fix a lexing bug and make utf8 use subscripts

uiua-lang · Sep 14, 2024 · 9b0390e · 9b0390e
1 parent 24154ac
commit 9b0390e
Show file tree

Hide file tree

Showing 4 changed files with 20 additions and 7 deletions.
diff --git a/site/primitives.json b/site/primitives.json
@@ -391,7 +391,6 @@
   },
   "both": {
     "glyph": "∩",
-    "args": 2,
     "outputs": 1,
     "modifier_args": 1,
     "class": "Planet",

diff --git a/src/compile/mod.rs b/src/compile/mod.rs
@@ -2049,10 +2049,12 @@ code:
     }
     #[allow(clippy::match_single_binding)]
     fn subscript(&mut self, sub: Subscript, span: CodeSpan, call: bool) -> UiuaResult {
-        self.experimental_error(&span, || {
-            "Subscripts are experimental. To use them, add \
-            `# Experimental!` to the top of the file."
-        });
+        if !matches!(sub.word.value, Word::Primitive(Primitive::Utf8)) {
+            self.experimental_error(&span, || {
+                "Subscripts are experimental. To use them, add \
+                `# Experimental!` to the top of the file."
+            });
+        }
         let n = sub.n.value;
         match sub.word.value {
             Word::Modified(m) => match m.modifier.value {
@@ -2106,6 +2108,12 @@ code:
                     self.primitive(Primitive::Mul, span.clone(), call)?;
                     self.primitive(Primitive::Floor, span, call)?;
                 }
+                Primitive::Utf8 => {
+                    if n != 8 {
+                        self.add_error(span.clone(), "Only UTF-8 is supported");
+                    }
+                    self.primitive(Primitive::Utf8, span, call)?
+                }
                 _ => {
                     self.add_error(
                         span.clone(),

diff --git a/src/format.rs b/src/format.rs
@@ -1060,7 +1060,7 @@ impl<'a> Formatter<'a> {
                 }
                 self.output.push(')');
             }
-            Word::Primitive(prim) => self.push(&word.span, &prim.to_string()),
+            Word::Primitive(prim) => self.format_primitive(*prim, &word.span),
             Word::Modified(m) => {
                 self.format_modifier(&m.modifier);
                 self.format_words(&m.operands, true, depth);
@@ -1216,6 +1216,12 @@ impl<'a> Formatter<'a> {
             }
         }
     }
+    fn format_primitive(&mut self, prim: Primitive, span: &CodeSpan) {
+        match prim {
+            Primitive::Utf8 => self.push(span, "utf"),
+            _ => self.push(span, &prim.to_string()),
+        }
+    }
     fn format_multiline_words(
         &mut self,
         mut lines: &[Vec<Sp<Word>>],

diff --git a/src/lex.rs b/src/lex.rs
@@ -1119,7 +1119,7 @@ impl<'a> Lexer<'a> {
                     // Try to parse as primitives
                     let lowercase_end = ident
                         .char_indices()
-                        .find(|(_, c)| !c.is_ascii_lowercase())
+                        .find(|(_, c)| !c.is_ascii_lowercase() && *c != '&')
                         .map_or(ident.len(), |(i, _)| i);
                     let lowercase = &ident[..lowercase_end];
                     if let Some(prims) = Primitive::from_format_name_multi(lowercase) {