fix: move to chunk based writing

spleeter/model/provider/github.py

@@ -18,7 +18,6 @@ import tarfile
 
 from os import environ
 from tempfile import TemporaryFile
-from shutil import copyfileobj
 
 import requests
 
@@ -60,14 +59,18 @@ class GithubModelProvider(ModelProvider):
             self._release,
             name)
         get_logger().info('Downloading model archive %s', url)
-        response = requests.get(url, stream=True)
+        with requests.get(url, stream=True) as response:
-        if response.status_code != 200:
+            # Note: check for error logging here or upstream ?
-            raise IOError(f'Resource {url} not found')
+            response.raise_for_status()
-        with TemporaryFile() as stream:
+            with TemporaryFile() as stream:
-            copyfileobj(response.raw, stream)
+                # Note: check for chunk size parameters ?
-            get_logger().info('Extracting downloaded %s archive', name)
+                for chunk in response.iter_content(chunk_size=8192):
-            stream.seek(0)
+                    if chunk:
-            tar = tarfile.open(fileobj=stream)
+                        stream.write(chunk)
-            tar.extractall(path=path)
+                get_logger().info('Extracting downloaded %s archive', name)
-            tar.close()
+                stream.seek(0)
+                tar = tarfile.open(fileobj=stream)
+                tar.extractall(path=path)
+                tar.close()
+            # TODO: perform checksum control
         get_logger().info('%s model file(s) extracted', name)