Fix filename decoding issue for zip files archived by macOS

jupyter_archive/handlers.py

@@ -223,17 +223,42 @@ async def get(self, archive_path, include_body=False):
         self.finish()
 
     def extract_archive(self, archive_path):
+        def try_macos_decode(fn):
+            try:
+                return fn.encode('cp437').decode('utf-8')
+            except UnicodeError:
+                return None
+
+        def try_windows_chinese_decode(fn):
+            try:
+                return fn.encode('cp437').decode('gbk')
+            except UnicodeError:
+                return None
+
+        def unzip(reader, destination):
+            # most of Windows implementations use DOS (OEM) encoding
+            # Mac OS zip utility uses utf-8, but it doesn't set utf-8 bit flags
+            # *nix zip utilities silently uses system encoding(utf-8 generally)
+            with reader as f:
+                for fn in f.namelist():
+                    extreact_path = pathlib.Path(f.extract(fn, path=destination))
+                    correct_filename = try_macos_decode(fn) or try_windows_chinese_decode(fn) or fn
+                    extreact_path.rename(os.path.join(destination, correct_filename))
 
         archive_destination = archive_path.parent
         self.log.info("Begin extraction of {} to {}.".format(archive_path, archive_destination))
 
         archive_reader = make_reader(archive_path)
-        with archive_reader as archive:
-            archive.extractall(archive_destination)
+        if isinstance(archive_reader, zipfile.ZipFile):
+            unzip(archive_reader, archive_destination)
+        else:
+            with archive_reader as archive:
+                archive.extractall(archive_destination)
 
         self.log.info("Finished extracting {} to {}.".format(archive_path, archive_destination))
 
 
+
 def setup_handlers(web_app):
     host_pattern = ".*$"
     base_url = web_app.settings["base_url"]