分享JMS流量统计面板

雪秋千 发表于 2025-3-14 16:25:13

本帖最后由雪秋千于 2025-3-14 18:05 编辑 
 
各位大侠有justmysocks的API的流量统计面板吗？自己用GPT写的太丑。 
下面是自己写的效果预览图 

<ignore_js_op>

<img src="https://hostloc.com/static/image/filetype/image.gif" border="0" class="vm" alt="" />


<a href="https://hostloc.com/forum.php?mod=attachment&aid=MTczMTg3fGZlZWU3MWY1fDE3NDIwODA4NTl8MHwxMzk2MDI1" target="_blank">Screenshot 2025-03-14 180331.png</a>

(65.11 KB, 下载次数: 2)

<div class="tip tip_4" id="attach_173187_menu" style="position: absolute; display: none" disautofocus="true">
<div class="tip_c xs0">
<div class="y">前天18:03 上传</div>
点击文件名下载附件

</div>
<div class="tip_horn"></div>
</div>

</ignore_js_op>

雪秋千 发表于 2025-3-14 18:02:58

先把我的放上来吧 
1. 首先是本地存储数据的，定期30秒运行一次。因为jms更新时间差不多就是半分钟。 <div class="blockcode"><div id="code_W4X"><ol><li>import requests <li>import json <li>from datetime import datetime <li> <li>API_URL = "https://justmysocks6.net/members/getbwcounter.php?service=" <li>HISTORY_FILE = "扶Q_usage_history.json" <li> <li>def fetch_bw_counter(): <li>response = requests.get(API_URL) <li>data = response.json() <li># data = {"monthly_bw_limit_b": 1000000000000, <li># "bw_counter_b": 651529129, <li># "bw_reset_day_of_month": 13} <li>return data["bw_counter_b"] <li> <li>def load_history(): <li>try: <li>with open(HISTORY_FILE, "r") as f: <li> return json.load(f) <li>except (IOError, ValueError): <li>return [] <li> <li>def save_history(history): <li>with open(HISTORY_FILE, "w") as f: <li>json.dump(history, f, indent=2) <li> <li>def record_usage(): <li># 1) Fetch the current usage <li>current_bw = fetch_bw_counter() <li>timestamp = datetime.utcnow().isoformat() <li> <li># 2) Append new record to history <li>history = load_history() <li>history.append({ <li>"timestamp": timestamp, <li>"bw_counter_b": current_bw <li>}) <li>save_history(history) <li> <li>if __name__ == "__main__": <li>record_usage() <li></ol></div>复制代码</div> 
 
2. 第二是显示面板，第一个面板是总体使用量，第二个是在不同时间的速率， <div class="blockcode"><div id="code_vEW"><ol><li>import json <li>import pandas as pd <li>from datetime import timedelta <li>import dash <li>from dash import dcc, html, Input, Output <li>import plotly.graph_objs as go <li> <li># ----- Utility Functions ----- <li> <li>def load_usage_data(file_path="扶Q_usage_history.json"): <li>"""Load usage data from JSON, localize to America/Los_Angeles, then convert to UTC. <li> Plotly will automatically render these timestamps in the visitor’s local time. <li>""" <li>with open(file_path, "r") as f: <li>data = json.load(f) <li>df = pd.DataFrame(data) <li># Assume timestamps in the file are in California time. <li>df['timestamp'] = pd.to_datetime(df['timestamp']).dt.tz_localize('America/Los_Angeles') <li># Convert to UTC for consistent plotting. <li>df['timestamp'] = df['timestamp'].dt.tz_convert('UTC') <li>df.sort_values('timestamp', inplace=True) <li>return df <li> <li>def convert_bytes(value_bytes): <li>""" <li>Convert a byte value to a human-friendly string using a 1000 conversion factor. <li>If the value in GB is less than 0.001, display in MB. <li>If in MB is less than 0.001, display in B. <li>""" <li>value_gb = value_bytes / 1e9 <li>if value_gb >= 0.001: <li>return f"{value_gb:.3f} GB" <li>value_mb = value_bytes / 1e6 <li>if value_mb >= 0.001: <li>return f"{value_mb:.3f} MB" <li>return f"{value_bytes} B" <li> <li>def aggregate_data(df, resolution, window): <li>""" <li>Aggregate usage data for a given resolution and time window. <li>resolution: a pandas offset alias, e.g., 'T' for minute, 'H' for hour, 'D' for day, 'W' for week. <li>window: timedelta object representing the lookback period. <li>""" <li>end_time = df['timestamp'].max() <li>start_time = end_time - window <li>df_window = df >= start_time].copy() <li>if df_window.empty: <li>return pd.DataFrame(columns=['timestamp', 'bw_counter_b']) <li>df_window.set_index('timestamp', inplace=True) <li>df_resampled = df_window.resample(resolution).last().dropna() <li>df_resampled.reset_index(inplace=True) <li>return df_resampled <li> <li>def compute_usage_rates(df): <li>""" <li>Compute the incremental usage (difference between consecutive bw_counter_b) <li>and time differences. Returns the DataFrame with a new column 'usage_diff'. <li>""" <li>df = df.copy() <li>df['usage_diff'] = df['bw_counter_b'].diff() <li>df['time_diff_sec'] = df['timestamp'].diff().dt.total_seconds() <li>df['usage_rate'] = df['usage_diff'] / df['time_diff_sec'] <li>return df <li> <li># ----- Dash App Setup ----- <li> <li>app = dash.Dash(__name__) <li>server = app.server <li> <li>app.layout = html.Div([ <li>html.H1("扶Q Data Usage Dashboard"), <li>html.Div([ <li>html.Button("Minutes", id="btn-minutes", n_clicks=0), <li>html.Button("Hourly", id="btn-hourly", n_clicks=0), <li>html.Button("Daily", id="btn-daily", n_clicks=0), <li>html.Button("Weekly", id="btn-weekly", n_clicks=0) <li>], style={'marginBottom': '20px'}), <li>html.Div(id="summary-stats", style={'marginBottom': '20px'}), <li>dcc.Graph(id="usage-graph"), <li>dcc.Graph(id="rate-graph"), <li>dcc.Interval(id="interval-update", interval=60*1000, n_intervals=0)# update every minute <li>]) <li> <li># ----- Callback to Update Graphs and Stats ----- <li> <li>@app.callback( <li>, <li> <li>) <li>def update_dashboard(n_min, n_hour, n_day, n_week, n_interval): <li>df = load_usage_data() <li> <li># Determine which button was most recently pressed <li>ctx = dash.callback_context <li>if not ctx.triggered: <li>resolution_choice = 'H' <li>window = timedelta(hours=24) <li>else: <li>button_id = ctx.triggered['prop_id'].split('.') <li>if button_id == "btn-minutes": <li> resolution_choice = 'T'# minute resolution <li> window = timedelta(hours=1) <li>elif button_id == "btn-hourly": <li> resolution_choice = 'H' <li> window = timedelta(hours=24) <li>elif button_id == "btn-daily": <li> resolution_choice = 'D' <li> window = timedelta(days=7) <li>elif button_id == "btn-weekly": <li> resolution_choice = 'W' <li> window = timedelta(weeks=4) <li>else: <li> resolution_choice = 'H' <li> window = timedelta(hours=24) <li> <li>df_agg = aggregate_data(df, resolution_choice, window) <li>df_rate = compute_usage_rates(df_agg) <li> <li># ----- Cumulative Usage Figure ----- <li>cum_fig = go.Figure() <li>cum_fig.add_trace(go.Scatter( <li>x=df_agg['timestamp'], <li>y=df_agg['bw_counter_b'] / 1e9,# cumulative usage in GB <li>mode='lines+markers', <li>name="Cumulative Usage (GB)", <li>connectgaps=False <li>)) <li>cum_fig.update_layout( <li>title="扶Q Cumulative Usage Over Time", <li>xaxis_title="Time", <li>yaxis_title="Usage (GB)", <li>hovermode="x unified" <li>) <li> <li># ----- Usage Rate Figure ----- <li>df_rate_clean = df_rate.dropna(subset=['usage_diff']) <li>if not df_rate_clean.empty: <li>max_diff = df_rate_clean['usage_diff'].max() <li>if max_diff / 1e9 >= 0.001: <li> factor = 1e9 <li> y_label = "Usage per Interval (GB)" <li>elif max_diff / 1e6 >= 0.001: <li> factor = 1e6 <li> y_label = "Usage per Interval (MB)" <li>else: <li> factor = 1 <li> y_label = "Usage per Interval (B)" <li>usage_diff_converted = df_rate_clean['usage_diff'] / factor <li>else: <li>usage_diff_converted = [] <li>y_label = "Usage per Interval" <li> <li>rate_fig = go.Figure() <li>rate_fig.add_trace(go.Scatter( <li>x=df_rate_clean['timestamp'], <li>y=usage_diff_converted, <li>mode='lines+markers', <li>name="Interval Usage", <li>connectgaps=False <li>)) <li>rate_fig.update_layout( <li>title="扶Q Usage Rate Over Time", <li>xaxis_title="Time", <li>yaxis_title=y_label, <li>hovermode="x unified" <li>) <li> <li># ----- Summary Statistics ----- <li>if not df_rate['usage_rate'].dropna().empty: <li>avg_rate = df_rate['usage_rate'].dropna().mean()# bytes per second <li>avg_per_min = convert_bytes(avg_rate * 60) <li>avg_per_hour = convert_bytes(avg_rate * 3600) <li>avg_per_day = convert_bytes(avg_rate * 3600 * 24) <li>avg_per_week = convert_bytes(avg_rate * 3600 * 24 * 7) <li>else: <li>avg_per_min = avg_per_hour = avg_per_day = avg_per_week = "N/A" <li> <li>summary = html.Div([ <li>html.P(f"Average Usage per Minute: {avg_per_min}"), <li>html.P(f"Average Usage per Hour: {avg_per_hour}"), <li>html.P(f"Average Usage per Day: {avg_per_day}"), <li>html.P(f"Average Usage per Week: {avg_per_week}") <li>]) <li> <li>return cum_fig, rate_fig, summary <li> <li>if __name__ == '__main__': <li>app.run_server(debug=True) <li></ol></div>复制代码</div>

页: [1]

全球主机交流论坛备用站's Archiver

分享JMS流量统计面板