Accept UTF-8 byte-order marks in OBJ and material files

README.md

@@ -53,7 +53,8 @@ scene = pywavefront.Wavefront('something.obj')
 A more complex example
 
 * `strict` (Default: `False`) will raise an exception if unsupported features are found in the obj or mtl file
-* `encoding` (Default: `utf-8`) of the obj and mtl file(s)
+* `encoding` (Default: `utf-8`) of the obj and mtl file(s).
+  UTF-8 files with or without a byte-order mark are accepted automatically.
 * `create_materials` (Default: `False`) will create materials if mtl file is missing or obj file references non-existing materials
 * `collect_faces` (Default: `False`) will collect triangle face data for every mesh. In case faces with more than three vertices are specified they will be triangulated. See the documentation of `ObjParser#consume_faces()` in [`obj.py`](https://github.com/pywavefront/PyWavefront/blob/master/pywavefront/obj.py).
 * `parse` (Default: `True`) decides if parsing should start immediately.

pywavefront/parser.py

@@ -77,17 +77,22 @@ def create_line_generator(self):
         Should only yield non-empty lines
         """
 
+        encoding = self.encoding
+        if encoding and codecs.lookup(encoding).name == 'utf-8':
+            # Consume a UTF-8 signature at the start, preserving U+FEFF in content.
+            encoding = 'utf-8-sig'
+
         if self.file_name.suffix == ".gz":
             # FIXME: Converting to str for now for py34 compatibility
-            gz = gzip.open(str(self.file_name), mode='rt', encoding=self.encoding)
+            gz = gzip.open(str(self.file_name), mode='rt', encoding=encoding)
 
             for line in gz.readlines():
                 yield line
 
             gz.close()
         else:
             # FIXME: Converting to str for now for py34 compatibility
-            file = open(str(self.file_name), mode='r', encoding=self.encoding)
+            file = open(str(self.file_name), mode='r', encoding=encoding)
 
             for line in file:
                 yield line

tests/test_encoding.py

@@ -0,0 +1,56 @@
+import codecs
+import gzip
+
+import pytest
+
+import pywavefront
+
+
+def write_source(path, text, compressed, bom=False, encoding='utf-8'):
+    data = text.encode(encoding)
+    if bom:
+        data = codecs.BOM_UTF8 + data
+    opener = gzip.open if compressed else open
+    with opener(str(path), 'wb') as stream:
+        stream.write(data)
+
+
+@pytest.mark.parametrize(
+    'obj_bom,mtl_bom', [(False, False), (True, False), (False, True), (True, True)])
+@pytest.mark.parametrize('compressed', [False, True])
+@pytest.mark.parametrize('encoding', [None, 'UTF8', 'utf-8-sig'])
+@pytest.mark.parametrize('strict', [False, True])
+def test_utf8_bom(tmp_path, obj_bom, mtl_bom, compressed, encoding, strict):
+    suffix = '.gz' if compressed else ''
+    obj_path = tmp_path / ('triangle.obj' + suffix)
+    mtl_path = tmp_path / ('triangle.mtl' + suffix)
+    # U+FEFF inside a material name is data, not a file signature.
+    name = 'caf\u00e9\ufeffpaint'
+    obj = ('v 0 0 0\nv 1 0 0\nv 0 1 0\nmtllib {0}\n'
+           'usemtl {1}\nf 1 2 3\n').format(mtl_path.name, name)
+    mtl = 'newmtl {0}\nKd 0.25 0.5 0.75\n'.format(name)
+    write_source(obj_path, obj, compressed, obj_bom)
+    write_source(mtl_path, mtl, compressed, mtl_bom)
+    kwargs = {'strict': strict, 'collect_faces': True}
+    if encoding is not None:
+        kwargs['encoding'] = encoding
+    scene = pywavefront.Wavefront(str(obj_path), **kwargs)
+    assert scene.vertices == [(0., 0., 0.), (1., 0., 0.), (0., 1., 0.)]
+    assert scene.mesh_list[0].faces == [[0, 1, 2]]
+    material = scene.materials[name]
+    assert scene.mesh_list[0].materials == [material]
+    assert material.diffuse[:3] == [0.25, 0.5, 0.75]
+    assert material.vertices == [0., 0., 0., 1., 0., 0., 0., 1., 0.]
+
+
+@pytest.mark.parametrize('encoding', ['utf-16', 'iso-8859-1'])
+@pytest.mark.parametrize('compressed', [False, True])
+def test_other_encodings(tmp_path, encoding, compressed):
+    suffix = '.gz' if compressed else ''
+    path = tmp_path / ('triangle.obj' + suffix)
+    obj = 'v 0 0 0\nv 1 0 0\nv 0 1 0\no caf\u00e9\nf 1 2 3\n'
+    write_source(path, obj, compressed, encoding=encoding)
+    scene = pywavefront.Wavefront(str(path), strict=True, encoding=encoding,
+                                  collect_faces=True)
+    assert scene.meshes['caf\u00e9'].faces == [[0, 1, 2]]
+    assert len(scene.vertices) == 3