Fix profiling scripts with a UTF-8 BOM (#1092)

Author: mameikagou

scalene/scalene_profiler.py

@@ -2027,7 +2027,7 @@ def run_profiler(
                     raise FileNotFoundError
                 # Use the full absolute path of the program being profiled, expanding ~ if need be.
                 prog_name = os.path.abspath(os.path.expanduser(progs[0]))
-                with open(prog_name, encoding="utf-8") as prog_being_profiled:
+                with open(prog_name, encoding="utf-8-sig") as prog_being_profiled:
                     # Read in the code and compile it.
                     code: Any = ""
                     try:

tests/test_issue1091_bom.py

@@ -0,0 +1,50 @@
+"""Regression test for issue #1091: UTF-8 BOM in a profiled script."""
+
+from __future__ import annotations
+
+import subprocess
+import sys
+from pathlib import Path
+
+import pytest
+
+
+@pytest.mark.parametrize(
+    ("name", "source", "expected"),
+    [
+        ("plain", b'print("PLAIN_OK")\n', "PLAIN_OK"),
+        ("bom", b'\xef\xbb\xbfprint("BOM_OK")\n', "BOM_OK"),
+        ("utf8_non_ascii", 'print("é_OK")\n'.encode(), "é_OK"),
+    ],
+)
+def test_cpu_only_cli_accepts_utf8_script_variants(
+    tmp_path: Path, name: str, source: bytes, expected: str
+) -> None:
+    """The real CLI must execute BOM and ordinary UTF-8 scripts successfully."""
+    script = tmp_path / f"{name}.py"
+    script.write_bytes(source)
+
+    result = subprocess.run(
+        [
+            sys.executable,
+            "-m",
+            "scalene",
+            "run",
+            "--cpu-only",
+            "--no-browser",
+            str(script),
+        ],
+        cwd=tmp_path,
+        check=False,
+        capture_output=True,
+        text=True,
+        timeout=60,
+    )
+    output = result.stdout + result.stderr
+
+    assert result.returncode == 0, (
+        f"Scalene failed for {name}: exit={result.returncode}\n{output[-3000:]}"
+    )
+    assert expected in result.stdout, (
+        f"Profiled {name} script did not run to completion:\n{output[-3000:]}"
+    )