Merge branch 'master' into patch-1

readthedocs/concepts/strings.rst

@@ -3,7 +3,7 @@ String-based Debugging
 ======================
 
 Debugging is *really* important. Telegram's API is really big and there
-is a lot of things that you should know. Such as, what attributes or fields
+are a lot of things that you should know. Such as, what attributes or fields
 does a result have? Well, the easiest thing to do is printing it:
 
 .. code-block:: python
@@ -32,7 +32,7 @@ Can we get better than the shown string, though? Yes!
 
     print(entity.stringify())
 
-Will show a much better:
+Will show a much better representation:
 
 .. code-block:: python
 

telethon/client/telegrambaseclient.py

@@ -1,4 +1,5 @@
 import abc
+import re
 import asyncio
 import collections
 import logging
@@ -161,11 +162,11 @@ class TelegramBaseClient(abc.ABC):
 
         device_model (`str`, optional):
             "Device model" to be sent when creating the initial connection.
-            Defaults to ``platform.node()``.
+            Defaults to 'PC (n)bit' derived from ``platform.uname().machine``, or its direct value if unknown.
 
         system_version (`str`, optional):
             "System version" to be sent when creating the initial connection.
-            Defaults to ``platform.system()``.
+            Defaults to ``platform.uname().release`` stripped of everything ahead of -.
 
         app_version (`str`, optional):
             "App version" to be sent when creating the initial connection.
@@ -320,11 +321,19 @@ class TelegramBaseClient(abc.ABC):
         # Used on connection. Capture the variables in a lambda since
         # exporting clients need to create this InvokeWithLayerRequest.
         system = platform.uname()
+
+        if system.machine in ('x86_64', 'AMD64'):
+            default_device_model = 'PC 64bit'
+        elif system.machine in ('i386','i686','x86'):
+            default_device_model = 'PC 32bit'
+        else:
+            default_device_model = system.machine
+        default_system_version = re.sub(r'-.+','',system.release)
         self._init_with = lambda x: functions.InvokeWithLayerRequest(
             LAYER, functions.InitConnectionRequest(
                 api_id=self.api_id,
-                device_model=device_model or system.system or 'Unknown',
+                device_model=device_model or default_device_model or 'Unknown',
-                system_version=system_version or system.release or '1.0',
+                system_version=system_version or default_system_version or '1.0',
                 app_version=app_version or self.__version__,
                 lang_code=lang_code,
                 system_lang_code=system_lang_code,

telethon/client/updates.py

@@ -467,6 +467,8 @@ class UpdateMethods:
         # the name of speed; we don't want to make it worse for all updates
         # just because albums may need it.
         for builder, callback in self._event_builders:
+            if isinstance(builder, events.Raw):
+                continue
             if not isinstance(event, builder.Event):
                 continue
 

telethon/utils.py

@@ -601,16 +601,52 @@ def get_message_id(message):
 
 
 def _get_metadata(file):
-    # `hachoir` only deals with paths to in-disk files, while
+    if not hachoir:
-    # `_get_extension` supports a few other things. The parser
+        return
-    # may also fail in any case and we don't want to crash if
+
+    stream = None
+    close_stream = True
+    seekable = True
+
+    # The parser may fail and we don't want to crash if
     # the extraction process fails.
-    if hachoir and isinstance(file, str) and os.path.isfile(file):
+    try:
-        try:
+        # Note: aiofiles are intentionally left out for simplicity
-            with hachoir.parser.createParser(file) as parser:
+        if isinstance(file, str):
-                return hachoir.metadata.extractMetadata(parser)
+            stream = open(file, 'rb')
-        except Exception as e:
+        elif isinstance(file, bytes):
-            _log.warning('Failed to analyze %s: %s %s', file, e.__class__, e)
+            stream = io.BytesIO(file)
+        else:
+            stream = file
+            close_stream = False
+            if getattr(file, 'seekable', None):
+                seekable = file.seekable()
+            else:
+                seekable = False
+
+        if not seekable:
+            return None
+
+        pos = stream.tell()
+        filename = getattr(file, 'name', '')
+
+        parser = hachoir.parser.guess.guessParser(hachoir.stream.InputIOStream(
+            stream,
+            source='file:' + filename,
+            tags=[],
+            filename=filename
+        ))
+
+        return hachoir.metadata.extractMetadata(parser)
+
+    except Exception as e:
+        _log.warning('Failed to analyze %s: %s %s', file, e.__class__, e)
+
+    finally:
+        if stream and close_stream:
+            stream.close()
+        elif stream and seekable:
+            stream.seek(pos)
 
 
 def get_attributes(file, *, attributes=None, mime_type=None,
@@ -803,15 +839,31 @@ def is_gif(file):
 
 
 def is_audio(file):
-    """Returns `True` if the file extension looks like an audio file."""
+    """Returns `True` if the file has an audio mime type."""
-    file = 'a' + _get_extension(file)
+    ext = _get_extension(file)
-    return (mimetypes.guess_type(file)[0] or '').startswith('audio/')
+    if not ext:
+        metadata = _get_metadata(file)
+        if metadata and metadata.has('mime_type'):
+            return metadata.get('mime_type').startswith('audio/')
+        else:
+            return False
+    else:
+        file = 'a' + ext
+        return (mimetypes.guess_type(file)[0] or '').startswith('audio/')
 
 
 def is_video(file):
-    """Returns `True` if the file extension looks like a video file."""
+    """Returns `True` if the file has a video mime type."""
-    file = 'a' + _get_extension(file)
+    ext = _get_extension(file)
-    return (mimetypes.guess_type(file)[0] or '').startswith('video/')
+    if not ext:
+        metadata = _get_metadata(file)
+        if metadata and metadata.has('mime_type'):
+            return metadata.get('mime_type').startswith('video/')
+        else:
+            return False
+    else:
+        file = 'a' + ext
+        return (mimetypes.guess_type(file)[0] or '').startswith('video/')
 
 
 def is_list_like(obj):