app.py

import boto3
import os
import re
import json
from pathlib import Path
import sqlite3
from huggingface_hub import Repository, HfFolder
import tqdm
import subprocess

from fastapi import FastAPI
from fastapi_utils.tasks import repeat_every

paginator = s3.get_paginator('list_objects_v2')

repo.git_pull()

def get_datasetdata(dataset_path):



def sync_rooms_to_dataset():
    for room_data_db in get_db(ROOMS_DATA_DB):

                    except Exception as e:
                        print(e)
                        continue

                objects += results
            room_data_db.commit()

            all_rows = [dict(row) for row in room_data_db.execute(
                "SELECT * FROM rooms_data WHERE room_id = ?", (room_id,)).fetchall()]
            data_path = S3_DATA_FOLDER / f"{room_id}.json"
            with open(data_path, 'w') as f:
                json.dump(all_rows, f, separators=(',', ':'))
    print("Updating repository")
    subprocess.Popen(
        "git add . && git commit --amend -m 'update' && git push --force", cwd=S3_DATA_FOLDER, shell=True)


app = FastAPI()


@app.post("/")
def read_root():
    return "Just a bot to sync data to huggingface datasets and tweet tha latest data"


@app.post("/sync")
def sync():
    sync_rooms_to_dataset()
    return "Synced data to huggingface datasets"


@app.on_event("startup")
@repeat_every(seconds=1800)
def repeat_sync():
    sync_rooms_to_dataset()
    return "Synced data to huggingface datasets"