From 69a34583ba4bab52998df8c21e3d875e82d2d76c Mon Sep 17 00:00:00 2001
From: MgnMtn <mematon1@sheffield.ac.uk>
Date: Sat, 1 Aug 2026 17:31:55 +0100
Subject: [PATCH 1/2] feat: pytest output

---
 multiplex/execute/pytest_runner.py | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/multiplex/execute/pytest_runner.py b/multiplex/execute/pytest_runner.py
index e4dfc41..1736bbb 100644
--- a/multiplex/execute/pytest_runner.py
+++ b/multiplex/execute/pytest_runner.py
@@ -21,7 +21,7 @@
 from util.rewrite_method import rewrite_method
 
 
-def _execute(project_root):
+def _execute(project_root, output_path=None, approach=None, label=None):
     """Run the project's tests with pytest.
 
     Returns True if pytest exits 0 (all tests pass) — i.e. the mutant is not
@@ -31,7 +31,8 @@ def _execute(project_root):
     command = [sys.executable, "-m", "pytest", "-q", str(project_root)]
     try:
         result = subprocess.run(
-            command, stdout=subprocess.PIPE, stderr=subprocess.STDOUT, check=False
+            command, stdout=subprocess.PIPE, stderr=subprocess.STDOUT,
+            text=True, check=False
         )
     except FileNotFoundError as exc:
         raise SystemExit(
@@ -39,6 +40,15 @@ def _execute(project_root):
             f"({sys.executable}). Install pytest to use the 'pytest' runtool."
         ) from exc
 
+    if output_path is not None and approach and label:
+        test_dir = Path(output_path, approach + "-test")
+        test_dir.mkdir(parents=True, exist_ok=True)
+        safe = str(label).replace(os.sep, "_")
+        (test_dir / f"{safe}_test.txt").write_text(
+            f"$ {' '.join(command)}\n# exit code: {result.returncode}\n\n"
+            f"{result.stdout}"
+        )
+
     return result.returncode == 0
 
 
@@ -59,7 +69,7 @@ def run_mutants(
 
     mutants = [["MUTANT", "EQUIVALENCE", "COMPILABLE", "SURVIVES"]]
 
-    if not _execute(project_root):
+    if not _execute(project_root, output_path, approach, "ORIGINAL"):
         raise IOError(
             "The original (unmutated) project did not pass pytest, so mutants "
             "cannot be evaluated against it. Run it manually to see why: "
@@ -83,7 +93,7 @@ def run_mutants(
 
         mutant_survives = False
         if mutant_compiles:
-            mutant_survives = _execute(project_root)
+            mutant_survives = _execute(project_root, output_path, approach, mutant_file)
         mutant_output.append(str(mutant_survives))
 
         mutants.append(mutant_output)

From 7210d7f1635204892a09a2fc40eac2bcf927e6aa Mon Sep 17 00:00:00 2001
From: MgnMtn <mematon1@sheffield.ac.uk>
Date: Sat, 1 Aug 2026 17:45:53 +0100
Subject: [PATCH 2/2] feat: pytest output

---
 multiplex/execute/pytest_runner.py  |  2 +-
 tests/execute/test_pytest_runner.py | 39 ++++++++++++++++++++++++++---
 2 files changed, 37 insertions(+), 4 deletions(-)

diff --git a/multiplex/execute/pytest_runner.py b/multiplex/execute/pytest_runner.py
index 1736bbb..518ed31 100644
--- a/multiplex/execute/pytest_runner.py
+++ b/multiplex/execute/pytest_runner.py
@@ -43,7 +43,7 @@ def _execute(project_root, output_path=None, approach=None, label=None):
     if output_path is not None and approach and label:
         test_dir = Path(output_path, approach + "-test")
         test_dir.mkdir(parents=True, exist_ok=True)
-        safe = str(label).replace(os.sep, "_")
+        safe = Path(str(label).replace(os.sep, "_")).stem
         (test_dir / f"{safe}_test.txt").write_text(
             f"$ {' '.join(command)}\n# exit code: {result.returncode}\n\n"
             f"{result.stdout}"
diff --git a/tests/execute/test_pytest_runner.py b/tests/execute/test_pytest_runner.py
index f2fddc0..c0a1a28 100644
--- a/tests/execute/test_pytest_runner.py
+++ b/tests/execute/test_pytest_runner.py
@@ -65,6 +65,37 @@ def raise_fnf(*a, **k):
     assert "pytest" in str(exc.value).lower()
 
 
+def _run_reporting(returncode, stdout):
+    def run(command, *args, **kwargs):
+        return types.SimpleNamespace(returncode=returncode, stdout=stdout)
+
+    return run
+
+
+def test_execute_writes_test_output_file_when_labelled(tmp_path, monkeypatch):
+    # Given an output path, approach and label, _execute persists the run's
+    # captured output to <approach>-test/<label without extension>_test.txt.
+    monkeypatch.setattr(
+        pytest_runner.subprocess, "run", _run_reporting(1, "1 failed, 2 passed\n")
+    )
+
+    assert pytest_runner._execute("proj", tmp_path, "basic", "mutant_killed.py") is False
+
+    # Same naming as the Defects4J runner: the extension is not carried over.
+    out_file = tmp_path / "basic-test" / "mutant_killed_test.txt"
+    assert out_file.exists()
+    body = out_file.read_text()
+    assert "1 failed, 2 passed" in body
+    assert "# exit code: 1" in body
+
+
+def test_execute_writes_no_output_file_without_a_label(tmp_path, monkeypatch):
+    monkeypatch.setattr(pytest_runner.subprocess, "run", _run_reporting(0, "2 passed\n"))
+
+    assert pytest_runner._execute("proj") is True
+    assert list(tmp_path.iterdir()) == []
+
+
 # --------------------------------------------------------------------------- #
 # run_mutants: full evaluation loop over a mutants directory                   #
 # --------------------------------------------------------------------------- #
@@ -127,7 +158,9 @@ def test_run_mutants_writes_summary_with_correct_classification(project, monkeyp
     # Stand in for `pytest`: the suite fails (mutant killed) only when the
     # injected source returns n + 1; every other state passes.
     monkeypatch.setattr(
-        pytest_runner, "_execute", lambda project_root: "return n + 1" not in Path(src).read_text()
+        pytest_runner,
+        "_execute",
+        lambda project_root, *args, **kwargs: "return n + 1" not in Path(src).read_text(),
     )
 
     _run(project)
@@ -142,7 +175,7 @@ def test_run_mutants_writes_summary_with_correct_classification(project, monkeyp
 
 
 def test_run_mutants_raises_when_baseline_fails(project, monkeypatch):
-    monkeypatch.setattr(pytest_runner, "_execute", lambda project_root: False)
+    monkeypatch.setattr(pytest_runner, "_execute", lambda project_root, *args, **kwargs: False)
     with pytest.raises(IOError):
         _run(project)
 
@@ -155,7 +188,7 @@ def test_run_mutants_skips_tests_for_noncompilable_mutant(project, monkeypatch):
 
     calls = {"n": 0}
 
-    def counting_execute(project_root):
+    def counting_execute(project_root, *args, **kwargs):
         calls["n"] += 1
         return True