feat: 从 URL 创建知识时自动提取 HTML 标题

internal/application/service/knowledge.go

@@ -833,6 +833,21 @@ func (s *knowledgeService) processDocumentFromURL(ctx context.Context,
 		return
 	}
 
+	// Update knowledge title if extracted from HTML
+	logger.GetLogger(ctx).Infof("DEBUG: resp.GetTitle()='%s', knowledge.Title='%s'", resp.GetTitle(), knowledge.Title)
+	if resp.GetTitle() != "" && knowledge.Title == "" {
+		knowledge.Title = resp.GetTitle()
+		logger.GetLogger(ctx).Infof("Extracted title from URL: %s", knowledge.Title)
+		// Update database immediately
+		if err := s.repo.UpdateKnowledge(ctx, knowledge); err != nil {
+			logger.GetLogger(ctx).WithField("error", err).Errorf("Failed to update knowledge title")
+		} else {
+			logger.GetLogger(ctx).Infof("DEBUG: Successfully updated knowledge title in database")
+		}
+	} else {
+		logger.GetLogger(ctx).Infof("DEBUG: Skipping title update - condition not met")
+	}
+
 	// Process and store chunks
 	s.processChunks(ctx, kb, knowledge, resp.Chunks)
 }

services/docreader/src/parser/base_parser.py

@@ -58,6 +58,7 @@ class ParseResult:
 
     text: str  # Extracted text content
     chunks: Optional[List[Chunk]] = None  # Chunk results
+    title: Optional[str] = None  # Document title (for URLs)
 
 
 class BaseParser(ABC):
@@ -672,7 +673,10 @@ def parse(self, content: bytes) -> ParseResult:
                     f"Skipping image processing for unsupported file type: {file_ext}"
                 )
 
-        return ParseResult(text=text, chunks=chunks)
+        # Include title if available (for WebParser)
+        title = getattr(self, 'extracted_title', None)
+        logger.info(f"DEBUG: Creating ParseResult with title='{title}'")
+        return ParseResult(text=text, chunks=chunks, title=title)
 
     def _split_into_units(self, text: str) -> List[str]:
         """

services/docreader/src/parser/web_parser.py

@@ -15,6 +15,7 @@ class WebParser(BaseParser):
 
     def __init__(self, title: str, **kwargs):
         self.title = title
+        self.extracted_title = None  # Store extracted title from HTML
         self.proxy = os.environ.get("WEB_PROXY", "")
         super().__init__(file_name=title, **kwargs)
         logger.info(f"Initialized WebParser with title: {title}")
@@ -110,6 +111,9 @@ def parse_into_text(self, content: bytes) -> Union[str, Tuple[str, Dict[str, Any
                     logger.info("No title found, using default")
 
             logger.info(f"Web page title: {title}")
+            # Store the extracted title for later use
+            self.extracted_title = title.strip() if title else ""
+            logger.info(f"DEBUG: Stored extracted_title='{self.extracted_title}'")
             text = "\n".join(
                 (line.strip() for line in text.splitlines() if line.strip())
             )

services/docreader/src/proto/docreader.proto

@@ -86,4 +86,5 @@ message Chunk {
 message ReadResponse {
   repeated Chunk chunks = 1; // 文档分块
   string error = 2;          // 错误信息
+  string title = 3;          // 文档标题 (仅从URL读取时返回)
 } 

services/docreader/src/server/server.py

@@ -262,13 +262,16 @@ def ReadFromURL(self, request, context):
                     context.set_details(error_msg)
                     return ReadResponse(error=error_msg)
 
-                # Convert to protobuf message, including image info
+                # Convert to protobuf message, including image info and title
                 logger.info(
-                    f"Successfully parsed URL {request.url}, returning {len(result.chunks)} chunks"
+                    f"Successfully parsed URL {request.url}, returning {len(result.chunks)} chunks, title: {result.title}"
                 )
 
+                response_title = result.title if result.title else ""
+                logger.info(f"DEBUG: Building ReadResponse with title='{response_title}'")
                 response = ReadResponse(
-                    chunks=[self._convert_chunk_to_proto(chunk) for chunk in result.chunks]
+                    chunks=[self._convert_chunk_to_proto(chunk) for chunk in result.chunks],
+                    title=response_title
                 )
                 logger.info(f"Response size: {response.ByteSize()} bytes")
                 return response