From 67b4d949fdee156a0529ae717d8d105b869e99b2 Mon Sep 17 00:00:00 2001
From: Moon <moon@shipoclu.com>
Date: Thu, 12 Jun 2025 12:53:37 +0900
Subject: [PATCH] use a library for content type detection instead of
handrolled
---
bot/pleroma_service.py | 33 ++++++++++++++-------------------
requirements.txt | 1 +
2 files changed, 15 insertions(+), 19 deletions(-)
diff --git a/bot/pleroma_service.py b/bot/pleroma_service.py
index d7c52b6..21320d1 100644
--- a/bot/pleroma_service.py
+++ b/bot/pleroma_service.py
@@ -1,7 +1,7 @@
from mastodon import Mastodon
from typing import List, Optional, Dict, Any, Union, BinaryIO
-import mimetypes
import io
+import filetype
from fediverse_service import FediverseService
from fediverse_types import (
FediverseNotification, FediversePost, FediverseUser, FediverseFile,
@@ -159,33 +159,28 @@ class PleromaService(FediverseService):
def upload_file(self, file_data: Union[BinaryIO, bytes], filename: Optional[str] = None) -> FediverseFile:
"""Upload a file to Pleroma instance"""
try:
- # Convert file_data to bytes if it's a stream for MIME detection
+ # Convert file_data to bytes for MIME detection
if hasattr(file_data, 'read'):
# Check if we can seek back
try:
current_pos = file_data.tell()
- header = file_data.read(8)
+ file_bytes = file_data.read()
file_data.seek(current_pos)
- except (io.UnsupportedOperation, OSError):
- # Non-seekable stream, read all data
- remaining_data = file_data.read()
- file_bytes = header + remaining_data
file_data = io.BytesIO(file_bytes)
- header = file_bytes[:8]
+ except (io.UnsupportedOperation, OSError):
+ # Non-seekable stream, already read all data
+ file_data = io.BytesIO(file_bytes)
else:
- header = file_data[:8] if len(file_data) >= 8 else file_data
+ file_bytes = file_data
+ file_data = io.BytesIO(file_bytes)
- # Determine mime type from file header
- if header.startswith(b'\xff\xd8\xff'):
- mime_type = 'image/jpeg'
- elif header.startswith(b'\x89PNG\r\n\x1a\n'):
- mime_type = 'image/png'
- elif header.startswith(b'GIF8'):
- mime_type = 'image/gif'
- elif header.startswith(b'RIFF') and len(header) >= 8 and b'WEBP' in header:
- mime_type = 'image/webp'
+ # Use filetype library for robust MIME detection
+ kind = filetype.guess(file_bytes)
+ if kind is not None:
+ mime_type = kind.mime
else:
- mime_type = 'image/jpeg' # Default fallback
+ # Fallback to image/jpeg if detection fails
+ mime_type = 'image/jpeg'
media = self.client.media_post(file_data, mime_type=mime_type, description=filename)
return self._convert_mastodon_file(media)
diff --git a/requirements.txt b/requirements.txt
index 3d237b4..4a749b2 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -7,3 +7,4 @@ MarkupSafe==3.0.2
Werkzeug==3.1.3
Misskey.py==4.1.0
Mastodon.py==1.8.1
+filetype==1.2.0