mirror of
https://github.com/danswer-ai/danswer.git
synced 2025-06-04 03:59:25 +02:00
24184024bb
* bump fastapi and starlette * bumping llama index and nltk and associated deps * bump to fix python-multipart * bump aiohttp * update package lock for examples/widget * bump black * sentencesplitter has changed namespaces * fix reorder import check, fix missing passlib * update package-lock.json * black formatter updated * reformatted again * change to black compatible reorder * change to black compatible reorder-python-imports fork * fix pytest dependency * black format again * we don't need cdk.txt. update packages to be consistent across all packages --------- Co-authored-by: Richard Kuo (Onyx) <rkuo@onyx.app> Co-authored-by: Richard Kuo <rkuo@rkuo.com>
104 lines
2.7 KiB
Python
104 lines
2.7 KiB
Python
import math
|
|
from datetime import datetime
|
|
|
|
from fastapi import APIRouter
|
|
from fastapi import Depends
|
|
from pydantic import BaseModel
|
|
from sqlalchemy.orm import Session
|
|
|
|
from onyx.context.search.models import SearchRequest
|
|
from onyx.context.search.pipeline import SearchPipeline
|
|
from onyx.db.engine import get_session
|
|
from onyx.db.models import User
|
|
from onyx.llm.factory import get_default_llms
|
|
from onyx.server.onyx_api.ingestion import api_key_dep
|
|
from onyx.utils.logger import setup_logger
|
|
|
|
|
|
logger = setup_logger()
|
|
|
|
|
|
router = APIRouter(prefix="/gpts")
|
|
|
|
|
|
def time_ago(dt: datetime) -> str:
|
|
# Calculate time difference
|
|
now = datetime.now()
|
|
diff = now - dt
|
|
|
|
# Convert difference to minutes
|
|
minutes = diff.total_seconds() / 60
|
|
|
|
# Determine the appropriate unit and calculate the age
|
|
if minutes < 60:
|
|
return f"~{math.floor(minutes)} minutes"
|
|
hours = minutes / 60
|
|
if hours < 24:
|
|
return f"~{math.floor(hours)} hours"
|
|
days = hours / 24
|
|
if days < 7:
|
|
return f"~{math.floor(days)} days"
|
|
weeks = days / 7
|
|
if weeks < 4:
|
|
return f"~{math.floor(weeks)} weeks"
|
|
months = days / 30
|
|
return f"~{math.floor(months)} months"
|
|
|
|
|
|
class GptSearchRequest(BaseModel):
|
|
query: str
|
|
|
|
|
|
class GptDocChunk(BaseModel):
|
|
title: str
|
|
content: str
|
|
source_type: str
|
|
link: str
|
|
metadata: dict[str, str | list[str]]
|
|
document_age: str
|
|
|
|
|
|
class GptSearchResponse(BaseModel):
|
|
matching_document_chunks: list[GptDocChunk]
|
|
|
|
|
|
@router.post("/gpt-document-search")
|
|
def gpt_search(
|
|
search_request: GptSearchRequest,
|
|
_: User | None = Depends(api_key_dep),
|
|
db_session: Session = Depends(get_session),
|
|
) -> GptSearchResponse:
|
|
llm, fast_llm = get_default_llms()
|
|
top_sections = SearchPipeline(
|
|
search_request=SearchRequest(
|
|
query=search_request.query,
|
|
),
|
|
user=None,
|
|
llm=llm,
|
|
fast_llm=fast_llm,
|
|
skip_query_analysis=True,
|
|
db_session=db_session,
|
|
).reranked_sections
|
|
|
|
return GptSearchResponse(
|
|
matching_document_chunks=[
|
|
GptDocChunk(
|
|
title=section.center_chunk.semantic_identifier,
|
|
content=section.center_chunk.content,
|
|
source_type=section.center_chunk.source_type,
|
|
link=(
|
|
section.center_chunk.source_links.get(0, "")
|
|
if section.center_chunk.source_links
|
|
else ""
|
|
),
|
|
metadata=section.center_chunk.metadata,
|
|
document_age=(
|
|
time_ago(section.center_chunk.updated_at)
|
|
if section.center_chunk.updated_at
|
|
else "Unknown"
|
|
),
|
|
)
|
|
for section in top_sections
|
|
],
|
|
)
|