Merge pull request #66 from ArcanaFramework/develop

Bug fix for converter registration

Author: tclose

.github/workflows/ci-cd.yml

@@ -39,11 +39,11 @@ jobs:
       with:
         python-version: ${{ matrix.python-version }}
     - name: Update build tools
-      run: python3 -m pip install --upgrade pip
+      run: python3 -m pip install --break-system-packages --upgrade pip
     - name: Install Package
-      run: python3 -m pip install -e .[test]
+      run: python3 -m pip install --break-system-packages -e .[test]
     - name: Install Extras Package
-      run: python3 -m pip install -e ./extras[test]
+      run: python3 -m pip install --break-system-packages -e ./extras[test]
     - name: Change out of root directory
       run: cd docs
     - name: Pytest
@@ -76,7 +76,7 @@ jobs:
       with:
         python-version: '3.12'
     - name: Install build tools
-      run: python3 -m pip install build twine
+      run: python3 -m pip install --break-system-packages build twine
     - name: Build source and wheel distributions
       run: python3 -m build ${{ matrix.pkg[1] }}
     - name: Check distributions
@@ -102,7 +102,7 @@ jobs:
         python-version: 3.x
     - name: Install dependencies
       run: |
-        python3 -m pip install --upgrade pip
+        python3 -m pip install --break-system-packages --upgrade pip
         pip install .[docs]
     - name: Build documentation
       run: |

fileformats/core/datatype.py

@@ -141,7 +141,7 @@ def from_mime(cls, mime_string):
             # treats it). Therefore, we loop through all subclasses across the different
             # namespaces to find one that matches the name.
             format_name = format_name[2:]  # remove "x-" prefix
-            matching_name = FileSet.formats_by_name[format_name]
+            matching_name = FileSet.formats_by_name.get(format_name, ())
             matching_name = [
                 m
                 for m in matching_name

fileformats/core/fileset.py

@@ -3,6 +3,7 @@
 import struct
 from enum import Enum, IntEnum
 from warnings import warn
+import inspect
 import tempfile
 from collections import Counter
 import typing as ty
@@ -625,7 +626,8 @@ def register_converter(
             raise FormatConversionError(
                 f"Cannot register converter from {source_format.__name__} "
                 f"to {cls.__name__}, {describe_task(task)}, because there is already "
-                f"one registered from {describe_task(prev_task)}"
+                f"one registered from {describe_task(prev_task)}:"
+                f"\n\n{inspect.getsource(task)}\n\nand{inspect.getsource(prev_task)}\n\n"
             )
         converters_dict[source_format] = converter_tuple
 
@@ -905,7 +907,10 @@ def mock(cls, *fspaths: ty.Tuple[ty.Union[Path, str]]) -> "FileSet":
 
     @classmethod
     def sample(
-        cls, dest_dir: ty.Optional[Path] = None, seed: int = 0, stem: str = None
+        cls,
+        dest_dir: ty.Optional[Path] = None,
+        seed: ty.Union[int, str] = 0,
+        stem: str = None,
     ) -> Self:
         """Return an sample instance of the file-set type for classes where the
         `test_data` extra has been implemented

fileformats/core/mixin.py

@@ -659,13 +659,13 @@ def namespace(cls):  # pylint: disable=no-self-argument
             namespace = super().namespace
         return namespace
 
-    @property
-    def type_name(self):
+    @classproperty
+    def type_name(cls):
         """Name of type including classifiers to be used in __repr__"""
-        if self.is_classified:
-            unclassified = self.unclassified.__name__
+        if cls.is_classified:
+            unclassified = cls.unclassified.__name__
         else:
-            unclassified = type(self).__name__
+            unclassified = type(cls).__name__
         return (
-            unclassified + "[" + ", ".join(t.type_name for t in self.classifiers) + "]"
+            unclassified + "[" + ", ".join(t.type_name for t in cls.classifiers) + "]"
         )

fileformats/core/sampling.py

@@ -20,7 +20,7 @@ class SampleFileGenerator:
     """
 
     dest_dir: Path
-    seed: int
+    seed: ty.Union[int, str]
     fname_stem: str
 
     FNAME_STEM_LENGTH = 24

fileformats/core/utils.py

@@ -14,7 +14,9 @@
 logger = logging.getLogger("fileformats")
 
 
-_excluded_subpackages = set(["core", "testing", "serialization", "archive", "document"])
+_excluded_subpackages = set(
+    ["core", "testing", "serialization", "archive", "document", "conftest"]
+)
 
 
 def include_testing_package(flag: bool = True):
@@ -137,7 +139,10 @@ def describe_task(task):
     if inspect.isfunction(task):
         import cloudpickle
 
-        task = cloudpickle.loads(task().inputs._func)
+        try:
+            task = cloudpickle.loads(task().inputs._func)
+        except Exception as e:
+            return f"{task} (Failed to load task function: {e})"
     src_file = inspect.getsourcefile(task)
     src_line = inspect.getsourcelines(task)[-1]
     return f"{task} (defined at line {src_line} of {src_file})"
@@ -146,9 +151,8 @@ def describe_task(task):
 def matching_source(task1, task2) -> bool:
     """Checks to see if the tasks share the same source code but are just getting reimported
     for some unknown reason"""
-    return (
-        inspect.getsourcefile(task1) == inspect.getsourcefile(task2)
-        and inspect.getsourcelines(task1)[-1] == inspect.getsourcelines(task2)[-1]
+    return inspect.getsource(inspect.getmodule(task1)) == inspect.getsource(
+        inspect.getmodule(task2)
     )