add doc support in knowledge base for unstructured (#17352)

api/core/rag/extractor/extract_processor.py

@@ -18,6 +18,7 @@ from core.rag.extractor.markdown_extractor import MarkdownExtractor
 from core.rag.extractor.notion_extractor import NotionExtractor
 from core.rag.extractor.pdf_extractor import PdfExtractor
 from core.rag.extractor.text_extractor import TextExtractor
+from core.rag.extractor.unstructured.unstructured_doc_extractor import UnstructuredWordExtractor
 from core.rag.extractor.unstructured.unstructured_eml_extractor import UnstructuredEmailExtractor
 from core.rag.extractor.unstructured.unstructured_epub_extractor import UnstructuredEpubExtractor
 from core.rag.extractor.unstructured.unstructured_markdown_extractor import UnstructuredMarkdownExtractor
@@ -104,7 +105,7 @@ class ExtractProcessor:
                 etl_type = dify_config.ETL_TYPE
                 extractor: Optional[BaseExtractor] = None
                 if etl_type == "Unstructured":
-                    unstructured_api_url = dify_config.UNSTRUCTURED_API_URL
+                    unstructured_api_url = dify_config.UNSTRUCTURED_API_URL or ""
                     unstructured_api_key = dify_config.UNSTRUCTURED_API_KEY or ""
 
                     if file_extension in {".xlsx", ".xls"}:
@@ -121,6 +122,8 @@ class ExtractProcessor:
                         extractor = HtmlExtractor(file_path)
                     elif file_extension == ".docx":
                         extractor = WordExtractor(file_path, upload_file.tenant_id, upload_file.created_by)
+                    elif file_extension == ".doc":
+                        extractor = UnstructuredWordExtractor(file_path, unstructured_api_url, unstructured_api_key)
                     elif file_extension == ".csv":
                         extractor = CSVExtractor(file_path, autodetect_encoding=True)
                     elif file_extension == ".msg":

api/core/rag/extractor/unstructured/unstructured_doc_extractor.py

@@ -10,14 +10,11 @@ logger = logging.getLogger(__name__)
 class UnstructuredWordExtractor(BaseExtractor):
     """Loader that uses unstructured to load word documents."""
 
-    def __init__(
-        self,
-        file_path: str,
-        api_url: str,
-    ):
+    def __init__(self, file_path: str, api_url: str, api_key: str = ""):
         """Initialize with file path."""
         self._file_path = file_path
         self._api_url = api_url
+        self._api_key = api_key
 
     def extract(self) -> list[Document]:
         from unstructured.__version__ import __version__ as __unstructured_version__
@@ -41,9 +38,10 @@ class UnstructuredWordExtractor(BaseExtractor):
             )
 
         if is_doc:
-            from unstructured.partition.doc import partition_doc
+            from unstructured.partition.api import partition_via_api
+
+            elements = partition_via_api(filename=self._file_path, api_url=self._api_url, api_key=self._api_key)
 
-            elements = partition_doc(filename=self._file_path)
         else:
             from unstructured.partition.docx import partition_docx