From 3115c2dfeb5477e2c0f7178307ca230a347c7b22 Mon Sep 17 00:00:00 2001
From: "copilot-swe-agent[bot]" <198982749+Copilot@users.noreply.github.com>
Date: Sun, 12 Apr 2026 17:54:43 +0000
Subject: [PATCH 1/2] Initial plan


From 0654aae79cfed7fb669778a0f37b8b2550a62f14 Mon Sep 17 00:00:00 2001
From: "copilot-swe-agent[bot]" <198982749+Copilot@users.noreply.github.com>
Date: Sun, 12 Apr 2026 17:58:34 +0000
Subject: [PATCH 2/2] Fix syntax error position reporting with comments in
 source code

Agent-Logs-Url: https://github.com/letsbuilda/bpp/sessions/b28bd439-112c-4c0c-851d-721db886836d

Co-authored-by: shenanigansd <54628770+shenanigansd@users.noreply.github.com>
---
 src/bpp/interpreter.py    | 14 +++++++-------
 src/bpp/tokens.py         | 17 +++++++++++++++++
 tests/test_interpreter.py |  3 +++
 3 files changed, 27 insertions(+), 7 deletions(-)

diff --git a/src/bpp/interpreter.py b/src/bpp/interpreter.py
index a5340ff..e496051 100644
--- a/src/bpp/interpreter.py
+++ b/src/bpp/interpreter.py
@@ -6,7 +6,7 @@
 from typing import TYPE_CHECKING, Self
 
 from .exceptions import BrainfuckSyntaxError
-from .tokens import Token, tokenize
+from .tokens import Token, _tokenize_with_positions
 
 if TYPE_CHECKING:
     from collections.abc import Sequence
@@ -20,7 +20,7 @@ class ResultState(Enum):
     JUMP_BACKWARD = auto()
 
 
-def validate_syntax(syntax: Sequence[Sequence[Token]]) -> None:
+def validate_syntax(syntax: Sequence[Sequence[tuple[Token, int]]]) -> None:
     """Validate the given syntax.
 
     Raises
@@ -32,16 +32,16 @@ def validate_syntax(syntax: Sequence[Sequence[Token]]) -> None:
     loop_started_at_line = 0
     loop_started_at_character = 0
     for line_index, line in enumerate(syntax):
-        for character_index, token in enumerate(line):
+        for token, char_position in line:
             if token == Token.LOOP_START:
                 loop_started_at_line = line_index
-                loop_started_at_character = character_index
+                loop_started_at_character = char_position
                 depth += 1
             if token == Token.LOOP_END:
                 if depth == 0:
                     msg = (
                         "Syntax error: Unexpected closing bracket in line "
-                        f"{line_index + 1} at char {character_index + 1}!"
+                        f"{line_index + 1} at char {char_position + 1}!"
                     )
                     raise BrainfuckSyntaxError(msg)
                 depth -= 1
@@ -140,9 +140,9 @@ def run(self: Self, code: str) -> str:
         -------
         The output of the code.
         """
-        syntax = [tokenize(line) for line in code.split("\n")]
+        syntax = [_tokenize_with_positions(line) for line in code.split("\n")]
         validate_syntax(syntax)
-        tokens = [token for line in syntax for token in line]
+        tokens = [token for line in syntax for token, _ in line]
 
         # Precompute matching bracket pairs to support nested loops.
         # validate_syntax guarantees brackets are balanced, so the stack is
diff --git a/src/bpp/tokens.py b/src/bpp/tokens.py
index e49bcef..c3f8703 100644
--- a/src/bpp/tokens.py
+++ b/src/bpp/tokens.py
@@ -47,3 +47,20 @@ def tokenize(code: str) -> Sequence[Token]:
             continue
         tokens.append(token)
     return tokens
+
+
+def _tokenize_with_positions(code: str) -> Sequence[tuple[Token, int]]:
+    """Convert text to (token, original_char_index) tuples.
+
+    Returns
+    -------
+    A sequence of (token, original_char_index) tuples preserving the
+    character's position in the original source string.
+    """
+    tokens = []
+    for position, character in enumerate(code):
+        token = Token.from_character(character)
+        if token is None:
+            continue
+        tokens.append((token, position))
+    return tokens
diff --git a/tests/test_interpreter.py b/tests/test_interpreter.py
index 6f1214d..2ed7901 100644
--- a/tests/test_interpreter.py
+++ b/tests/test_interpreter.py
@@ -51,6 +51,9 @@ def test_can_input_letter(monkeypatch) -> None:  # type: ignore[no-untyped-def]
     [
         ("[", "Syntax error: Unclosed bracket in line 1 at char 1!"),
         ("]", "Syntax error: Unexpected closing bracket in line 1 at char 1!"),
+        # Comments before brackets should not affect reported position
+        (">comment[", r"Syntax error: Unclosed bracket in line 1 at char 9!"),
+        (">comment]", r"Syntax error: Unexpected closing bracket in line 1 at char 9!"),
     ],
 )
 def test_invalid_syntax_fails(code: str, error_message: str) -> None: