fix: Ignore some emtpy page_content when append to split_documents (#2898)

2 years ago · 696efe494e
parent 4419d357c4
commit 696efe494e
1 changed files with 4 additions and 3 deletions
--- a/api/core/rag/index_processor/processor/paragraph_index_processor.py
+++ b/api/core/rag/index_processor/processor/paragraph_index_processor.py
@ -45,9 +45,10 @@ class ParagraphIndexProcessor(BaseIndexProcessor):
                    # delete Spliter character
                    page_content = document_node.page_content
                    if page_content.startswith(".") or page_content.startswith("。"):
-                        page_content = page_content[1:]
+                        page_content = page_content[1:].strip()
                    else:
                        page_content = page_content
+                    if len(page_content) > 0:
                        document_node.page_content = page_content
                        split_documents.append(document_node)
            all_documents.extend(split_documents)