[torchcodec] simple video decoder constructor now accepts path objects (

#60)

Summary:
Pull Request resolved: #60

Addresses TODO in code. `SimpleVideoDecoder` now accepts a `pathlib.Path` object directly. We still have to call a string constructor on it, as the underlying core function needs to take a string, but this is a usability improvement.

Reviewed By: ahmadsharif1

Differential Revision: D59325727

fbshipit-source-id: 2d2e1b8414046fb48d5d2d6631b0abda56a8b86b

src/torchcodec/decoders/_simple_video_decoder.py

@@ -1,3 +1,4 @@
+from pathlib import Path
 from typing import Union
 
 import torch
@@ -9,17 +10,18 @@ class SimpleVideoDecoder:
     """TODO: Add docstring."""
 
     def __init__(self, source: Union[str, bytes, torch.Tensor]):
-        # TODO: support Path objects.
         if isinstance(source, str):
             self._decoder = core.create_from_file(source)
+        elif isinstance(source, Path):
+            self._decoder = core.create_from_file(str(source))
         elif isinstance(source, bytes):
             self._decoder = core.create_from_bytes(source)
         elif isinstance(source, torch.Tensor):
             self._decoder = core.create_from_tensor(source)
         else:
             raise TypeError(
                 f"Unknown source type: {type(source)}. "
-                "Supported types are str, bytes and Tensor."
+                "Supported types are str, Path, bytes and Tensor."
             )
 
         core.add_video_stream(self._decoder)

test/decoders/simple_video_decoder_test.py

@@ -7,10 +7,12 @@
 
 
 class TestSimpleDecoder:
-    @pytest.mark.parametrize("source_kind", ("path", "tensor", "bytes"))
+    @pytest.mark.parametrize("source_kind", ("str", "path", "tensor", "bytes"))
     def test_create(self, source_kind):
-        if source_kind == "path":
+        if source_kind == "str":
             source = str(NASA_VIDEO.path)
+        elif source_kind == "path":
+            source = NASA_VIDEO.path
         elif source_kind == "tensor":
             source = NASA_VIDEO.to_tensor()
         elif source_kind == "bytes":
@@ -35,7 +37,7 @@ def test_create_fails(self):
             decoder = SimpleVideoDecoder(123)  # noqa
 
     def test_getitem_int(self):
-        decoder = SimpleVideoDecoder(str(NASA_VIDEO.path))
+        decoder = SimpleVideoDecoder(NASA_VIDEO.path)
 
         ref_frame0 = NASA_VIDEO.get_tensor_by_index(0)
         ref_frame1 = NASA_VIDEO.get_tensor_by_index(1)
@@ -48,7 +50,7 @@ def test_getitem_int(self):
         assert_tensor_equal(ref_frame_last, decoder[-1])
 
     def test_getitem_slice(self):
-        decoder = SimpleVideoDecoder(str(NASA_VIDEO.path))
+        decoder = SimpleVideoDecoder(NASA_VIDEO.path)
 
         # ensure that the degenerate case of a range of size 1 works
 
@@ -185,7 +187,7 @@ def test_getitem_slice(self):
             assert_tensor_equal(sliced, ref)
 
     def test_getitem_fails(self):
-        decoder = SimpleVideoDecoder(str(NASA_VIDEO.path))
+        decoder = SimpleVideoDecoder(NASA_VIDEO.path)
 
         with pytest.raises(IndexError, match="out of bounds"):
             frame = decoder[1000]  # noqa
@@ -197,7 +199,7 @@ def test_getitem_fails(self):
             frame = decoder["0"]  # noqa
 
     def test_iteration(self):
-        decoder = SimpleVideoDecoder(str(NASA_VIDEO.path))
+        decoder = SimpleVideoDecoder(NASA_VIDEO.path)
 
         ref_frame0 = NASA_VIDEO.get_tensor_by_index(0)
         ref_frame1 = NASA_VIDEO.get_tensor_by_index(1)