From 3f9551596457a56d1727705967aeb027165ba9d0 Mon Sep 17 00:00:00 2001
From: Araozu <fernando@araozu.dev>
Date: Sun, 24 Nov 2024 07:32:37 -0500
Subject: [PATCH] feat: lex comments, forbid CR

---
 build.zig                |  1 +
 src/01_lexic/comment.zig | 77 ++++++++++++++++++++++++++++++++++++++++
 src/01_lexic/root.zig    |  9 +++++
 src/01_lexic/token.zig   |  2 ++
 4 files changed, 89 insertions(+)
 create mode 100644 src/01_lexic/comment.zig

diff --git a/build.zig b/build.zig
index cbc5822..7678f5d 100644
--- a/build.zig
+++ b/build.zig
@@ -96,6 +96,7 @@ pub fn build(b: *std.Build) void {
         "src/01_lexic/identifier.zig",
         "src/01_lexic/datatype.zig",
         "src/01_lexic/operator.zig",
+        "src/01_lexic/comment.zig",
         "src/01_lexic/token.zig",
         "src/01_lexic/utils.zig",
     };
diff --git a/src/01_lexic/comment.zig b/src/01_lexic/comment.zig
new file mode 100644
index 0000000..d0c3fcb
--- /dev/null
+++ b/src/01_lexic/comment.zig
@@ -0,0 +1,77 @@
+const std = @import("std");
+const assert = std.debug.assert;
+const token = @import("./token.zig");
+const utils = @import("./utils.zig");
+
+const Token = token.Token;
+const TokenType = token.TokenType;
+const LexError = token.LexError;
+const LexReturn = token.LexReturn;
+
+pub fn lex(input: []const u8, start: usize) LexError!?LexReturn {
+    const cap = input.len;
+    assert(start < cap);
+
+    // there should be at least 2 characters
+    if (start + 1 >= cap) {
+        return null;
+    }
+
+    if (input[start] == '/' and input[start + 1] == '/') {
+        var current_pos = start + 2;
+
+        // consume all bytes until newline (LF)
+        while (current_pos < cap and input[current_pos] != '\n') {
+            // check for CR, and throw error
+            if (input[current_pos] == '\r') {
+                return LexError.CRLF;
+            }
+            current_pos += 1;
+        }
+
+        return .{ Token.init(input[start..current_pos], TokenType.Comment, start), current_pos };
+    } else {
+        return null;
+    }
+}
+
+test "should lex comment until EOF" {
+    const input = "// aea";
+    const output = try lex(input, 0);
+
+    if (output) |tuple| {
+        const t = tuple[0];
+        try std.testing.expectEqualDeep("// aea", t.value);
+        try std.testing.expectEqual(6, tuple[1]);
+    } else {
+        try std.testing.expect(false);
+    }
+}
+
+test "should lex comment until newline (LF)" {
+    const input = "// my comment\n// other comment";
+    const output = try lex(input, 0);
+
+    if (output) |tuple| {
+        const t = tuple[0];
+        try std.testing.expectEqualDeep("// my comment", t.value);
+        try std.testing.expectEqual(13, tuple[1]);
+    } else {
+        try std.testing.expect(false);
+    }
+}
+
+test "shouldn lex incomplete comment" {
+    const input = "/aa";
+    const output = try lex(input, 0);
+    try std.testing.expect(output == null);
+}
+
+test "should fail on CRLF" {
+    const input = "// my comment\x0D\x0A// other comment";
+    _ = lex(input, 0) catch |err| {
+        try std.testing.expectEqual(LexError.CRLF, err);
+        return;
+    };
+    try std.testing.expect(false);
+}
diff --git a/src/01_lexic/root.zig b/src/01_lexic/root.zig
index a3f3fe1..cee5514 100644
--- a/src/01_lexic/root.zig
+++ b/src/01_lexic/root.zig
@@ -5,6 +5,7 @@ const identifier = @import("./identifier.zig");
 const datatype = @import("./datatype.zig");
 const token = @import("./token.zig");
 const operator = @import("./operator.zig");
+const comment = @import("./comment.zig");
 
 const TokenType = token.TokenType;
 const Token = token.Token;
@@ -44,6 +45,14 @@ pub fn tokenize(input: []const u8, alloc: std.mem.Allocator) !void {
 
             try tokens.append(t);
         }
+        // attempt to lex a comment
+        else if (try comment.lex(input, actual_next_pos)) |tuple| {
+            assert(tuple[1] > current_pos);
+            const t = tuple[0];
+            current_pos = tuple[1];
+
+            try tokens.append(t);
+        }
         // attempt to lex an operator
         else if (try operator.lex(input, actual_next_pos)) |tuple| {
             assert(tuple[1] > current_pos);
diff --git a/src/01_lexic/token.zig b/src/01_lexic/token.zig
index 18a8284..17b176d 100644
--- a/src/01_lexic/token.zig
+++ b/src/01_lexic/token.zig
@@ -3,6 +3,7 @@ pub const TokenType = enum {
     Float,
     Identifier,
     Operator,
+    Comment,
 };
 
 pub const Token = struct {
@@ -24,6 +25,7 @@ pub const LexError = error{
     Incomplete,
     IncompleteFloatingNumber,
     IncompleteScientificNumber,
+    CRLF,
 };
 
 /// Contains the lexed token and the next position