pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2025-12-08 12:32:30 +08:00
Code Issues Packages Projects Releases Wiki Activity

Feat: add datasource Dropbox (#11488)

Browse Source 
### What problem does this PR solve?

Add datasource Dropbox.

### Type of change

- [x] New Feature (non-breaking change which adds functionality)
This commit is contained in:
Yongteng Lei
2025-11-25 09:40:03 +08:00
committed by GitHub
parent d5f8548200
commit d1744aaaf3
9 changed files with 197 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
rag/svr/sync_data_source.py
View File

@ -37,7 +37,7 @@ from api.db.services.connector_service import ConnectorService, SyncLogsService
from api.db.services.knowledgebase_service import KnowledgebaseService
from common import settings
from common.config_utils import show_configs
from common.data_source import BlobStorageConnector, NotionConnector, DiscordConnector, GoogleDriveConnector, MoodleConnector, JiraConnector
from common.data_source import BlobStorageConnector, NotionConnector, DiscordConnector, GoogleDriveConnector, MoodleConnector, JiraConnector, DropboxConnector
from common.constants import FileSource, TaskStatus
from common.data_source.config import INDEX_BATCH_SIZE
from common.data_source.confluence_connector import ConfluenceConnector
@ -211,6 +211,27 @@ class Gmail(SyncBase):
pass
class Dropbox(SyncBase):
SOURCE_NAME: str = FileSource.DROPBOX
async def _generate(self, task: dict):
self.connector = DropboxConnector(batch_size=self.conf.get("batch_size", INDEX_BATCH_SIZE))
self.connector.load_credentials(self.conf["credentials"])
if task["reindex"] == "1" or not task["poll_range_start"]:
document_generator = self.connector.load_from_state()
begin_info = "totally"
else:
poll_start = task["poll_range_start"]
document_generator = self.connector.poll_source(
poll_start.timestamp(), datetime.now(timezone.utc).timestamp()
)
begin_info = f"from {poll_start}"
logging.info(f"[Dropbox] Connect to Dropbox {begin_info}")
return document_generator
class GoogleDrive(SyncBase):
SOURCE_NAME: str = FileSource.GOOGLE_DRIVE
@ -454,7 +475,8 @@ func_factory = {
FileSource.SHAREPOINT: SharePoint,
FileSource.SLACK: Slack,
FileSource.TEAMS: Teams,
FileSource.MOODLE: Moodle
FileSource.MOODLE: Moodle,
FileSource.DROPBOX: Dropbox,
}