Fix uploading files with unicode filenames (#1214)

* Patch urllib3RequestField to make it *not* support RFC2231

* Add new required test data file

* Fix on py2

* Remove weird legacy code from inputfile

Not needed anymore, and also makes it not work... so that was not great

telegram/files/inputfile.py

@@ -22,7 +22,6 @@
 import imghdr
 import mimetypes
 import os
-import sys
 from uuid import uuid4
 
 from telegram import TelegramError
@@ -74,10 +73,6 @@ class InputFile(object):
         if not self.filename or '.' not in self.filename:
             self.filename = self.mimetype.replace('/', '.')
 
-        if sys.version_info < (3,):
-            if isinstance(self.filename, unicode):  # flake8: noqa  pylint: disable=E0602
-                self.filename = self.filename.encode('utf-8', 'replace')
-
     @property
     def field_tuple(self):
         return self.filename, self.input_file_content, self.mimetype

telegram/utils/request.py

@@ -36,6 +36,7 @@ try:
     import telegram.vendor.ptb_urllib3.urllib3.contrib.appengine as appengine
     from telegram.vendor.ptb_urllib3.urllib3.connection import HTTPConnection
     from telegram.vendor.ptb_urllib3.urllib3.util.timeout import Timeout
+    from telegram.vendor.ptb_urllib3.urllib3.fields import RequestField
 except ImportError:  # pragma: no cover
     warnings.warn("python-telegram-bot wasn't properly installed. Please refer to README.rst on "
                   "how to properly install.")
@@ -45,6 +46,20 @@ from telegram import (InputFile, TelegramError, InputMedia)
 from telegram.error import (Unauthorized, NetworkError, TimedOut, BadRequest, ChatMigrated,
                             RetryAfter, InvalidToken)
 
+
+def _render_part(self, name, value):
+    """
+    Monkey patch urllib3.urllib3.fields.RequestField to make it *not* support RFC2231 compliant
+    Content-Disposition headers since telegram servers don't understand it. Instead just escape
+    \ and " and replace any \n and \r with a space.
+    """
+    value = value.replace(u'\\', u'\\\\').replace(u'"', u'\\"')
+    value = value.replace(u'\r', u' ').replace(u'\n', u' ')
+    return u'%s="%s"' % (name, value)
+
+
+RequestField._render_part = _render_part
+
 logging.getLogger('urllib3').setLevel(logging.WARNING)
 
 USER_AGENT = 'Python Telegram Bot (https://github.com/python-telegram-bot/python-telegram-bot)'

tests/test_photo.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-#
+# -*- coding: utf-8 -*-
 # A library that provides a Python interface to the Telegram Bot API
 # Copyright (C) 2015-2018
 # Leandro Toledo de Souza <devs@python-telegram-bot.org>
@@ -195,6 +195,21 @@ class TestPhoto(object):
         assert isinstance(photo.file_id, str)
         assert photo.file_id != ''
 
+    @flaky(3, 1)
+    @pytest.mark.timeout(10)
+    def test_send_file_unicode_filename(self, bot, chat_id):
+        """
+        Regression test for https://github.com/python-telegram-bot/python-telegram-bot/issues/1202
+        """
+        with open(u'tests/data/测试.png', 'rb') as f:
+            message = bot.send_photo(photo=f, chat_id=chat_id)
+
+        photo = message.photo[-1]
+
+        assert isinstance(photo, PhotoSize)
+        assert isinstance(photo.file_id, str)
+        assert photo.file_id != ''
+
     @flaky(3, 1)
     @pytest.mark.timeout(10)
     def test_send_bytesio_jpg_file(self, bot, chat_id):