fix: improve dashboard and knowledge base file handling

astrbot/cli/utils/dashboard.py

@@ -15,7 +15,7 @@ class DashboardManager:
         from astrbot.core.utils.io import download_dashboard, get_dashboard_version
 
         if self._bundled_dist.is_dir():
-            click.echo("Dashboard is bundled with the package – skipping download.")
+            click.echo("Dashboard is bundled with the package - skipping download.")
             return
 
         try:
@@ -41,7 +41,10 @@ class DashboardManager:
                             click.echo(f"Failed to install dashboard: {e}")
 
                 case str():
-                    if VersionComparator.compare_version(VERSION, dashboard_version) <= 0:
+                    if (
+                        VersionComparator.compare_version(VERSION, dashboard_version)
+                        <= 0
+                    ):
                         click.echo("Dashboard is already up to date")
                         return
                     try:

astrbot/core/knowledge_base/kb_helper.py

@@ -233,15 +233,12 @@ class KBHelper:
         await self._ensure_vec_db()
         doc_id = str(uuid.uuid4())
         media_paths: list[Path] = []
+        saved_file_path: Path | None = None
         file_size = 0
 
-        # file_path = self.kb_files_dir / f"{doc_id}.{file_type}"
-        # async with aiofiles.open(file_path, "wb") as f:
-        #     await f.write(file_content)
-
         try:
-            chunks_text = []
-            saved_media = []
+            chunks_text: list[str] = []
+            saved_media: list[KBMedia] = []
 
             if pre_chunked_text is not None:
                 # 如果提供了预分块文本,直接使用
@@ -256,6 +253,9 @@ class KBHelper:
                     )
 
                 file_size = len(file_content)
+                saved_file_path = self.kb_files_dir / f"{doc_id}.{file_type}"
+                async with aiofiles.open(saved_file_path, "wb") as f:
+                    await f.write(file_content)
 
                 # 阶段1: 解析文档
                 if progress_callback:
@@ -326,17 +326,15 @@ class KBHelper:
                 doc_name=file_name,
                 file_type=file_type,
                 file_size=file_size,
-                # file_path=str(file_path),
-                file_path="",
+                file_path=str(saved_file_path) if saved_file_path else "",
                 chunk_count=len(chunks_text),
-                media_count=0,
+                media_count=len(saved_media),
             )
             async with self.kb_db.get_db() as session:
                 async with session.begin():
                     session.add(doc)
                     for media in saved_media:
                         session.add(media)
-                    await session.commit()
 
                 await session.refresh(doc)
 
@@ -347,8 +345,14 @@ class KBHelper:
             return doc
         except Exception as e:
             logger.error(f"上传文档失败: {e}")
-            # if file_path.exists():
-            #     file_path.unlink()
+
+            if saved_file_path and saved_file_path.exists():
+                try:
+                    saved_file_path.unlink()
+                except Exception as file_error:
+                    logger.warning(
+                        f"清理原始文档文件失败 {saved_file_path}: {file_error}"
+                    )
 
             for media_path in media_paths:
                 try:
@@ -357,7 +361,7 @@ class KBHelper:
                 except Exception as me:
                     logger.warning(f"清理多媒体文件失败 {media_path}: {me}")
 
-            raise e
+            raise
 
     async def list_documents(
         self,
@@ -412,7 +416,6 @@ class KBHelper:
         async with self.kb_db.get_db() as session:
             async with session.begin():
                 session.add(doc)
-                await session.commit()
             await session.refresh(doc)
 
     async def get_chunks_by_doc_id(

astrbot/core/utils/io.py

@@ -58,7 +58,6 @@ def save_temp_img(img: Image.Image | bytes) -> str:
     timestamp = f"{int(time.time())}_{uuid.uuid4().hex[:8]}"
     p = os.path.join(temp_dir, f"io_temp_img_{timestamp}.jpg")
 
-
     if isinstance(img, Image.Image):
         img.save(p)
     else:
@@ -67,8 +66,6 @@ def save_temp_img(img: Image.Image | bytes) -> str:
     return p
 
 
-
-
 async def download_image_by_url(
     url: str,
     post: bool = False,
@@ -207,7 +204,6 @@ async def download_file(url: str, path: str, show_progress: bool = False) -> Non
         logger.info("下载完成")
 
 
-
 def file_to_base64(file_path: str) -> str:
     with open(file_path, "rb") as f:
         data_bytes = f.read()
@@ -215,7 +211,6 @@ def file_to_base64(file_path: str) -> str:
     return "base64://" + base64_str
 
 
-
 def get_local_ip_addresses() -> list[IPv4Address | IPv6Address]:
     net_interfaces = psutil.net_if_addrs()
     network_ips: list[IPv4Address | IPv6Address] = []
@@ -233,7 +228,6 @@ def get_local_ip_addresses() -> list[IPv4Address | IPv6Address]:
     return network_ips
 
 
-
 async def get_public_ip_address() -> list[IPv4Address | IPv6Address]:
     urls = [
         "https://api64.ipify.org",
@@ -264,7 +258,6 @@ async def get_public_ip_address() -> list[IPv4Address | IPv6Address]:
     return list(found_ips.values())
 
 
-
 async def get_dashboard_version():
     # First check user data directory (manually updated / downloaded dashboard).
     dist_dir = os.path.join(get_astrbot_data_path(), "dist")
@@ -282,7 +275,6 @@ async def get_dashboard_version():
     return None
 
 
-
 async def download_dashboard(
     path: str | None = None,
     extract_path: str = "data",