From 6255eb95737919a92e24459df1fc2ac4be168be8 Mon Sep 17 00:00:00 2001
From: Dmitry <shirshov.dima@gmail.com>
Date: Sat, 1 Feb 2025 20:40:43 +0300
Subject: [PATCH] feat(comment): support single-line simple prefix comments
 (#23)

---
 src/lexer.rs                     | 18 ++++++++++++++++++
 src/parser.rs                    |  3 +++
 tests/input/strip_comment.jsonc  |  5 +++++
 tests/output/strip_comment.jsonc |  5 +++++
 4 files changed, 31 insertions(+)
 create mode 100644 tests/input/strip_comment.jsonc
 create mode 100644 tests/output/strip_comment.jsonc

diff --git a/src/lexer.rs b/src/lexer.rs
index 81fa196..a19ae16 100644
--- a/src/lexer.rs
+++ b/src/lexer.rs
@@ -5,6 +5,11 @@ pub enum Paired {
     File,    // start, end
 }
 
+#[derive(Debug, Clone, Copy, Eq, PartialEq)]
+pub enum Comment {
+    Oneline, // // or # or ;
+}
+
 #[derive(Debug, Clone, Copy, Eq, PartialEq)]
 pub enum Lexem {
     String,
@@ -14,6 +19,7 @@ pub enum Lexem {
     Colon,
     Else,
     WhiteSpace,
+    Comment(Comment),
 }
 
 fn joinable(lexem: Lexem) -> bool {
@@ -39,6 +45,7 @@ impl From<u8> for Lexem {
             b',' => Lexem::Comma,
             b':' | b'=' => Lexem::Colon,
             b'\'' | b'"' | b'`' => Lexem::String,
+            b';' | b'#' | b'/' => Lexem::Comment(Comment::Oneline),
             _ => {
                 if character.is_ascii_whitespace() {
                     Lexem::WhiteSpace
@@ -158,6 +165,17 @@ impl Lexer {
         if cfg!(debug_assertions) {
             eprintln!("{character} {}", character as char);
         }
+        if let Some(Lexem::Comment(Comment::Oneline)) = self.lexem {
+            if character == b'\r' || character == b'\n' {
+                self.lexem = Some(Lexem::WhiteSpace);
+                return Some((
+                    Lexem::Comment(Comment::Oneline),
+                    std::mem::replace(&mut self.state, vec![character]),
+                ));
+            }
+            self.state.push(character);
+            return None;
+        }
         if let Some(Lexem::String) = self.lexem {
             let first_char = self.state.first().cloned().unwrap_or_default();
             let last_char = self.state.last().cloned().unwrap_or_default();
diff --git a/src/parser.rs b/src/parser.rs
index d586199..5f2c730 100644
--- a/src/parser.rs
+++ b/src/parser.rs
@@ -143,6 +143,9 @@ impl Parser {
             eprintln!("LEXEM {lexem:?} {value:?}");
         }
         let mut result = Vec::new();
+        if let Lexem::Comment(_) = lexem {
+            return result;
+        }
         if let Lexem::WhiteSpace = lexem {
             self.whitespace.extend(value);
             return result;
diff --git a/tests/input/strip_comment.jsonc b/tests/input/strip_comment.jsonc
new file mode 100644
index 0000000..72347c3
--- /dev/null
+++ b/tests/input/strip_comment.jsonc
@@ -0,0 +1,5 @@
+{
+    // c-style oneline comment
+    # python-style oneline comment
+    ; assembler-style comment
+}
diff --git a/tests/output/strip_comment.jsonc b/tests/output/strip_comment.jsonc
new file mode 100644
index 0000000..abd2979
--- /dev/null
+++ b/tests/output/strip_comment.jsonc
@@ -0,0 +1,5 @@
+{
+    
+    
+    
+}