Add support for nested loops

[Copilot]

• Fix validate_syntax to use a depth counter instead of a boolean flag so nested brackets are validated correctly
• Track the most recently seen [ in error messages (reports innermost unclosed bracket for deep nesting)
• Fix run to precompute matching bracket pairs and use them for JUMP_FORWARD / JUMP_BACKWARD jumps
• Add comment explaining the stack invariant in run
• Create examples/hello-world-nested-loops.bf with the content from the issue
• Add the new example file to the parametrized test in test_example_files.py
• Fix KeyError crash: use memory.get(position, 0) in output_current_byte and loop-condition checks in handle_token so uninitialized sparse-dict cells are treated as 0
• Replace single-letter variable names (i, j) with descriptive names (token_index, open_bracket_index)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.04%. Comparing base (483dd3a) to head (552a4d3).
⚠️ Report is 1 commits behind head on main.
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #71      +/-   ##
==========================================
+ Coverage   98.54%   99.04%   +0.49%     
==========================================
  Files          10       10              
  Lines         206      209       +3     
  Branches       31       31              
==========================================
+ Hits          203      207       +4     
+ Partials        3        2       -1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[shenanigansd]

@copilot fix this new test failure:

===================================================================== FAILURES ======================================================================
___________________________________________________ test_example_files[example_file1-hello world] ___________________________________________________

example_file = PosixPath('examples/hello-world-nested-loops.bf'), expected_result = 'hello world'

    @pytest.mark.parametrize(
        ("example_file", "expected_result"),
        [
            (Path("./examples/hello-world.bf"), "Hello, World!"),
            (Path("./examples/hello-world-nested-loops.bf"), "hello world"),
            (Path("./examples/jump.bf"), ""),
            (Path("./examples/decrement_not_in_memory.bf"), ""),
        ],
    )
    def test_example_files(example_file: Path, expected_result: str) -> None:
        """Test example files."""
        source = example_file.read_text()
        interpreter = Interpreter()
>       assert interpreter.run(source) == expected_result
               ^^^^^^^^^^^^^^^^^^^^^^^

tests/test_example_files.py:24: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
src/bpp/interpreter.py:162: in run
    match self.handle_token(token):
          ^^^^^^^^^^^^^^^^^^^^^^^^
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

self = <bpp.interpreter.Interpreter object at 0x7c2e29cff5b0>, token = <Token.LOOP_START: '['>

    def handle_token(self: Self, token: Token) -> ResultState:  # noqa: C901
        """Handle a single token.
    
        Returns
        -------
        The state of the interpreter after handling the token
        """
        match token:
            case Token.INCREMENT_POINTER:
                self.increment_pointer()
            case Token.DECREMENT_POINTER:
                self.decrement_pointer()
            case Token.INCREMENT_BYTE:
                self.increment_byte_at_current_pointer()
            case Token.DECREMENT_BYTE:
                self.decrement_byte_at_current_pointer()
            case Token.OUTPUT_BYTE:
                self.output_current_byte()
            case Token.INPUT_BYTE:
                self.get_input()
            case Token.LOOP_START:
>               if self.memory[self.current_position] == 0:
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
E               KeyError: 1

src/bpp/interpreter.py:129: KeyError
============================================================== short test summary info ==============================================================
FAILED tests/test_example_files.py::test_example_files[example_file1-hello world] - KeyError: 1
=========================================================== 1 failed, 17 passed in 0.21s ============================================================
@shenanigansd ➜ /workspaces/bpp (copilot/add-support-for-nested-loops) $ 

[Copilot]

@copilot fix this new test failure:

===================================================================== FAILURES ======================================================================

...

Fixed in cb13182. The failure was a KeyError: the interpreter stores memory as a sparse dict, but output_current_byte and the loop-condition checks in handle_token read directly from the dict (self.memory[pos]) without a default. The nested-loops program accesses cells that haven't been written yet. Changed all three read sites to use self.memory.get(self.current_position, 0) to treat uninitialized cells as 0, consistent with how the write methods already behave.

[shenanigansd]

@copilot replace all single letter variable names with full words.

[Copilot]

@copilot replace all single letter variable names with full words.

Done in 552a4d3. Renamed i → token_index and j → open_bracket_index in the bracket-map loop inside run().