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(document-converters): offload sync embed_text/embed_texts to thread

Browse source 
generate_document_summary and create_document_chunks are async helpers
called from the chat path and from many connector indexers. Both wrapped
embed_text/embed_texts directly inside the coroutine, blocking the event
loop for the full duration of the embedding call.
This commit is contained in:
CREDO23 2026-05-20 10:03:42 +02:00
parent 52d425f170
commit a3d6fa6196
1 changed files with 5 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

7
surfsense_backend/app/utils/document_converters.py
View file

@ -1,3 +1,4 @@
import asyncio
import hashlib import hashlib
import logging import logging
import threading import threading
@ -221,7 +222,9 @@ async def generate_document_summary(
else: else:
enhanced_summary_content = summary_content enhanced_summary_content = summary_content
summary_embedding = embed_text(enhanced_summary_content) summary_embedding = await asyncio.to_thread(
embed_text, enhanced_summary_content
)
return enhanced_summary_content, summary_embedding return enhanced_summary_content, summary_embedding
@ -237,7 +240,7 @@ async def create_document_chunks(content: str) -> list[Chunk]:
List of Chunk objects with embeddings List of Chunk objects with embeddings
""" """
chunk_texts = [c.text for c in config.chunker_instance.chunk(content)] chunk_texts = [c.text for c in config.chunker_instance.chunk(content)]
chunk_embeddings = embed_texts(chunk_texts) chunk_embeddings = await asyncio.to_thread(embed_texts, chunk_texts)
return [ return [
Chunk(content=text, embedding=emb) Chunk(content=text, embedding=emb)
for text, emb in zip(chunk_texts, chunk_embeddings, strict=False) for text, emb in zip(chunk_texts, chunk_embeddings, strict=False)