Fix: webdav connector (#12380)

### What problem does this PR solve?

fix webdav #11422

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

common/data_source/__init__.py

@@ -34,7 +34,6 @@ from .google_drive.connector import GoogleDriveConnector
 from .jira.connector import JiraConnector
 from .sharepoint_connector import SharePointConnector
 from .teams_connector import TeamsConnector
-from .webdav_connector import WebDAVConnector
 from .moodle_connector import MoodleConnector
 from .airtable_connector import AirtableConnector
 from .asana_connector import AsanaConnector
@@ -62,7 +61,6 @@ __all__ = [
     "JiraConnector",
     "SharePointConnector",
     "TeamsConnector",
-    "WebDAVConnector",
     "MoodleConnector",
     "BlobType",
     "DocumentSource",

common/data_source/webdav_connector.py

@@ -82,10 +82,6 @@ class WebDAVConnector(LoadConnector, PollConnector):
                 base_url=self.base_url,
                 auth=(username, password)
             )
-            
-            # Test connection
-            self.client.exists(self.remote_path)
-            
         except Exception as e:
             logging.error(f"Failed to connect to WebDAV server: {e}")
             raise ConnectorMissingCredentialError(
@@ -308,60 +304,79 @@ class WebDAVConnector(LoadConnector, PollConnector):
             yield batch
 
     def validate_connector_settings(self) -> None:
-        """Validate WebDAV connector settings
-        
-        Raises:
-            ConnectorMissingCredentialError: If credentials are not loaded
-            ConnectorValidationError: If settings are invalid
+        """Validate WebDAV connector settings.
+
+        Validation should exercise the same code-paths used by the connector
+        (directory listing / PROPFIND), avoiding exists() which may probe with
+        methods that differ across servers.
         """
         if self.client is None:
-            raise ConnectorMissingCredentialError(
-                "WebDAV credentials not loaded."
-            )
+            raise ConnectorMissingCredentialError("WebDAV credentials not loaded.")
 
         if not self.base_url:
-            raise ConnectorValidationError(
-                "No base URL was provided in connector settings."
-            )
+            raise ConnectorValidationError("No base URL was provided in connector settings.")
+
+        # Normalize directory path: for collections, many servers behave better with trailing '/'
+        test_path = self.remote_path or "/"
+        if not test_path.startswith("/"):
+            test_path = f"/{test_path}"
+        if test_path != "/" and not test_path.endswith("/"):
+            test_path = f"{test_path}/"
 
         try:
-            if not self.client.exists(self.remote_path):
-                raise ConnectorValidationError(
-                    f"Remote path '{self.remote_path}' does not exist on WebDAV server."
-                )
+            # Use the same behavior as real sync: list directory with details (PROPFIND)
+            self.client.ls(test_path, detail=True)
 
         except Exception as e:
-            error_message = str(e)
-            
-            if "401" in error_message or "unauthorized" in error_message.lower():
-                raise CredentialExpiredError(
-                    "WebDAV credentials appear invalid or expired."
-                )
-            
-            if "403" in error_message or "forbidden" in error_message.lower():
+            # Prefer structured status codes if present on the exception/response
+            status = None
+            for attr in ("status_code", "code"):
+                v = getattr(e, attr, None)
+                if isinstance(v, int):
+                    status = v
+                    break
+            if status is None:
+                resp = getattr(e, "response", None)
+                v = getattr(resp, "status_code", None)
+                if isinstance(v, int):
+                    status = v
+
+            # If we can classify by status code, do it
+            if status == 401:
+                raise CredentialExpiredError("WebDAV credentials appear invalid or expired.")
+            if status == 403:
                 raise InsufficientPermissionsError(
                     f"Insufficient permissions to access path '{self.remote_path}' on WebDAV server."
                 )
-            
-            if "404" in error_message or "not found" in error_message.lower():
+            if status == 404:
                 raise ConnectorValidationError(
                     f"Remote path '{self.remote_path}' does not exist on WebDAV server."
                 )
 
+            # Fallback: avoid brittle substring matching that caused false positives.
+            # Provide the original exception for diagnosis.
             raise ConnectorValidationError(
-                f"Unexpected WebDAV client error: {e}"
+                f"WebDAV validation failed for path '{test_path}': {repr(e)}"
             )
 
 
+
 if __name__ == "__main__":
     credentials_dict = {
         "username": os.environ.get("WEBDAV_USERNAME"),
         "password": os.environ.get("WEBDAV_PASSWORD"),
     }
 
+    credentials_dict = {
+        "username": "user",
+        "password": "pass",
+    }
+
+
+
     connector = WebDAVConnector(
-        base_url=os.environ.get("WEBDAV_URL") or "https://webdav.example.com",
-        remote_path=os.environ.get("WEBDAV_PATH") or "/",
+        base_url="http://172.17.0.1:8080/",
+        remote_path="/",
     )
 
     try:

rag/svr/sync_data_source.py

@@ -46,7 +46,6 @@ from common.data_source import (
     MoodleConnector,
     JiraConnector,
     DropboxConnector,
-    WebDAVConnector,
     AirtableConnector,
     AsanaConnector,
     ImapConnector,
@@ -55,6 +54,7 @@ from common.data_source import (
 from common.constants import FileSource, TaskStatus
 from common.data_source.config import INDEX_BATCH_SIZE
 from common.data_source.models import ConnectorFailure
+from common.data_source.webdav_connector import WebDAVConnector
 from common.data_source.confluence_connector import ConfluenceConnector
 from common.data_source.gmail_connector import GmailConnector
 from common.data_source.box_connector import BoxConnector
@@ -696,7 +696,12 @@ class WebDAV(SyncBase):
             self.conf.get("remote_path", "/"),
             begin_info
         ))
-        return document_batch_generator
+
+        async def async_wrapper():
+            for document_batch in document_batch_generator:
+                yield document_batch
+                
+        return async_wrapper()
 
 
 class Moodle(SyncBase):