Don't distinguish between str and bytes when serializing

This makes it easier to use some requests like ReqPqRequest which
needs a string of bytes, not a valid utf-8 string per se.

telethon/tl/tlobject.py

@@ -87,6 +87,9 @@ class TLObject:
     @staticmethod
     def serialize_bytes(data):
         """Write bytes by using Telegram guidelines"""
+        if isinstance(data, str):
+            data = data.encode('utf-8')
+
         r = []
         if len(data) < 254:
             padding = (len(data) + 1) % 4
@@ -112,10 +115,6 @@ class TLObject:
         r.append(bytes(padding))
         return b''.join(r)
 
-    @staticmethod
-    def serialize_string(string):
-        return TLObject.serialize_bytes(string.encode('utf-8'))
-
     # These should be overrode
     def to_dict(self, recursive=True):
         return {}

telethon_generator/tl_generator.py

@@ -493,7 +493,7 @@ class TLGenerator:
             builder.write("struct.pack('<d', {})".format(name))
 
         elif 'string' == arg.type:
-            builder.write('TLObject.serialize_string({})'.format(name))
+            builder.write('TLObject.serialize_bytes({})'.format(name))
 
         elif 'Bool' == arg.type:
             # 0x997275b5 if boolean else 0xbc799737