Handle problematic characters in Workunit.store_output_folder

bfabric/docs/changelog.md

@@ -10,6 +10,14 @@ Versioning currently follows `X.Y.Z` where
 
 ## \[Unreleased\]
 
+### Fixed
+
+- Handle problematic characters in `Workunit.store_output_folder`.
+
+### Added
+
+- Generic functionality in `bfabric.utils.path_safe_name` to validate names for use in paths.
+
 ## \[1.13.22\] - 2025-02-19
 
 ### Fixed

bfabric/src/bfabric/entities/workunit.py

@@ -10,6 +10,7 @@
 from bfabric.entities.core.has_container_mixin import HasContainerMixin
 from bfabric.entities.core.has_many import HasMany
 from bfabric.entities.core.has_one import HasOne
+from bfabric.utils.path_safe_name import path_safe_name
 
 if TYPE_CHECKING:
     from bfabric import Bfabric
@@ -52,8 +53,8 @@ def store_output_folder(self) -> Path:
         return Path(
             f"{self.application.storage['projectfolderprefix']}{self.container.id}",
             "bfabric",
-            self.application["technology"].replace(" ", "_"),
-            self.application["name"].replace(" ", "_"),
+            path_safe_name(self.application["technology"]),
+            path_safe_name(self.application["name"]),
             date.strftime("%Y/%Y-%m/%Y-%m-%d/"),
             f"workunit_{self.id}",
         )

bfabric/src/bfabric/utils/path_safe_name.py

@@ -0,0 +1,42 @@
+import re
+from typing import Annotated
+from pydantic import BeforeValidator
+
+# Regex pattern to match any characters that aren't alphanumeric, underscore, or hyphen
+_regex = re.compile(r"[^a-zA-Z0-9_-]+")
+
+
+def path_safe_name(name: str) -> str:
+    """
+    Convert a string to a filesystem-safe format by replacing unsafe characters with underscores.
+
+    Args:
+        name: The string to be converted to a path-safe format
+
+    Returns:
+        A string with all characters not matching [a-zA-Z0-9_-] replaced with underscores
+
+    Example:
+        >>> path_safe_name("Hello World!")
+        'Hello_World_'
+        >>> path_safe_name("file:name.txt")
+        'file_name_txt'
+    """
+    return _regex.sub("_", name)
+
+
+# Custom type that automatically converts strings to path-safe format
+PathSafeStr = Annotated[str, BeforeValidator(path_safe_name)]
+"""
+A custom Pydantic type that automatically sanitizes strings for safe use in file paths.
+When a field is annotated with this type, any unsafe characters will be replaced with underscores.
+
+Example:
+    >>> from pydantic import BaseModel
+    >>> class FileModel(BaseModel):
+    ...     filename: PathSafeStr
+    ...
+    >>> file = FileModel(filename="my file:name.txt")
+    >>> file.filename
+    'my_file_name_txt'
+"""

tests/bfabric/entities/test_workunit.py

@@ -105,6 +105,20 @@ def test_store_output_folder(mocker, mock_workunit) -> None:
     assert Path("xyz12/bfabric/tech/my_app/2024/2024-01/2024-01-02/workunit_30000") == mock_workunit.store_output_folder
 
 
+def test_store_output_folder_when_unsafe_chars(mocker, mock_workunit) -> None:
+    mock_application = mocker.MagicMock(storage={"projectfolderprefix": "xyz"})
+    mock_application.__getitem__.side_effect = {
+        "technology": "tech & tech",
+        "name": "my app (:",
+    }.__getitem__
+    mocker.patch.object(mock_workunit, "application", mock_application)
+    mocker.patch.object(Workunit, "container", mocker.PropertyMock(return_value=mocker.MagicMock(id=12)))
+    assert (
+        Path("xyz12/bfabric/tech_tech/my_app_/2024/2024-01/2024-01-02/workunit_30000")
+        == mock_workunit.store_output_folder
+    )
+
+
 def test_repr() -> None:
     workunit = Workunit({"id": 30000}, client=None)
     assert repr(workunit) == "Workunit({'id': 30000}, client=None)"