Sanitize profile filename (#21395)

* sanitize filename

* add testing

* changelog

* fix comment

---------

Co-authored-by: Bhimraj Yadav <bhimrajyadav977@gmail.com>

requirements/pytorch/extra.txt

@@ -5,7 +5,7 @@
 matplotlib>3.1, <3.11.0
 omegaconf >=2.2.3, <2.4.0
 hydra-core >=1.2.0, <1.4.0
-jsonargparse[signatures,jsonnet] >=4.39.0, <4.44.0
+jsonargparse[signatures,jsonnet] >=4.39.0, <4.45.0
 rich >=12.3.0, <14.3.0
 tensorboardX >=2.2, <2.7.0  # min version is set by torch.onnx missing attribute
 bitsandbytes >=0.45.2,<0.47.0; platform_system != "Darwin"

src/lightning/pytorch/CHANGELOG.md

@@ -16,12 +16,19 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 -
 
+### Deprecated
+
+- Deprecated `to_torchscript` method due to deprecation of TorchScript in PyTorch ([#21397](https://github.com/Lightning-AI/pytorch-lightning/pull/21397))
+
 ### Removed
 
 - Removed support for Python 3.9 due to end-of-life status ([#21398](https://github.com/Lightning-AI/pytorch-lightning/pull/21398))
 
 ### Fixed
 
+- Sanitize profiler filenames when saving to avoid crashes due to invalid characters ([#21395](https://github.com/Lightning-AI/pytorch-lightning/pull/21395))
+
+
 - Fix `StochasticWeightAveraging` with infinite epochs ([#21396](https://github.com/Lightning-AI/pytorch-lightning/pull/21396))
 
 

src/lightning/pytorch/core/module.py

@@ -64,7 +64,7 @@ from lightning.pytorch.utilities import GradClipAlgorithmType
 from lightning.pytorch.utilities.exceptions import MisconfigurationException
 from lightning.pytorch.utilities.imports import _TORCH_GREATER_EQUAL_2_6, _TORCHMETRICS_GREATER_EQUAL_0_9_1
 from lightning.pytorch.utilities.model_helpers import _restricted_classmethod
-from lightning.pytorch.utilities.rank_zero import WarningCache, rank_zero_warn
+from lightning.pytorch.utilities.rank_zero import WarningCache, rank_zero_deprecation, rank_zero_warn
 from lightning.pytorch.utilities.signature_utils import is_param_in_hook_signature
 from lightning.pytorch.utilities.types import (
     _METRIC,
@@ -1498,6 +1498,11 @@ class LightningModule(
         scripted you should override this method. In case you want to return multiple modules, we recommend using a
         dictionary.
 
+        .. deprecated::
+            ``LightningModule.to_torchscript`` has been deprecated in v2.7 and will be removed in v2.8.
+            TorchScript is deprecated in PyTorch. Use ``torch.export.export()`` for model exporting instead.
+            See https://pytorch.org/docs/stable/export.html for more information.
+
         Args:
             file_path: Path where to save the torchscript. Default: None (no file saved).
             method: Whether to use TorchScript's script or trace method. Default: 'script'
@@ -1536,6 +1541,11 @@ class LightningModule(
             defined or not.
 
         """
+        rank_zero_deprecation(
+            "`LightningModule.to_torchscript` has been deprecated in v2.7 and will be removed in v2.8. "
+            "TorchScript is deprecated in PyTorch. Use `torch.export.export()` for model exporting instead. "
+            "See https://pytorch.org/docs/stable/export.html for more information."
+        )
         mode = self.training
 
         if method == "script":

src/lightning/pytorch/profilers/profiler.py

@@ -15,6 +15,7 @@
 
 import logging
 import os
+import re
 from abc import ABC, abstractmethod
 from collections.abc import Generator
 from contextlib import contextmanager
@@ -80,7 +81,6 @@ class Profiler(ABC):
         self,
         action_name: Optional[str] = None,
         extension: str = ".txt",
-        split_token: str = "-",  # noqa: S107
     ) -> str:
         args = []
         if self._stage is not None:
@@ -91,7 +91,14 @@ class Profiler(ABC):
             args.append(str(self._local_rank))
         if action_name is not None:
             args.append(action_name)
-        return split_token.join(args) + extension
+        base = "-".join(args)
+        # Replace a set of path-unsafe characters across platforms with '_'
+        base = re.sub(r"[\\/:*?\"<>|\n\r\t]", "_", base)
+        base = re.sub(r"_+", "_", base)
+        base = base.strip()
+        if not base:
+            base = "profile"
+        return base + extension
 
     def _prepare_streams(self) -> None:
         if self._write_stream is not None:

tests/tests_pytorch/helpers/test_models.py

@@ -46,7 +46,8 @@ def test_models(tmp_path, data_class, model_class):
     if dm is not None:
         trainer.test(model, datamodule=dm)
 
-    model.to_torchscript()
+    with pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"):
+        model.to_torchscript()
     if data_class:
         model.to_onnx(os.path.join(tmp_path, "my-model.onnx"), input_sample=dm.sample)
 

tests/tests_pytorch/models/test_torchscript.py

@@ -21,6 +21,7 @@ from fsspec.implementations.local import LocalFileSystem
 
 from lightning.fabric.utilities.cloud_io import get_filesystem
 from lightning.fabric.utilities.imports import _IS_WINDOWS, _TORCH_GREATER_EQUAL_2_4
+from lightning.fabric.utilities.rank_zero import LightningDeprecationWarning
 from lightning.pytorch.core.module import LightningModule
 from lightning.pytorch.demos.boring_classes import BoringModel
 from tests_pytorch.helpers.advanced_models import BasicGAN, ParityModuleRNN
@@ -36,7 +37,8 @@ def test_torchscript_input_output(modelclass):
     if isinstance(model, BoringModel):
         model.example_input_array = torch.randn(5, 32)
 
-    script = model.to_torchscript()
+    with pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"):
+        script = model.to_torchscript()
     assert isinstance(script, torch.jit.ScriptModule)
 
     model.eval()
@@ -59,7 +61,8 @@ def test_torchscript_example_input_output_trace(modelclass):
     if isinstance(model, BoringModel):
         model.example_input_array = torch.randn(5, 32)
 
-    script = model.to_torchscript(method="trace")
+    with pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"):
+        script = model.to_torchscript(method="trace")
     assert isinstance(script, torch.jit.ScriptModule)
 
     model.eval()
@@ -74,7 +77,8 @@ def test_torchscript_input_output_trace():
     """Test that traced LightningModule forward works with example_inputs."""
     model = BoringModel()
     example_inputs = torch.randn(1, 32)
-    script = model.to_torchscript(example_inputs=example_inputs, method="trace")
+    with pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"):
+        script = model.to_torchscript(example_inputs=example_inputs, method="trace")
     assert isinstance(script, torch.jit.ScriptModule)
 
     model.eval()
@@ -99,7 +103,8 @@ def test_torchscript_device(device_str):
     model = BoringModel().to(device)
     model.example_input_array = torch.randn(5, 32)
 
-    script = model.to_torchscript()
+    with pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"):
+        script = model.to_torchscript()
     assert next(script.parameters()).device == device
     script_output = script(model.example_input_array.to(device))
     assert script_output.device == device
@@ -121,7 +126,8 @@ def test_torchscript_device_with_check_inputs(device_str):
 
     check_inputs = torch.rand(5, 32)
 
-    script = model.to_torchscript(method="trace", check_inputs=check_inputs)
+    with pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"):
+        script = model.to_torchscript(method="trace", check_inputs=check_inputs)
     assert isinstance(script, torch.jit.ScriptModule)
 
 
@@ -129,11 +135,13 @@ def test_torchscript_retain_training_state():
     """Test that torchscript export does not alter the training mode of original model."""
     model = BoringModel()
     model.train(True)
-    script = model.to_torchscript()
+    with pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"):
+        script = model.to_torchscript()
     assert model.training
     assert not script.training
     model.train(False)
-    _ = model.to_torchscript()
+    with pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"):
+        _ = model.to_torchscript()
     assert not model.training
     assert not script.training
 
@@ -142,7 +150,8 @@ def test_torchscript_retain_training_state():
 def test_torchscript_properties(modelclass):
     """Test that scripted LightningModule has unnecessary methods removed."""
     model = modelclass()
-    script = model.to_torchscript()
+    with pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"):
+        script = model.to_torchscript()
     assert not hasattr(model, "batch_size") or hasattr(script, "batch_size")
     assert not hasattr(model, "learning_rate") or hasattr(script, "learning_rate")
     assert not callable(getattr(script, "training_step", None))
@@ -153,7 +162,8 @@ def test_torchscript_save_load(tmp_path, modelclass):
     """Test that scripted LightningModule is correctly saved and can be loaded."""
     model = modelclass()
     output_file = str(tmp_path / "model.pt")
-    script = model.to_torchscript(file_path=output_file)
+    with pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"):
+        script = model.to_torchscript(file_path=output_file)
     loaded_script = torch.jit.load(output_file)
     assert torch.allclose(next(script.parameters()), next(loaded_script.parameters()))
 
@@ -170,7 +180,8 @@ def test_torchscript_save_load_custom_filesystem(tmp_path, modelclass):
 
     model = modelclass()
     output_file = os.path.join(_DUMMY_PRFEIX, _PREFIX_SEPARATOR, tmp_path, "model.pt")
-    script = model.to_torchscript(file_path=output_file)
+    with pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"):
+        script = model.to_torchscript(file_path=output_file)
 
     fs = get_filesystem(output_file)
     with fs.open(output_file, "rb") as f:
@@ -184,7 +195,10 @@ def test_torchcript_invalid_method():
     model = BoringModel()
     model.train(True)
 
-    with pytest.raises(ValueError, match="only supports 'script' or 'trace'"):
+    with (
+        pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"),
+        pytest.raises(ValueError, match="only supports 'script' or 'trace'"),
+    ):
         model.to_torchscript(method="temp")
 
 
@@ -193,7 +207,10 @@ def test_torchscript_with_no_input():
     model = BoringModel()
     model.example_input_array = None
 
-    with pytest.raises(ValueError, match="requires either `example_inputs` or `model.example_input_array`"):
+    with (
+        pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"),
+        pytest.raises(ValueError, match="requires either `example_inputs` or `model.example_input_array`"),
+    ):
         model.to_torchscript(method="trace")
 
 
@@ -224,6 +241,17 @@ def test_torchscript_script_recursively():
 
     lm = Parent()
     assert not lm._jit_is_scripting
-    script = lm.to_torchscript(method="script")
+    with pytest.deprecated_call(match="has been deprecated in v2.7 and will be removed in v2.8"):
+        script = lm.to_torchscript(method="script")
     assert not lm._jit_is_scripting
     assert isinstance(script, torch.jit.RecursiveScriptModule)
+
+
+def test_to_torchscript_deprecation():
+    """Test that to_torchscript raises a deprecation warning."""
+    model = BoringModel()
+    model.example_input_array = torch.randn(5, 32)
+
+    with pytest.warns(LightningDeprecationWarning, match="has been deprecated in v2.7 and will be removed in v2.8"):
+        script = model.to_torchscript()
+    assert isinstance(script, torch.jit.ScriptModule)

tests/tests_pytorch/profilers/test_profiler.py

@@ -322,6 +322,33 @@ def test_advanced_profiler_dump_states(tmp_path):
     assert len(data) > 0
 
 
+@pytest.mark.parametrize("char", ["/", "\\", ":", "*", "?", '"', "<", ">", "|", "\n", "\r", "\t"])
+def test_advanced_profiler_dump_states_sanitizes_filename(tmp_path, char):
+    """Profiler should sanitize action names to produce filesystem-safe .prof filenames.
+
+    This guards against errors when callbacks or actions include path-unsafe characters (e.g., metric names with '/').
+
+    """
+    profiler = AdvancedProfiler(dirpath=tmp_path, dump_stats=True)
+    action_name = f"before{char}after"
+    with profiler.profile(action_name):
+        pass
+
+    profiler.describe()
+
+    prof_files = [f for f in os.listdir(tmp_path) if f.endswith(".prof")]
+    assert len(prof_files) == 1
+    prof_name = prof_files[0]
+
+    # Ensure none of the path-unsafe characters are present in the produced filename
+    forbidden = ["/", "\\", ":", "*", "?", '"', "<", ">", "|", "\n", "\r", "\t"]
+    for bad in forbidden:
+        assert bad not in prof_name
+
+    # File should be non-empty
+    assert (tmp_path / prof_name).read_bytes()
+
+
 def test_advanced_profiler_value_errors(advanced_profiler):
     """Ensure errors are raised where expected."""
     action = "test"