From 7d7faa068ea1f5ea23a18a4f3133075c7912d6ea Mon Sep 17 00:00:00 2001
From: Nick Fiacco <nicholas.r.fiacco@gmail.com>
Date: Thu, 4 Apr 2024 10:50:07 -0700
Subject: [PATCH] Use recursive strategy by default for SharePoint (#12557)
Most of the other readers (including OneDrive which is the most similar) use
recursive file reading by default. Do the same thing here.
Also treat empty strings as the value not being present.
---
.../CHANGELOG.md | 4 ++++
.../readers/microsoft_sharepoint/base.py | 14 +++++++-------
.../pyproject.toml | 2 +-
3 files changed, 12 insertions(+), 8 deletions(-)
diff --git a/llama-index-integrations/readers/llama-index-readers-microsoft-sharepoint/CHANGELOG.md b/llama-index-integrations/readers/llama-index-readers-microsoft-sharepoint/CHANGELOG.md
index 9c578a182a..70508c6ff0 100644
--- a/llama-index-integrations/readers/llama-index-readers-microsoft-sharepoint/CHANGELOG.md
+++ b/llama-index-integrations/readers/llama-index-readers-microsoft-sharepoint/CHANGELOG.md
@@ -1,5 +1,9 @@
# CHANGELOG
+## [0.1.7] - 2024-04-03
+
+- Use recursive strategy by default for reading from a folder
+
## [0.1.6] - 2024-04-01
- Allow passing arguments for sitename and folder path during construction of reader
diff --git a/llama-index-integrations/readers/llama-index-readers-microsoft-sharepoint/llama_index/readers/microsoft_sharepoint/base.py b/llama-index-integrations/readers/llama-index-readers-microsoft-sharepoint/llama_index/readers/microsoft_sharepoint/base.py
index 88ab7cee4c..09875f6a0c 100644
--- a/llama-index-integrations/readers/llama-index-readers-microsoft-sharepoint/llama_index/readers/microsoft_sharepoint/base.py
+++ b/llama-index-integrations/readers/llama-index-readers-microsoft-sharepoint/llama_index/readers/microsoft_sharepoint/base.py
@@ -335,7 +335,7 @@ class SharePointReader(BasePydanticReader):
self._drive_id = self._get_drive_id()
- if sharepoint_folder_id is None:
+ if not sharepoint_folder_id:
sharepoint_folder_id = self._get_sharepoint_folder_id(
sharepoint_folder_path
)
@@ -378,7 +378,7 @@ class SharePointReader(BasePydanticReader):
sharepoint_site_name: Optional[str] = None,
sharepoint_folder_path: Optional[str] = None,
sharepoint_folder_id: Optional[str] = None,
- recursive: bool = False,
+ recursive: bool = True,
) -> List[Document]:
"""
Loads the files from the specified folder in the SharePoint site.
@@ -395,20 +395,20 @@ class SharePointReader(BasePydanticReader):
Exception: If an error occurs while accessing SharePoint site.
"""
# If no arguments are provided to load_data, default to the object attributes
- if sharepoint_site_name is None:
+ if not sharepoint_site_name:
sharepoint_site_name = self.sharepoint_site_name
- if sharepoint_folder_path is None:
+ if not sharepoint_folder_path:
sharepoint_folder_path = self.sharepoint_folder_path
- if sharepoint_folder_id is None:
+ if not sharepoint_folder_id:
sharepoint_folder_id = self.sharepoint_folder_id
# TODO: make both of these values optional — and just default to the client ID defaults
- if sharepoint_site_name is None:
+ if not sharepoint_site_name:
raise ValueError("sharepoint_site_name must be provided.")
- if sharepoint_folder_path is None and sharepoint_folder_id is None:
+ if not sharepoint_folder_path and not sharepoint_folder_id:
raise ValueError(
"sharepoint_folder_path or sharepoint_folder_id must be provided."
)
diff --git a/llama-index-integrations/readers/llama-index-readers-microsoft-sharepoint/pyproject.toml b/llama-index-integrations/readers/llama-index-readers-microsoft-sharepoint/pyproject.toml
index 05609b48c3..9372ddfc18 100644
--- a/llama-index-integrations/readers/llama-index-readers-microsoft-sharepoint/pyproject.toml
+++ b/llama-index-integrations/readers/llama-index-readers-microsoft-sharepoint/pyproject.toml
@@ -29,7 +29,7 @@ license = "MIT"
maintainers = ["arun-soliton"]
name = "llama-index-readers-microsoft-sharepoint"
readme = "README.md"
-version = "0.1.6"
+version = "0.1.7"
[tool.poetry.dependencies]
python = ">=3.8.1,<4.0"
--
GitLab