app.py

from fastapi import FastAPI, Request
from huggingface_hub import HfApi
from langchain_community.vectorstores.faiss import FAISS
from langchain_community.embeddings import OllamaEmbeddings


import os

app = FastAPI()

@app.get("/")
def greet_json():
    with open("test.txt", "w") as f:
        f.write("HELLO WORLD")
    api = HfApi()

    api.upload_file(
    
        path_or_fileobj="test.txt",
    
        path_in_repo="test.txt",
    
        repo_id="jonathanjordan21/test-dataset",
    
        repo_type="dataset",

        token = os.getenv("HF_WRITE_TOKEN")
    
    )
    return {"Hello": "World!"}


@app.get("/{rand_int}/")
async def main(rand_int: str, request: Request):
    with open("test.txt", "w") as f:
        f.write(rand_int)
    api = HfApi()

    api.upload_file(
    
        path_or_fileobj="test.txt",
    
        path_in_repo="test.txt",
    
        repo_id="jonathanjordan21/test-dataset",
    
        repo_type="dataset",

        token = os.getenv("HF_WRITE_TOKEN")
    
    )
    return {"raw_url": str(request.url), "message":rand_int}




@app.get("/vecs/vecs")
async def main2(request: Request):
    emb = OllamaEmbeddings(model="intfloat/multilingual-e5-large-instruct",
                base_url="https://lintasmediadanawa-hf-llm-api.hf.space",
                embed_instruction="",
                query_instruction="Instruct: Given a web search query, retrieve relevant passages that answer the query\nQuery: ",
                headers={"Authorization": f"Bearer {os.getenv("HF_WRITE_TOKEN")}"},
            )
    vecs = FAISS.from_texts(["haha"], emb)
    api = HfApi()

    vecs.save_local("faiss_index")

    api.upload_folder(

        folder_path="faiss_index",
    
        repo_id="jonathanjordan21/test-dataset",

        path_in_repo="vecs/faiss_index",
    
        repo_type="dataset",

        token = os.getenv("HF_WRITE_TOKEN")
    
    )

    return {"raw_url": str(request.url)}