pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2025-12-08 20:42:30 +08:00
Code Issues Packages Projects Releases Wiki Activity

Introduced task priority (#6118)

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

Introduced task priority

### Type of change

- [x] New Feature (non-breaking change which adds functionality)
This commit is contained in:
Zhichang Yu
2025-03-14 23:43:46 +08:00
committed by GitHub
parent 1842ca0334
commit 89a69eed72
9 changed files with 59 additions and 42 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
api/db/services/task_service.py
View File

@ -28,7 +28,7 @@ from api.db.services.common_service import CommonService
from api.db.services.document_service import DocumentService
from api.utils import current_timestamp, get_uuid
from deepdoc.parser.excel_parser import RAGFlowExcelParser
from rag.settings import SVR_QUEUE_NAME
from rag.settings import get_svr_queue_name
from rag.utils.storage_factory import STORAGE_IMPL
from rag.utils.redis_conn import REDIS_CONN
from api import settings
@ -289,7 +289,7 @@ class TaskService(CommonService):
).execute()
def queue_tasks(doc: dict, bucket: str, name: str):
def queue_tasks(doc: dict, bucket: str, name: str, priority: int):
"""Create and queue document processing tasks.
This function creates processing tasks for a document based on its type and configuration.
@ -301,6 +301,7 @@ def queue_tasks(doc: dict, bucket: str, name: str):
doc (dict): Document dictionary containing metadata and configuration.
bucket (str): Storage bucket name where the document is stored.
name (str): File name of the document.
priority (int, optional): Priority level for task queueing (default is 0).
Note:
- For PDF documents, tasks are created per page range based on configuration
@ -358,6 +359,7 @@ def queue_tasks(doc: dict, bucket: str, name: str):
task_digest = hasher.hexdigest()
task["digest"] = task_digest
task["progress"] = 0.0
task["priority"] = priority
prev_tasks = TaskService.get_tasks(doc["id"])
ck_num = 0
@ -380,7 +382,7 @@ def queue_tasks(doc: dict, bucket: str, name: str):
unfinished_task_array = [task for task in parse_task_array if task["progress"] < 1.0]
for unfinished_task in unfinished_task_array:
assert REDIS_CONN.queue_product(
SVR_QUEUE_NAME, message=unfinished_task
get_svr_queue_name(priority), message=unfinished_task
), "Can't access Redis. Please check the Redis' status."