::new("(පරිගණකය)*", "vm").unwrap();
+        assert!(regex.is_match("පරිගණකය"));
+        assert!(regex.is_match(""));
     }
 
     #[test]
     fn invalid_vm() {
-        for test in ["a(b", "*", ")c", "*", "+"] {
+        for test in ["a(b", "*", ")c", "+"] {
             let regex = RustRegex::new(test, "vm");
             assert!(regex.is_err());
         }
@@ -407,7 +425,7 @@ mod tests {
     }
 
     #[test]
-    fn multibyte_derivative() {
+    fn nonascii_derivative() {
         let regex = RustRegex::new("あ|い*", "derivative").unwrap();
         assert!(regex.is_match("あ"));
         assert!(regex.is_match("い"));
@@ -461,11 +479,20 @@ mod tests {
         assert!(regex.is_match("♥"));
         assert!(!regex.is_match("♡"));
         assert!(!regex.is_match("👎️"));
+
+        let regex = RustRegex::new("ගවයා|ng'ombe", "derivative").unwrap();
+        assert!(regex.is_match("ගවයා"));
+        assert!(regex.is_match("ng'ombe"));
+        assert!(!regex.is_match("ගවයාng'ombe"));
+
+        let regex = RustRegex::new("(පරිගණකය)*", "derivative").unwrap();
+        assert!(regex.is_match("පරිගණකය"));
+        assert!(regex.is_match(""));
     }
 
     #[test]
     fn invalid_derivative() {
-        for test in ["a(b", "*", ")c", "*", "+"] {
+        for test in ["a(b", "*", ")c", "+"] {
             let regex = RustRegex::new(test, "derivative");
             assert!(regex.is_err());
         }