Lexer: comments groundwork

4 files changed, 39 insertions, 6 deletions

diff --git a/grammar.ron b/grammar.ron
index 482e00f91..71e354dac 100644
--- a/grammar.ron
+++ b/grammar.ron
@@ -48,5 +48,8 @@ Grammar(
         "AMPERSAND",
         "PIPE",
         "THIN_ARROW",
+        "COMMENT",
+        "DOC_COMMENT",
+        "SHEBANG",
     ]
 )
\ No newline at end of file
diff --git a/src/lexer/comments.rs b/src/lexer/comments.rs
new file mode 100644
index 000000000..c61c85824
--- /dev/null
+++ b/src/lexer/comments.rs
@@ -0,0 +1,11 @@
+use lexer::ptr::Ptr;
+
+use {SyntaxKind};
+
+pub(crate) fn scan_shebang(ptr: &mut Ptr) -> bool {
+    false
+}
+
+pub(crate) fn scan_comment(ptr: &mut Ptr) -> Option<SyntaxKind> {
+    None
+}
\ No newline at end of file
diff --git a/src/lexer/mod.rs b/src/lexer/mod.rs
index 3e49b1c2b..f46746bee 100644
--- a/src/lexer/mod.rs
+++ b/src/lexer/mod.rs
@@ -13,6 +13,9 @@ use self::numbers::scan_number;
 mod strings;
 use self::strings::{is_string_literal_start, scan_char, scan_byte_char_or_string, scan_string, scan_raw_string};
 
+mod comments;
+use self::comments::{scan_shebang, scan_comment};
+
 pub fn next_token(text: &str) -> Token {
     assert!(!text.is_empty());
     let mut ptr = Ptr::new(text);
@@ -23,16 +26,26 @@ pub fn next_token(text: &str) -> Token {
 }
 
 fn next_token_inner(c: char, ptr: &mut Ptr) -> SyntaxKind {
-    let ident_start = is_ident_start(c) && !is_string_literal_start(c, ptr.next(), ptr.nnext());
-    if ident_start {
-        return scan_ident(c, ptr);
-    }
-
     if is_whitespace(c) {
         ptr.bump_while(is_whitespace);
         return WHITESPACE;
     }
 
+    match c {
+        '#' => if scan_shebang(ptr) {
+            return SHEBANG;
+        }
+        '/' => if let Some(kind) = scan_comment(ptr) {
+            return kind;
+        }
+        _ => (),
+    }
+
+    let ident_start = is_ident_start(c) && !is_string_literal_start(c, ptr.next(), ptr.nnext());
+    if ident_start {
+        return scan_ident(c, ptr);
+    }
+
     if is_dec_digit(c) {
         let kind = scan_number(c, ptr);
         scan_literal_suffix(ptr);
diff --git a/src/syntax_kinds.rs b/src/syntax_kinds.rs
index 83fabe403..ec2a036b9 100644
--- a/src/syntax_kinds.rs
+++ b/src/syntax_kinds.rs
@@ -49,8 +49,11 @@ pub const PERCENT: SyntaxKind = SyntaxKind(44);
 pub const AMPERSAND: SyntaxKind = SyntaxKind(45);
 pub const PIPE: SyntaxKind = SyntaxKind(46);
 pub const THIN_ARROW: SyntaxKind = SyntaxKind(47);
+pub const COMMENT: SyntaxKind = SyntaxKind(48);
+pub const DOC_COMMENT: SyntaxKind = SyntaxKind(49);
+pub const SHEBANG: SyntaxKind = SyntaxKind(50);
 
-static INFOS: [SyntaxInfo; 48] = [
+static INFOS: [SyntaxInfo; 51] = [
     SyntaxInfo { name: "ERROR" },
     SyntaxInfo { name: "IDENT" },
     SyntaxInfo { name: "UNDERSCORE" },
@@ -99,6 +102,9 @@ static INFOS: [SyntaxInfo; 48] = [
     SyntaxInfo { name: "AMPERSAND" },
     SyntaxInfo { name: "PIPE" },
     SyntaxInfo { name: "THIN_ARROW" },
+    SyntaxInfo { name: "COMMENT" },
+    SyntaxInfo { name: "DOC_COMMENT" },
+    SyntaxInfo { name: "SHEBANG" },
 ];
 
 pub(crate) fn syntax_info(kind: SyntaxKind) -> &'static SyntaxInfo {