unstructured-ingest 0.0.2__py3-none-any.whl → 0.0.2.dev0__py3-none-any.whl

unstructured_ingest/v2/processes/connectors/elasticsearch.py

@@ -7,12 +7,10 @@ from pathlib import Path
 from time import time
 from typing import TYPE_CHECKING, Any, Generator, Optional
 
+from unstructured.documents.elements import DataSourceMetadata
+
 from unstructured_ingest.enhanced_dataclass import EnhancedDataClassJsonMixin, enhanced_field
-from unstructured_ingest.error import (
-    DestinationConnectionError,
-    SourceConnectionError,
-    SourceConnectionNetworkError,
-)
+from unstructured_ingest.error import SourceConnectionError, SourceConnectionNetworkError
 from unstructured_ingest.utils.data_prep import flatten_dict, generator_batching_wbytes
 from unstructured_ingest.utils.dep_check import requires_dependencies
 from unstructured_ingest.v2.interfaces import (
@@ -22,7 +20,6 @@ from unstructured_ingest.v2.interfaces import (
     DownloaderConfig,
     DownloadResponse,
     FileData,
-    FileDataSourceMetadata,
     Indexer,
     IndexerConfig,
     UploadContent,
@@ -124,14 +121,11 @@ class ElasticsearchIndexerConfig(IndexerConfig):
 class ElasticsearchIndexer(Indexer):
     connection_config: ElasticsearchConnectionConfig
     index_config: ElasticsearchIndexerConfig
+    client: "ElasticsearchClient" = field(init=False)
     connector_type: str = CONNECTOR_TYPE
 
-    def precheck(self) -> None:
-        try:
-            self.connection_config.get_client()
-        except Exception as e:
-            logger.error(f"failed to validate connection: {e}", exc_info=True)
-            raise SourceConnectionError(f"failed to validate connection: {e}")
+    def __post_init__(self):
+        self.client = self.connection_config.get_client()
 
     @requires_dependencies(["elasticsearch"], extras="elasticsearch")
     def load_scan(self):
@@ -144,9 +138,8 @@ class ElasticsearchIndexer(Indexer):
         scan = self.load_scan()
 
         scan_query: dict = {"stored_fields": [], "query": {"match_all": {}}}
-        client = self.connection_config.get_client()
         hits = scan(
-            client,
+            self.client,
             query=scan_query,
             scroll="1m",
             index=self.index_config.index_name,
@@ -175,7 +168,7 @@ class ElasticsearchIndexer(Indexer):
             yield FileData(
                 identifier=identified,
                 connector_type=CONNECTOR_TYPE,
-                metadata=FileDataSourceMetadata(
+                metadata=DataSourceMetadata(
                     url=f"{self.connection_config.hosts[0]}/{self.index_config.index_name}",
                     date_processed=str(time()),
                 ),
@@ -241,7 +234,7 @@ class ElasticsearchDownloader(Downloader):
             file_data=FileData(
                 identifier=filename_id,
                 connector_type=CONNECTOR_TYPE,
-                metadata=FileDataSourceMetadata(
+                metadata=DataSourceMetadata(
                     version=str(result["_version"]) if "_version" in result else None,
                     date_processed=str(time()),
                     record_locator={
@@ -346,13 +339,6 @@ class ElasticsearchUploader(Uploader):
     upload_config: ElasticsearchUploaderConfig
     connection_config: ElasticsearchConnectionConfig
 
-    def precheck(self) -> None:
-        try:
-            self.connection_config.get_client()
-        except Exception as e:
-            logger.error(f"failed to validate connection: {e}", exc_info=True)
-            raise DestinationConnectionError(f"failed to validate connection: {e}")
-
     @requires_dependencies(["elasticsearch"], extras="elasticsearch")
     def load_parallel_bulk(self):
         from elasticsearch.helpers import parallel_bulk

unstructured_ingest/v2/processes/connectors/fsspec/fsspec.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 import contextlib
+import fnmatch
 from dataclasses import dataclass, field
 from datetime import datetime
 from pathlib import Path
@@ -8,6 +9,8 @@ from time import time
 from typing import TYPE_CHECKING, Any, Generator, Optional, TypeVar
 from uuid import NAMESPACE_DNS, uuid5
 
+from unstructured.documents.elements import DataSourceMetadata
+
 from unstructured_ingest.enhanced_dataclass import enhanced_field
 from unstructured_ingest.error import SourceConnectionError, SourceConnectionNetworkError
 from unstructured_ingest.v2.interfaces import (
@@ -17,7 +20,6 @@ from unstructured_ingest.v2.interfaces import (
     DownloaderConfig,
     DownloadResponse,
     FileData,
-    FileDataSourceMetadata,
     Indexer,
     IndexerConfig,
     SourceIdentifiers,
@@ -72,6 +74,7 @@ class FileConfig(Base):
 @dataclass
 class FsspecIndexerConfig(FileConfig, IndexerConfig):
     recursive: bool = False
+    file_glob: Optional[list[str]] = None
 
 
 @dataclass
@@ -106,7 +109,17 @@ class FsspecIndexer(Indexer):
             **self.connection_config.get_access_config(),
         )
 
-    def precheck(self) -> None:
+    def does_path_match_glob(self, path: str) -> bool:
+        if self.index_config.file_glob is None:
+            return True
+        patterns = self.index_config.file_glob
+        for pattern in patterns:
+            if fnmatch.filter([path], pattern):
+                return True
+        logger.debug(f"The file {path!r} is discarded as it does not match any given glob.")
+        return False
+
+    def check_connection(self):
         from fsspec import get_filesystem_class
 
         try:
@@ -144,10 +157,10 @@ class FsspecIndexer(Indexer):
             else:
                 raise TypeError(f"unhandled response type from find: {type(found)}")
 
-    def get_metadata(self, path: str) -> FileDataSourceMetadata:
+    def get_metadata(self, path: str) -> DataSourceMetadata:
         date_created = None
         date_modified = None
-        file_size = None
+
         try:
             created: Optional[Any] = self.fs.created(path)
             if created:
@@ -167,8 +180,6 @@ class FsspecIndexer(Indexer):
                     date_modified = str(modified)
         except NotImplementedError:
             pass
-        with contextlib.suppress(AttributeError):
-            file_size = self.fs.size(path)
 
         version = self.fs.checksum(path)
         metadata: dict[str, str] = {}
@@ -178,19 +189,15 @@ class FsspecIndexer(Indexer):
             "protocol": self.index_config.protocol,
             "remote_file_path": self.index_config.remote_url,
         }
-        file_stat = self.fs.stat(path=path)
-        if file_id := file_stat.get("id"):
-            record_locator["file_id"] = file_id
         if metadata:
             record_locator["metadata"] = metadata
-        return FileDataSourceMetadata(
+        return DataSourceMetadata(
             date_created=date_created,
             date_modified=date_modified,
             date_processed=str(time()),
             version=str(version),
             url=f"{self.index_config.protocol}://{path}",
             record_locator=record_locator,
-            filesize_bytes=file_size,
         )
 
     def sterilize_info(self, path) -> dict:
@@ -198,7 +205,8 @@ class FsspecIndexer(Indexer):
         return sterilize_dict(data=info)
 
     def run(self, **kwargs: Any) -> Generator[FileData, None, None]:
-        files = self.list_files()
+        raw_files = self.list_files()
+        files = [f for f in raw_files if self.does_path_match_glob(f)]
         for file in files:
             # Note: we remove any remaining leading slashes (Box introduces these)
             # to get a valid relative path
@@ -247,6 +255,13 @@ class FsspecDownloader(Downloader):
             **self.connection_config.get_access_config(),
         )
 
+    def get_download_path(self, file_data: FileData) -> Path:
+        return (
+            self.download_dir / Path(file_data.source_identifiers.relative_path)
+            if self.download_config
+            else Path(file_data.source_identifiers.rel_path)
+        )
+
     def run(self, file_data: FileData, **kwargs: Any) -> DownloadResponse:
         download_path = self.get_download_path(file_data=file_data)
         download_path.parent.mkdir(parents=True, exist_ok=True)

unstructured_ingest/v2/processes/connectors/fsspec/s3.py

@@ -5,15 +5,11 @@ from pathlib import Path
 from time import time
 from typing import Any, Generator, Optional
 
-from unstructured.utils import requires_dependencies
+from unstructured.documents.elements import DataSourceMetadata
 
 from unstructured_ingest.enhanced_dataclass import enhanced_field
-from unstructured_ingest.v2.interfaces import (
-    DownloadResponse,
-    FileData,
-    FileDataSourceMetadata,
-    UploadContent,
-)
+from unstructured_ingest.utils.dep_check import requires_dependencies
+from unstructured_ingest.v2.interfaces import DownloadResponse, FileData, UploadContent
 from unstructured_ingest.v2.processes.connector_registry import (
     DestinationRegistryEntry,
     SourceRegistryEntry,
@@ -70,10 +66,9 @@ class S3Indexer(FsspecIndexer):
     index_config: S3IndexerConfig
     connector_type: str = CONNECTOR_TYPE
 
-    def get_metadata(self, path: str) -> FileDataSourceMetadata:
+    def get_metadata(self, path: str) -> DataSourceMetadata:
         date_created = None
         date_modified = None
-        file_size = None
         try:
             modified: Optional[datetime] = self.fs.modified(path)
             if modified:
@@ -81,8 +76,6 @@ class S3Indexer(FsspecIndexer):
                 date_modified = str(modified.timestamp())
         except NotImplementedError:
             pass
-        with contextlib.suppress(AttributeError):
-            file_size = self.fs.size(path)
 
         version = None
         info: dict[str, Any] = self.fs.info(path)
@@ -97,14 +90,13 @@ class S3Indexer(FsspecIndexer):
         }
         if metadata:
             record_locator["metadata"] = metadata
-        return FileDataSourceMetadata(
+        return DataSourceMetadata(
             date_created=date_created,
             date_modified=date_modified,
             date_processed=str(time()),
             version=version,
             url=f"{self.index_config.protocol}://{path}",
             record_locator=record_locator,
-            filesize_bytes=file_size,
         )
 
     @requires_dependencies(["s3fs", "fsspec"], extras="s3")

unstructured_ingest/v2/processes/connectors/google_drive.py

@@ -1,16 +1,15 @@
 import io
 import os
 from dataclasses import dataclass, field
+from pathlib import Path
 from typing import TYPE_CHECKING, Any, Generator, Optional, Union
 
 from dateutil import parser
+from unstructured.documents.elements import DataSourceMetadata
 from unstructured.file_utils.google_filetype import GOOGLE_DRIVE_EXPORT_TYPES
 
 from unstructured_ingest.enhanced_dataclass import enhanced_field
-from unstructured_ingest.error import (
-    SourceConnectionError,
-    SourceConnectionNetworkError,
-)
+from unstructured_ingest.error import SourceConnectionNetworkError
 from unstructured_ingest.utils.dep_check import requires_dependencies
 from unstructured_ingest.utils.string_and_date_utils import json_to_dict
 from unstructured_ingest.v2.interfaces import (
@@ -19,7 +18,6 @@ from unstructured_ingest.v2.interfaces import (
     Downloader,
     DownloaderConfig,
     FileData,
-    FileDataSourceMetadata,
     Indexer,
     IndexerConfig,
     SourceIdentifiers,
@@ -123,13 +121,6 @@ class GoogleDriveIndexer(Indexer):
         ]
     )
 
-    def precheck(self) -> None:
-        try:
-            self.connection_config.get_files_service()
-        except Exception as e:
-            logger.error(f"failed to validate connection: {e}", exc_info=True)
-            raise SourceConnectionError(f"failed to validate connection: {e}")
-
     @staticmethod
     def is_dir(record: dict) -> bool:
         return record.get("mimeType") == "application/vnd.google-apps.folder"
@@ -164,7 +155,7 @@ class GoogleDriveIndexer(Indexer):
             connector_type=CONNECTOR_TYPE,
             identifier=file_id,
             source_identifiers=source_identifiers,
-            metadata=FileDataSourceMetadata(
+            metadata=DataSourceMetadata(
                 url=url,
                 version=version,
                 date_created=str(date_created_dt.timestamp()),
@@ -281,6 +272,11 @@ class GoogleDriveDownloader(Downloader):
     )
     connector_type: str = CONNECTOR_TYPE
 
+    def get_download_path(self, file_data: FileData) -> Path:
+        rel_path = file_data.source_identifiers.relative_path
+        rel_path = rel_path[1:] if rel_path.startswith("/") else rel_path
+        return self.download_dir / Path(rel_path)
+
     @SourceConnectionNetworkError.wrap
     def _get_content(self, downloader: "MediaIoBaseDownload") -> bool:
         downloaded = False

unstructured_ingest/v2/processes/connectors/local.py

@@ -1,9 +1,12 @@
 import glob
+import itertools
 import shutil
 from dataclasses import dataclass, field
 from pathlib import Path
 from time import time
-from typing import Any, Generator
+from typing import Any, Generator, Optional
+
+from unstructured.documents.elements import DataSourceMetadata
 
 from unstructured_ingest.v2.interfaces import (
     AccessConfig,
@@ -12,7 +15,6 @@ from unstructured_ingest.v2.interfaces import (
     DownloaderConfig,
     DownloadResponse,
     FileData,
-    FileDataSourceMetadata,
     Indexer,
     IndexerConfig,
     SourceIdentifiers,
@@ -43,6 +45,7 @@ class LocalConnectionConfig(ConnectionConfig):
 class LocalIndexerConfig(IndexerConfig):
     input_path: str
     recursive: bool = False
+    file_glob: Optional[list[str]] = None
 
     @property
     def path(self) -> Path:
@@ -61,11 +64,16 @@ class LocalIndexer(Indexer):
         input_path = self.index_config.path
         if input_path.is_file():
             return [Path(s) for s in glob.glob(f"{self.index_config.path}")]
-        if self.index_config.recursive:
-            return list(input_path.rglob("*"))
-        return list(input_path.glob("*"))
+        glob_fn = input_path.rglob if self.index_config.recursive else input_path.glob
+        if not self.index_config.file_glob:
+            return list(glob_fn("*"))
+        return list(
+            itertools.chain.from_iterable(
+                glob_fn(pattern) for pattern in self.index_config.file_glob
+            )
+        )
 
-    def get_file_metadata(self, path: Path) -> FileDataSourceMetadata:
+    def get_file_metadata(self, path: Path) -> DataSourceMetadata:
         stats = path.stat()
         try:
             date_modified = str(stats.st_mtime)
@@ -85,20 +93,12 @@ class LocalIndexer(Indexer):
         except Exception as e:
             logger.warning(f"Couldn't detect file mode: {e}")
             permissions_data = None
-
-        try:
-            filesize_bytes = stats.st_size
-        except Exception as e:
-            logger.warning(f"Couldn't detect file size: {e}")
-            filesize_bytes = None
-
-        return FileDataSourceMetadata(
+        return DataSourceMetadata(
             date_modified=date_modified,
             date_created=date_created,
             date_processed=str(time()),
             permissions_data=permissions_data,
             record_locator={"path": str(path.resolve())},
-            filesize_bytes=filesize_bytes,
         )
 
     def run(self, **kwargs: Any) -> Generator[FileData, None, None]:

unstructured_ingest/v2/processes/connectors/mongodb.py

@@ -6,7 +6,6 @@ from typing import TYPE_CHECKING, Any, Optional
 from unstructured.__version__ import __version__ as unstructured_version
 
 from unstructured_ingest.enhanced_dataclass import enhanced_field
-from unstructured_ingest.error import DestinationConnectionError
 from unstructured_ingest.utils.data_prep import batch_generator
 from unstructured_ingest.utils.dep_check import requires_dependencies
 from unstructured_ingest.v2.interfaces import (
@@ -86,15 +85,11 @@ class MongoDBUploaderConfig(UploaderConfig):
 class MongoDBUploader(Uploader):
     upload_config: MongoDBUploaderConfig
     connection_config: MongoDBConnectionConfig
+    client: Optional["MongoClient"] = field(init=False)
     connector_type: str = CONNECTOR_TYPE
 
-    def precheck(self) -> None:
-        try:
-            client = self.create_client()
-            client.admin.command("ping")
-        except Exception as e:
-            logger.error(f"failed to validate connection: {e}", exc_info=True)
-            raise DestinationConnectionError(f"failed to validate connection: {e}")
+    def __post_init__(self):
+        self.client = self.create_client()
 
     @requires_dependencies(["pymongo"], extras="mongodb")
     def create_client(self) -> "MongoClient":
@@ -128,8 +123,7 @@ class MongoDBUploader(Uploader):
             f"collection {self.connection_config.collection} "
             f"at {self.connection_config.host}",
         )
-        client = self.create_client()
-        db = client[self.connection_config.database]
+        db = self.client[self.connection_config.database]
         collection = db[self.connection_config.collection]
         for chunk in batch_generator(elements_dict, self.upload_config.batch_size):
             collection.insert_many(chunk)

unstructured_ingest/v2/processes/connectors/onedrive.py

@@ -5,6 +5,7 @@ from time import time
 from typing import TYPE_CHECKING, Any, Generator, Optional
 
 from dateutil import parser
+from unstructured.documents.elements import DataSourceMetadata
 
 from unstructured_ingest.enhanced_dataclass import enhanced_field
 from unstructured_ingest.error import SourceConnectionError, SourceConnectionNetworkError
@@ -16,7 +17,6 @@ from unstructured_ingest.v2.interfaces import (
     DownloaderConfig,
     DownloadResponse,
     FileData,
-    FileDataSourceMetadata,
     Indexer,
     IndexerConfig,
     SourceIdentifiers,
@@ -87,18 +87,6 @@ class OnedriveIndexer(Indexer):
     connection_config: OnedriveConnectionConfig
     index_config: OnedriveIndexerConfig
 
-    def precheck(self) -> None:
-        try:
-            token_resp: dict = self.connection_config.get_token()
-            if error := token_resp.get("error"):
-                raise SourceConnectionError(
-                    "{} ({})".format(error, token_resp.get("error_description"))
-                )
-            self.connection_config.get_client()
-        except Exception as e:
-            logger.error(f"failed to validate connection: {e}", exc_info=True)
-            raise SourceConnectionError(f"failed to validate connection: {e}")
-
     def list_objects(self, folder, recursive) -> list["DriveItem"]:
         drive_items = folder.children.get().execute_query()
         files = [d for d in drive_items if d.is_file]
@@ -148,7 +136,7 @@ class OnedriveIndexer(Indexer):
             source_identifiers=SourceIdentifiers(
                 fullpath=server_path, filename=drive_item.name, rel_path=rel_path
             ),
-            metadata=FileDataSourceMetadata(
+            metadata=DataSourceMetadata(
                 url=drive_item.parent_reference.path + "/" + drive_item.name,
                 version=drive_item.etag,
                 date_modified=str(date_modified_dt.timestamp()) if date_modified_dt else None,

unstructured_ingest/v2/processes/connectors/pinecone.py

@@ -123,12 +123,9 @@ class PineconeUploader(Uploader):
     connection_config: PineconeConnectionConfig
     connector_type: str = CONNECTOR_TYPE
 
-    def precheck(self):
-        try:
-            self.connection_config.get_index()
-        except Exception as e:
-            logger.error(f"failed to validate connection: {e}", exc_info=True)
-            raise DestinationConnectionError(f"failed to validate connection: {e}")
+    @DestinationConnectionError.wrap
+    def check_connection(self):
+        _ = self.connection_config.get_index()
 
     @requires_dependencies(["pinecone"], extras="pinecone")
     def upsert_batch(self, batch):

unstructured_ingest/v2/processes/connectors/salesforce.py

@@ -18,9 +18,10 @@ from textwrap import dedent
 from typing import TYPE_CHECKING, Any, Generator, Type
 
 from dateutil import parser
+from unstructured.documents.elements import DataSourceMetadata
 
 from unstructured_ingest.enhanced_dataclass import enhanced_field
-from unstructured_ingest.error import SourceConnectionError, SourceConnectionNetworkError
+from unstructured_ingest.error import SourceConnectionNetworkError
 from unstructured_ingest.utils.dep_check import requires_dependencies
 from unstructured_ingest.v2.interfaces import (
     AccessConfig,
@@ -29,7 +30,6 @@ from unstructured_ingest.v2.interfaces import (
     DownloaderConfig,
     DownloadResponse,
     FileData,
-    FileDataSourceMetadata,
     Indexer,
     IndexerConfig,
     SourceIdentifiers,
@@ -132,13 +132,6 @@ class SalesforceIndexer(Indexer):
             if record_type not in ACCEPTED_CATEGORIES:
                 raise ValueError(f"{record_type} not currently an accepted Salesforce category")
 
-    def precheck(self) -> None:
-        try:
-            self.connection_config.get_client()
-        except Exception as e:
-            logger.error(f"failed to validate connection: {e}", exc_info=True)
-            raise SourceConnectionError(f"failed to validate connection: {e}")
-
     def get_file_extension(self, record_type) -> str:
         if record_type == "EmailMessage":
             extension = ".eml"
@@ -179,7 +172,7 @@ class SalesforceIndexer(Indexer):
                                 filename=record_with_extension,
                                 fullpath=f"{record['attributes']['type']}/{record_with_extension}",
                             ),
-                            metadata=FileDataSourceMetadata(
+                            metadata=DataSourceMetadata(
                                 url=record["attributes"]["url"],
                                 version=str(parser.parse(record["SystemModstamp"]).timestamp()),
                                 date_created=str(parser.parse(record["CreatedDate"]).timestamp()),
@@ -214,6 +207,11 @@ class SalesforceDownloader(Downloader):
     )
     connector_type: str = CONNECTOR_TYPE
 
+    def get_download_path(self, file_data: FileData) -> Path:
+        rel_path = file_data.source_identifiers.relative_path
+        rel_path = rel_path[1:] if rel_path.startswith("/") else rel_path
+        return self.download_dir / Path(rel_path)
+
     def _xml_for_record(self, record: OrderedDict) -> str:
         """Creates partitionable xml file from a record"""
         import xml.etree.ElementTree as ET

unstructured_ingest/v2/processes/connectors/sharepoint.py

@@ -6,8 +6,10 @@ from time import time
 from typing import TYPE_CHECKING, Any, Generator, Optional
 from urllib.parse import quote
 
+from unstructured.documents.elements import DataSourceMetadata
+
 from unstructured_ingest.enhanced_dataclass import EnhancedDataClassJsonMixin, enhanced_field
-from unstructured_ingest.error import SourceConnectionError, SourceConnectionNetworkError
+from unstructured_ingest.error import SourceConnectionNetworkError
 from unstructured_ingest.utils.dep_check import requires_dependencies
 from unstructured_ingest.v2.interfaces import (
     AccessConfig,
@@ -16,7 +18,6 @@ from unstructured_ingest.v2.interfaces import (
     DownloaderConfig,
     DownloadResponse,
     FileData,
-    FileDataSourceMetadata,
     Indexer,
     IndexerConfig,
     SourceIdentifiers,
@@ -133,14 +134,6 @@ class SharepointIndexer(Indexer):
     connection_config: SharepointConnectionConfig
     index_config: SharepointIndexerConfig = field(default_factory=lambda: SharepointIndexerConfig())
 
-    def precheck(self) -> None:
-        try:
-            site_client = self.connection_config.get_client()
-            site_client.site_pages.pages.get().execute_query()
-        except Exception as e:
-            logger.error(f"failed to validate connection: {e}", exc_info=True)
-            raise SourceConnectionError(f"failed to validate connection: {e}")
-
     def list_files(self, folder: "Folder", recursive: bool = False) -> list["File"]:
         if not recursive:
             folder.expand(["Files"]).get().execute_query()
@@ -194,7 +187,7 @@ class SharepointIndexer(Indexer):
                 fullpath=file_path,
                 rel_path=file_path.replace(self.index_config.path, ""),
             ),
-            metadata=FileDataSourceMetadata(
+            metadata=DataSourceMetadata(
                 url=url,
                 version=version,
                 date_modified=str(date_modified_dt.timestamp()) if date_modified_dt else None,
@@ -229,7 +222,7 @@ class SharepointIndexer(Indexer):
                 fullpath=fullpath,
                 rel_path=rel_path,
             ),
-            metadata=FileDataSourceMetadata(
+            metadata=DataSourceMetadata(
                 url=absolute_url,
                 version=f"{file.major_version}.{file.minor_version}",
                 date_modified=str(date_modified_dt.timestamp()) if date_modified_dt else None,
@@ -347,9 +340,10 @@ class SharepointDownloader(Downloader):
     connector_type: str = CONNECTOR_TYPE
 
     def get_download_path(self, file_data: FileData) -> Path:
-        download_path = super().get_download_path(file_data=file_data)
-
         content_type = file_data.additional_metadata.get("sharepoint_content_type")
+        rel_path = file_data.source_identifiers.fullpath
+        rel_path = rel_path[1:] if rel_path.startswith("/") else rel_path
+        download_path = self.download_dir / Path(rel_path)
         if content_type == SharepointContentType.SITEPAGE.value:
             # Update output extension to html if site page
             download_path = download_path.with_suffix(".html")