fix

pkg/dash/func/render.py

@@ -2,10 +2,12 @@ from datetime import datetime, timedelta
 from dash.dependencies import Input, Output
 from pkg.dash.app_init import app
 from pkg.config import render_data
-from pkg.tool import aggregate_data, minutes_to_time, get_tweets_since_last_friday
+from pkg.tool import aggregate_data, minutes_to_time, get_tweets_since_last_friday,get_pace_and_total_tweets
 from dash import dcc
 import plotly.graph_objs as go
 import pandas as pd
+import pytz
+import numpy as np
 
 @app.callback(
     [Output('tabs-content', 'children'),

pkg/tool.py

@@ -96,3 +96,54 @@ def format_time_str(days_to_next_friday):
     return f"{days}d {hours:02d}h {minutes:02d}m {seconds:02d}s ({total_hours}h)"
 
 
+def get_pace_and_total_tweets(target_time: datetime) -> tuple[float, int]:
+    est = pytz.timezone('US/Eastern')
+
+    # 如果 target_time 没有时区信息，假设为 EST
+    if target_time.tzinfo is None:
+        target_time = est.localize(target_time)
+
+    # 计算上周五 12:00 AM EST
+    target_date = target_time.date()
+    days_since_last_friday = (target_date.weekday() + 3) % 7  # 距离上周五的天数
+    last_friday = target_time - timedelta(days=days_since_last_friday)
+    last_friday_midnight = last_friday.replace(hour=0, minute=0, second=0, microsecond=0)
+
+    # 计算下周五 12:00 AM EST
+    days_to_next_friday = (4 - target_date.weekday()) % 7
+    next_friday = target_time + timedelta(days=days_to_next_friday)
+    next_friday_midnight = next_friday.replace(hour=0, minute=0, second=0, microsecond=0)
+    if target_time > next_friday_midnight:
+        next_friday_midnight += timedelta(days=7)
+
+    # 从 global_agg_df 中筛选从上周五 12:00 AM 到 target_time 的数据
+    if hasattr(render_data, 'global_agg_df') and not render_data.global_agg_df.empty:
+        multi_data_agg = render_data.global_agg_df[
+            (render_data.global_agg_df['date'] >= last_friday_midnight.date()) &
+            (render_data.global_agg_df['date'] <= target_date)
+            ].copy()
+    else:
+        multi_data_agg = pd.DataFrame()
+
+    if multi_data_agg.empty:
+        total_tweets = 0
+    else:
+        # 使用 minute_of_day 转换为时间戳并筛选到 target_time 之前
+        multi_data_agg['timestamp'] = pd.to_datetime(multi_data_agg['date'].astype(str)) + \
+                                      pd.to_timedelta(multi_data_agg['minute_of_day'], unit='m')
+        multi_data_agg['timestamp'] = multi_data_agg['timestamp'].dt.tz_localize(est)
+        multi_data_agg = multi_data_agg[multi_data_agg['timestamp'] <= target_time]
+        total_tweets = multi_data_agg['tweet_count'].sum() if 'tweet_count' in multi_data_agg else 0
+
+    # 计算 Pace
+    days_elapsed = (target_time - last_friday_midnight).total_seconds() / (24 * 60 * 60)
+    days_remaining = (next_friday_midnight - target_time).total_seconds() / (24 * 60 * 60)
+
+    if days_elapsed > 0 and total_tweets > 0:
+        daily_avg = total_tweets / days_elapsed
+        pace = daily_avg * days_remaining + total_tweets
+    else:
+        pace = float(total_tweets)  # 如果没有数据或时间未开始，Pace 等于当前推文数
+
+    return round(pace, 2), int(total_tweets)
+