elon_py/pkg/tool.py

from datetime import datetime, timedelta
import pandas as pd
from pkg.config import render_data
import pytz

def aggregate_data(data, interval):
    all_minutes = pd.DataFrame({'interval_group': range(0, 1440, interval)})
    result = []

    if data.empty or 'date' not in data.columns:
        complete_data = all_minutes.copy()
        complete_data['tweet_count'] = 0
        complete_data['date'] = datetime.now().date()
        return complete_data

    for date in data['date'].unique():
        day_data = data[data['date'] == date].copy()
        day_data['interval_group'] = (day_data['minute_of_day'] // interval) * interval
        agg = day_data.groupby('interval_group').size().reset_index(name='tweet_count')
        complete_data = all_minutes.merge(agg, on='interval_group', how='left').fillna({'tweet_count': 0})
        complete_data['date'] = date
        result.append(complete_data)

    if not result:
        complete_data = all_minutes.copy()
        complete_data['tweet_count'] = 0
        complete_data['date'] = data['date'].iloc[0] if not data.empty else datetime.now().date()
        return complete_data

    return pd.concat(result, ignore_index=True)


def generate_xticks(interval):
    if interval <= 5:
        tick_step = 60
    elif interval <= 10:
        tick_step = 60
    elif interval <= 30:
        tick_step = 120
    else:
        tick_step = 240
    ticks = list(range(0, 1440, tick_step))
    tick_labels = [f"{m // 60:02d}:{m % 60:02d}" for m in ticks]
    return ticks, tick_labels


def minutes_to_time(minutes):
    hours = minutes // 60
    mins = minutes % 60
    return f"{hours:02d}:{mins:02d}"


def get_tweets_since_last_friday():
    est = pytz.timezone('US/Eastern')
    now_est = datetime.now(est)
    today = now_est.date()
    days_since_friday = (today.weekday() - 4) % 7
    this_friday = today - timedelta(days=days_since_friday)
    this_friday_datetime = est.localize(datetime.combine(this_friday, datetime.strptime("12:00", "%H:%M").time()))
    last_friday = this_friday - timedelta(days=7)
    last_friday_datetime = est.localize(datetime.combine(last_friday, datetime.strptime("12:00", "%H:%M").time()))
    if now_est < this_friday_datetime:
        start_datetime = last_friday_datetime
    else:
        start_datetime = this_friday_datetime
    if hasattr(render_data, 'global_df') and not render_data.global_df.empty:
        df = render_data.global_df.copy()
        mask = df['datetime_est'] >= start_datetime
        filtered_df = df[mask]
        tweet_count = len(filtered_df)
        return int(tweet_count)
    return 0


def get_time_since_last_tweet():
    est = pytz.timezone('US/Eastern')
    now_est = datetime.now(est)
    if (not hasattr(render_data, 'global_df') or
            render_data.global_df is None or
            render_data.global_df.empty):
        return 0.0
    df = render_data.global_df
    if 'datetime_est' not in df.columns:
        return 0.0
    latest_tweet_time = df['datetime_est'].max()
    time_diff = now_est - latest_tweet_time
    days_diff = time_diff.total_seconds() / (24 * 60 * 60)  # 转换为天数
    return days_diff

def format_time_str(days_to_next_friday):
    total_seconds = days_to_next_friday * 24 * 60 * 60
    days = int(total_seconds // (24 * 60 * 60))
    hours = int((total_seconds % (24 * 60 * 60)) // (60 * 60))
    minutes = int((total_seconds % (60 * 60)) // 60)
    seconds = int(total_seconds % 60)
    total_hours = round(days_to_next_friday * 24, 2)
    return f"{days}d {hours:02d}h {minutes:02d}m {seconds:02d}s ({total_hours}h)"


def get_hourly_weighted_array():
    est = pytz.timezone('US/Eastern')
    now = datetime.now(est).date()
    last_7_days = [now - timedelta(days=i) for i in range(7)]

    multi_data_agg = render_data.global_agg_df[
        render_data.global_agg_df['date'].isin(last_7_days)].copy()

    if multi_data_agg.empty:
        return [1 / 24] * 24

    agg_data = aggregate_data(multi_data_agg, 60)
    one_day_data = agg_data.groupby('interval_group')['tweet_count'].sum().reset_index()
    tweet_count_total = one_day_data['tweet_count'].sum()

    hourly_rates = [0] * 24
    for _, row in one_day_data.iterrows():
        minute = row['interval_group']
        hour = int(minute // 60)
        if hour < 24:
            hourly_rates[hour] = row['tweet_count'] / tweet_count_total if tweet_count_total > 0 else 0

    total_rate = sum(hourly_rates)
    if total_rate > 0:
        hourly_rates = [rate / total_rate for rate in hourly_rates]
    else:
        hourly_rates = [1 / 24] * 24
    return hourly_rates