Make _open_file() return a context manager

Summary: Make the `_open_file()` function return a context manager and remove the associated file closure

Reviewed By: nikhilaravi

Differential Revision: D20720506

fbshipit-source-id: 7d96ceb2fd64b6ee3985d0b0faf8d8bf791b1966

pytorch3d/io/mtl_io.py

@@ -379,13 +379,13 @@ def _bilinear_interpolation_grid_sample(
     return out.permute(0, 2, 3, 1)
 
 
-def load_mtl(f_mtl, material_names: List, data_dir: str, device="cpu"):
+def load_mtl(f, material_names: List, data_dir: str, device="cpu"):
     """
     Load texture images and material reflectivity values for ambient, diffuse
     and specular light (Ka, Kd, Ks, Ns).
 
     Args:
-        f_mtl: a file like object of the material information.
+        f: a file-like object of the material information.
         material_names: a list of the material names found in the .obj file.
         data_dir: the directory where the material texture files are located.
 
@@ -414,39 +414,36 @@ def load_mtl(f_mtl, material_names: List, data_dir: str, device="cpu"):
     texture_images = {}
     material_name = ""
 
-    f_mtl, new_f = _open_file(f_mtl)
-    lines = [line.strip() for line in f_mtl]
-    for line in lines:
-        if len(line.split()) != 0:
-            if line.split()[0] == "newmtl":
-                material_name = line.split()[1]
-                material_colors[material_name] = {}
-            if line.split()[0] == "map_Kd":
-                # Texture map.
-                texture_files[material_name] = line.split()[1]
-            if line.split()[0] == "Kd":
-                # RGB diffuse reflectivity
-                kd = np.array(list(line.split()[1:4])).astype(np.float32)
-                kd = torch.from_numpy(kd).to(device)
-                material_colors[material_name]["diffuse_color"] = kd
-            if line.split()[0] == "Ka":
-                # RGB ambient reflectivity
-                ka = np.array(list(line.split()[1:4])).astype(np.float32)
-                ka = torch.from_numpy(ka).to(device)
-                material_colors[material_name]["ambient_color"] = ka
-            if line.split()[0] == "Ks":
-                # RGB specular reflectivity
-                ks = np.array(list(line.split()[1:4])).astype(np.float32)
-                ks = torch.from_numpy(ks).to(device)
-                material_colors[material_name]["specular_color"] = ks
-            if line.split()[0] == "Ns":
-                # Specular exponent
-                ns = np.array(list(line.split()[1:4])).astype(np.float32)
-                ns = torch.from_numpy(ns).to(device)
-                material_colors[material_name]["shininess"] = ns
-
-    if new_f:
-        f_mtl.close()
+    with _open_file(f) as f:
+        lines = [line.strip() for line in f]
+        for line in lines:
+            if len(line.split()) != 0:
+                if line.split()[0] == "newmtl":
+                    material_name = line.split()[1]
+                    material_colors[material_name] = {}
+                if line.split()[0] == "map_Kd":
+                    # Texture map.
+                    texture_files[material_name] = line.split()[1]
+                if line.split()[0] == "Kd":
+                    # RGB diffuse reflectivity
+                    kd = np.array(list(line.split()[1:4])).astype(np.float32)
+                    kd = torch.from_numpy(kd).to(device)
+                    material_colors[material_name]["diffuse_color"] = kd
+                if line.split()[0] == "Ka":
+                    # RGB ambient reflectivity
+                    ka = np.array(list(line.split()[1:4])).astype(np.float32)
+                    ka = torch.from_numpy(ka).to(device)
+                    material_colors[material_name]["ambient_color"] = ka
+                if line.split()[0] == "Ks":
+                    # RGB specular reflectivity
+                    ks = np.array(list(line.split()[1:4])).astype(np.float32)
+                    ks = torch.from_numpy(ks).to(device)
+                    material_colors[material_name]["specular_color"] = ks
+                if line.split()[0] == "Ns":
+                    # Specular exponent
+                    ns = np.array(list(line.split()[1:4])).astype(np.float32)
+                    ns = torch.from_numpy(ns).to(device)
+                    material_colors[material_name]["shininess"] = ns
 
     # Only keep the materials referenced in the obj.
     for name in material_names:

pytorch3d/io/obj_io.py

@@ -72,7 +72,7 @@ def _format_faces_indices(faces_indices, max_index, device, pad_value=None):
 
 
 def load_obj(
-    f_obj,
+    f,
     load_textures=True,
     create_texture_atlas: bool = False,
     texture_atlas_size: int = 4,
@@ -211,14 +211,13 @@ def load_obj(
               None.
     """
     data_dir = "./"
-    if isinstance(f_obj, (str, bytes, os.PathLike)):
+    if isinstance(f, (str, bytes, os.PathLike)):
         # pyre-fixme[6]: Expected `_PathLike[Variable[typing.AnyStr <: [str,
         #  bytes]]]` for 1st param but got `Union[_PathLike[typing.Any], bytes, str]`.
-        data_dir = os.path.dirname(f_obj)
-    f_obj, new_f = _open_file(f_obj, "r")
-    try:
-        return _load(
-            f_obj,
+        data_dir = os.path.dirname(f)
+    with _open_file(f, "r") as f:
+        return _load_obj(
+            f,
             data_dir,
             load_textures=load_textures,
             create_texture_atlas=create_texture_atlas,
@@ -226,9 +225,6 @@ def load_obj(
             texture_wrap=texture_wrap,
             device=device,
         )
-    finally:
-        if new_f:
-            f_obj.close()
 
 
 def load_objs_as_meshes(files: list, device=None, load_textures: bool = True):
@@ -338,7 +334,7 @@ def _parse_face(
         faces_materials_idx.append(material_idx)
 
 
-def _load(
+def _load_obj(
     f_obj,
     data_dir,
     load_textures: bool = True,
@@ -523,12 +519,8 @@ def save_obj(f, verts, faces, decimal_places: Optional[int] = None):
         message = "Argument 'faces' should either be empty or of shape (num_faces, 3)."
         raise ValueError(message)
 
-    f, new_f = _open_file(f, "w")
-    try:
+    with _open_file(f, "w") as f:
         return _save(f, verts, faces, decimal_places)
-    finally:
-        if new_f:
-            f.close()
 
 
 # TODO (nikhilar) Speed up this function.

pytorch3d/io/ply_io.py

@@ -603,13 +603,8 @@ def _load_ply_raw(f) -> Tuple[_PlyHeader, dict]:
                   uniformly-sized list, then the value will be a 2D numpy array.
                   If not, it is a list of the relevant property values.
     """
-    f, new_f = _open_file(f, "rb")
-    try:
+    with _open_file(f, "rb") as f:
         header, elements = _load_ply_raw_stream(f)
-    finally:
-        if new_f:
-            f.close()
-
     return header, elements
 
 
@@ -800,9 +795,5 @@ def save_ply(
         message = "Argument 'verts_normals' should either be empty or of shape (num_verts, 3)."
         raise ValueError(message)
 
-    f, new_f = _open_file(f, "w")
-    try:
+    with _open_file(f, "w") as f:
         _save_ply(f, verts, faces, verts_normals, decimal_places)
-    finally:
-        if new_f:
-            f.close()

pytorch3d/io/utils.py

 # Copyright (c) Facebook, Inc. and its affiliates. All rights reserved.
 
+import contextlib
 import pathlib
+from typing import IO, ContextManager
 
 import numpy as np
 from fvcore.common.file_io import PathManager
 from PIL import Image
 
 
-# TODO(plabatut): Replace with a context manager
-def _open_file(f, mode="r"):
-    new_f = False
+def _open_file(f, mode="r") -> ContextManager[IO]:
     if isinstance(f, str):
-        new_f = True
         f = open(f, mode)
+        return contextlib.closing(f)
     elif isinstance(f, pathlib.Path):
-        new_f = True
         f = f.open(mode)
-    return f, new_f
+        return contextlib.closing(f)
+    else:
+        return contextlib.nullcontext(f)
 
 
 def _read_image(file_name: str, format=None):