apunkt/SurfSense
1
0
Fork 0
mirror of https://github.com/MODSetter/SurfSense.git synced 2026-05-27 19:25:15 +02:00
Code Issues Projects Releases Packages Wiki Activity

perf(kb-persistence): offload sync embed_texts to thread

Browse source 
_create_document and _update_document run on the chat critical path
when the filesystem subagent writes via the user's chat turn. Both
called embed_texts synchronously inside an async coroutine, blocking
the event loop for the duration of the embed.
This commit is contained in:
CREDO23 2026-05-20 10:03:14 +02:00
parent 4fa85a9a94
commit 52d425f170
1 changed files with 5 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

9
surfsense_backend/app/agents/new_chat/middleware/kb_persistence.py
View file

@ -32,6 +32,7 @@ exact same routine when ``aafter_agent`` was skipped (e.g. client disconnect).
from __future__ import annotations from __future__ import annotations
import asyncio
import logging import logging
from datetime import UTC, datetime from datetime import UTC, datetime
from typing import Any from typing import Any
@ -249,11 +250,11 @@ async def _create_document(
session.add(doc) session.add(doc)
await session.flush() await session.flush()
summary_embedding = embed_texts([content])[0] summary_embedding = (await asyncio.to_thread(embed_texts, [content]))[0]
doc.embedding = summary_embedding doc.embedding = summary_embedding
chunks = chunk_text(content) chunks = chunk_text(content)
if chunks: if chunks:
chunk_embeddings = embed_texts(chunks) chunk_embeddings = await asyncio.to_thread(embed_texts, chunks)
session.add_all( session.add_all(
[ [
Chunk(document_id=doc.id, content=text, embedding=embedding) Chunk(document_id=doc.id, content=text, embedding=embedding)
@ -295,13 +296,13 @@ async def _update_document(
search_space_id, search_space_id,
) )
summary_embedding = embed_texts([content])[0] summary_embedding = (await asyncio.to_thread(embed_texts, [content]))[0]
document.embedding = summary_embedding document.embedding = summary_embedding
await session.execute(delete(Chunk).where(Chunk.document_id == document.id)) await session.execute(delete(Chunk).where(Chunk.document_id == document.id))
chunks = chunk_text(content) chunks = chunk_text(content)
if chunks: if chunks:
chunk_embeddings = embed_texts(chunks) chunk_embeddings = await asyncio.to_thread(embed_texts, chunks)
session.add_all( session.add_all(
[ [
Chunk(document_id=document.id, content=text, embedding=embedding) Chunk(document_id=document.id, content=text, embedding=embedding)