Support \u escapes in strings

schala-lang/src/parsing/combinator.rs

@@ -841,8 +841,15 @@ fn string_literal(input: Span) -> ParseResult<ExpressionKind> {
 }
 
 fn bare_string_literal(input: Span) -> ParseResult<String> {
-    let string_escape_transforms =
+    let string_escape_transforms = alt((
-        alt((value("\\", tag("\\")), value("\"", tag("\"")), value("\n", tag("n")), value("\t", tag("t"))));
+        value('\\', tag("\\")),
+        value('"', tag("\"")),
+        value('\n', tag("n")),
+        value('\t', tag("t")),
+        map(delimited(tag(r#"u{"#), recognize(digit_group_hex), tag("}")), |value| {
+            char::from_u32(u32::from_str_radix(value.fragment(), 16).unwrap()).unwrap()
+        }),
+    ));
     alt((
         map(tag(r#""""#), |_| String::new()),
         preceded(

schala-lang/src/parsing/peg_parser.rs

@@ -423,7 +423,7 @@ peg::parser! {
         rule pattern_literal() -> Pattern =
             "true" { Pattern::Literal(PatternLiteral::BoolPattern(true)) } /
             "false" { Pattern::Literal(PatternLiteral::BoolPattern(false)) } /
-            s:bare_string_literal() { Pattern::Literal(PatternLiteral::StringPattern(Rc::new(s.to_string()))) } /
+            s:bare_string_literal() { Pattern::Literal(PatternLiteral::StringPattern(Rc::new(s))) } /
             sign:("-"?)  num:(float_literal() / nat_literal()) {
                 let neg = sign.is_some();
                 Pattern::Literal(PatternLiteral::NumPattern { neg, num })
@@ -447,19 +447,18 @@ peg::parser! {
             }
 
         rule string_literal() -> ExpressionKind =
-            prefix:identifier()? s:bare_string_literal(){ ExpressionKind::StringLiteral{ s: Rc::new(s.to_string()),
+            prefix:identifier()? s:bare_string_literal(){ ExpressionKind::StringLiteral{ s: Rc::new(s),
                 prefix: prefix.map(rc_string)
             } }
 
-        rule bare_string_literal() -> &'input str =
+        rule bare_string_literal() -> String =
-            "\"" s:$(string_component()*) "\"" { s }
+            "\"" chars:string_component()* "\"" { chars.into_iter().collect::<String>() }
 
-        rule string_component() -> &'input str =
+        rule string_component() -> char =
-            r#"\\"# { "\\" } /
+            !(r#"""# / r#"\"#) ch:$([_]) { ch.chars().next().unwrap() } /
-            r#"\""# { "\"" } /
+            r#"\u{"# value:$(['0'..='9' | 'a'..='f' | 'A'..='F']+) "}" { char::from_u32(u32::from_str_radix(value, 16).unwrap()).unwrap() } /
-            r#"\t"# { "\t" } /
+            r#"\n"# { '\n' } / r#"\t"# { '\t' } / r#"\""# { '"' } / r#"\\"# { '\\' } /
-            r#"\n"# { "\n" } /
+      expected!("Valid escape sequence")
-            ch:$([^ '"' ]) { ch }
 
         rule bool_literal() -> ExpressionKind =
             "true" { ExpressionKind::BoolLiteral(true) } / "false" { ExpressionKind::BoolLiteral(false) }

schala-lang/src/parsing/test.rs

@@ -189,8 +189,8 @@ fn string_literals() {
         r#"b"some bytestring""#,
         expr(StringLiteral { s: rc("some bytestring"), prefix: Some(rc("b")) })
     );
-    //NOTE I'm not 100% sure this case is correct, but I'll deal with it later
+    assert_expr!(r#""Do \n \" escapes work\t""#, expr(strlit("Do \n \" escapes work\t")));
-    //assert_expr!(r#""Do \n \" escapes work\t""#, expr(StringLiteral(rc("Do \n \" escapes work\t"))));
+    assert_expr!(r#""Georgian letter jani \u{10ef}""#, expr(strlit("Georgian letter jani ჯ")));
 }
 
 #[test]