|
|
@@ -0,0 +1,272 @@
|
|
|
+import sys
|
|
|
+import os
|
|
|
+import json
|
|
|
+from datetime import datetime, timedelta
|
|
|
+
|
|
|
+import streamlit as st
|
|
|
+import pandas as pd
|
|
|
+from sqlalchemy import create_engine
|
|
|
+from dotenv import load_dotenv
|
|
|
+import plotly.express as px
|
|
|
+
|
|
|
+# ===== 项目路径 =====
|
|
|
+sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))
|
|
|
+from app.core.config import settings
|
|
|
+
|
|
|
+# ===== 页面配置 =====
|
|
|
+st.set_page_config(
|
|
|
+ page_title="微信指数趋势看板",
|
|
|
+ page_icon="📈",
|
|
|
+ layout="wide",
|
|
|
+ initial_sidebar_state="expanded",
|
|
|
+# menu_items={
|
|
|
+# 'Get Help': 'https://www.extremelycoolapp.com/help',
|
|
|
+# 'Report a bug': "https://www.extremelycoolapp.com/bug",
|
|
|
+# 'About': "# This is a header. This is an *extremely* cool app!"
|
|
|
+# }
|
|
|
+)
|
|
|
+
|
|
|
+load_dotenv(os.path.join(os.path.dirname(__file__), '../.env'))
|
|
|
+
|
|
|
+# ===== 字段映射 =====
|
|
|
+FIELD_MAPPING = {
|
|
|
+ 'total_score': '总指数',
|
|
|
+ 'finder_score': '视频号',
|
|
|
+ 'query_score': '搜一搜',
|
|
|
+ 'mpdoc_score': '公众号',
|
|
|
+ 'live_score': '直播',
|
|
|
+ 'miniapp_score': '小程序'
|
|
|
+}
|
|
|
+
|
|
|
+# ======================
|
|
|
+# 数据层
|
|
|
+# ======================
|
|
|
+
|
|
|
+@st.cache_resource
|
|
|
+def get_db_connection():
|
|
|
+ return create_engine(settings.DATABASE_URL.replace('+aiomysql', '+pymysql'))
|
|
|
+
|
|
|
+
|
|
|
+@st.cache_data(ttl=3600)
|
|
|
+def get_keywords():
|
|
|
+ engine = get_db_connection()
|
|
|
+ sql = """
|
|
|
+ SELECT id, keyword
|
|
|
+ FROM wx_trend_keywords
|
|
|
+ WHERE is_active = 1
|
|
|
+ ORDER BY priority DESC
|
|
|
+ """
|
|
|
+ with engine.connect() as conn:
|
|
|
+ return pd.read_sql(sql, conn)
|
|
|
+
|
|
|
+
|
|
|
+@st.cache_data(ttl=3600)
|
|
|
+def get_trend_data(keyword_ids, start_date, end_date):
|
|
|
+ if not keyword_ids:
|
|
|
+ return pd.DataFrame()
|
|
|
+
|
|
|
+ engine = get_db_connection()
|
|
|
+ s_date = start_date.strftime("%Y%m%d")
|
|
|
+ e_date = end_date.strftime("%Y%m%d")
|
|
|
+
|
|
|
+ ids_tuple = tuple(keyword_ids) if len(keyword_ids) > 1 else f"({keyword_ids[0]})"
|
|
|
+
|
|
|
+ sql = f"""
|
|
|
+ SELECT
|
|
|
+ k.keyword,
|
|
|
+ d.ymd,
|
|
|
+ d.channel_score
|
|
|
+ FROM wx_trend_data d
|
|
|
+ JOIN wx_trend_keywords k ON d.keyword_id = k.id
|
|
|
+ WHERE d.keyword_id IN {ids_tuple}
|
|
|
+ AND d.ymd BETWEEN '{s_date}' AND '{e_date}'
|
|
|
+ ORDER BY d.ymd ASC
|
|
|
+ """
|
|
|
+
|
|
|
+ with engine.connect() as conn:
|
|
|
+ df = pd.read_sql(sql, conn)
|
|
|
+
|
|
|
+ if df.empty:
|
|
|
+ return df
|
|
|
+
|
|
|
+ if isinstance(df['channel_score'].iloc[0], str):
|
|
|
+ df['channel_score'] = df['channel_score'].apply(json.loads)
|
|
|
+
|
|
|
+ score_df = pd.json_normalize(df['channel_score'])
|
|
|
+ result = pd.concat([df[['keyword', 'ymd']], score_df], axis=1)
|
|
|
+ result['date'] = pd.to_datetime(result['ymd'], format='%Y%m%d')
|
|
|
+ return result
|
|
|
+
|
|
|
+
|
|
|
+# ======================
|
|
|
+# Sidebar
|
|
|
+# ======================
|
|
|
+
|
|
|
+st.sidebar.header("🔍 筛选条件")
|
|
|
+
|
|
|
+kw_df = get_keywords()
|
|
|
+selected_keywords = st.sidebar.multiselect(
|
|
|
+ "关键词",
|
|
|
+ options=kw_df['keyword'].tolist(),
|
|
|
+ default=kw_df['keyword'].iloc[:3].tolist() if not kw_df.empty else []
|
|
|
+)
|
|
|
+
|
|
|
+selected_ids = kw_df[kw_df['keyword'].isin(selected_keywords)]['id'].tolist()
|
|
|
+
|
|
|
+date_range = st.sidebar.date_input(
|
|
|
+ "日期范围",
|
|
|
+ value=(datetime.now() - timedelta(days=30), datetime.now()),
|
|
|
+ max_value=datetime.now()
|
|
|
+)
|
|
|
+
|
|
|
+selected_metrics = st.sidebar.multiselect(
|
|
|
+ "展示维度",
|
|
|
+ options=list(FIELD_MAPPING.keys()),
|
|
|
+ format_func=lambda x: FIELD_MAPPING[x],
|
|
|
+ default=['total_score']
|
|
|
+)
|
|
|
+
|
|
|
+# ======================
|
|
|
+# Main
|
|
|
+# ======================
|
|
|
+
|
|
|
+st.title("微信指数趋势洞察")
|
|
|
+
|
|
|
+if not selected_ids:
|
|
|
+ st.warning("请至少选择一个关键词")
|
|
|
+ st.stop()
|
|
|
+
|
|
|
+if len(date_range) != 2:
|
|
|
+ st.warning("请选择完整的日期范围")
|
|
|
+ st.stop()
|
|
|
+
|
|
|
+with st.spinner("正在加载数据..."):
|
|
|
+ df = get_trend_data(selected_ids, date_range[0], date_range[1])
|
|
|
+
|
|
|
+if df.empty:
|
|
|
+ st.warning("所选时间段内暂无数据")
|
|
|
+ st.stop()
|
|
|
+
|
|
|
+# ======================
|
|
|
+# KPI 卡片
|
|
|
+# ======================
|
|
|
+
|
|
|
+st.markdown("### 💡 最新指数概览")
|
|
|
+
|
|
|
+cols = st.columns(min(len(selected_keywords), 4))
|
|
|
+
|
|
|
+for i, kw in enumerate(selected_keywords):
|
|
|
+ kw_df = df[df['keyword'] == kw].sort_values("date")
|
|
|
+ if kw_df.empty:
|
|
|
+ continue
|
|
|
+
|
|
|
+ latest = kw_df.iloc[-1]
|
|
|
+ prev = kw_df.iloc[-2] if len(kw_df) > 1 else None
|
|
|
+
|
|
|
+ delta = ""
|
|
|
+ if prev is not None and prev['total_score'] > 0:
|
|
|
+ pct = (latest['total_score'] - prev['total_score']) / prev['total_score'] * 100
|
|
|
+ delta = f"{pct:+.2f}%"
|
|
|
+
|
|
|
+ with cols[i % 4]:
|
|
|
+ st.metric(
|
|
|
+ label=f"{kw}（总指数）",
|
|
|
+ value=f"{int(latest['total_score']):,}",
|
|
|
+ delta=delta
|
|
|
+ )
|
|
|
+
|
|
|
+st.divider()
|
|
|
+
|
|
|
+# ======================
|
|
|
+# 趋势图（专业版）
|
|
|
+# ======================
|
|
|
+
|
|
|
+st.markdown("### 📈 趋势对比分析")
|
|
|
+
|
|
|
+compare_metric = st.selectbox(
|
|
|
+ "对比指标",
|
|
|
+ options=selected_metrics,
|
|
|
+ format_func=lambda x: FIELD_MAPPING[x]
|
|
|
+)
|
|
|
+
|
|
|
+fig = px.line(
|
|
|
+ df,
|
|
|
+ x="date",
|
|
|
+ y=compare_metric,
|
|
|
+ color="keyword",
|
|
|
+ title=f"各关键词【{FIELD_MAPPING[compare_metric]}】趋势对比"
|
|
|
+)
|
|
|
+
|
|
|
+# —— 核心专业配置 ——
|
|
|
+fig.update_traces(
|
|
|
+ mode="lines",
|
|
|
+ line=dict(width=3),
|
|
|
+ hovertemplate=(
|
|
|
+ "日期：%{x|%Y.%m.%d}<br>"
|
|
|
+ "关键词：%{text}<br>"
|
|
|
+ "指数：%{y:,.0f}"
|
|
|
+ "<extra></extra>"
|
|
|
+ ),
|
|
|
+ text=df['keyword'] # 使用 text 参数传递关键词
|
|
|
+)
|
|
|
+
|
|
|
+fig.update_layout(
|
|
|
+ hovermode="x unified",
|
|
|
+ margin=dict(t=80, l=60, r=40, b=50),
|
|
|
+
|
|
|
+ legend=dict(
|
|
|
+ orientation="h",
|
|
|
+ yanchor="bottom",
|
|
|
+ y=1.05,
|
|
|
+ xanchor="right",
|
|
|
+ x=1,
|
|
|
+ title="关键词（点击可隐藏）"
|
|
|
+ ),
|
|
|
+
|
|
|
+ xaxis=dict(
|
|
|
+ title="日期",
|
|
|
+ tickformat="%Y.%m.%d",
|
|
|
+ showgrid=False,
|
|
|
+ tickfont=dict(size=12)
|
|
|
+ ),
|
|
|
+
|
|
|
+ yaxis=dict(
|
|
|
+ title="指数",
|
|
|
+ showgrid=True,
|
|
|
+ gridcolor="rgba(255,255,255,0.06)",
|
|
|
+ tickfont=dict(size=12),
|
|
|
+ tickformat=",d"
|
|
|
+ )
|
|
|
+)
|
|
|
+fig.update_traces(
|
|
|
+ mode="lines",
|
|
|
+ line=dict(width=3),
|
|
|
+ hovertemplate=(
|
|
|
+ # "日期：%{x|%Y.%m.%d}<br>"
|
|
|
+ "关键词：%{text}<br>"
|
|
|
+ "指数：%{y:,.0f}"
|
|
|
+ "<extra></extra>"
|
|
|
+ ),
|
|
|
+ text=df['keyword']
|
|
|
+)
|
|
|
+
|
|
|
+st.plotly_chart(fig, width='stretch')
|
|
|
+
|
|
|
+# st.caption(
|
|
|
+# "说明：微信指数为相对热度指标，用于趋势对比分析，不代表绝对搜索量。"
|
|
|
+# )
|
|
|
+
|
|
|
+# ======================
|
|
|
+# 原始数据
|
|
|
+# ======================
|
|
|
+
|
|
|
+with st.expander("📄 查看原始数据"):
|
|
|
+ display_df = df[['date', 'keyword'] + selected_metrics].copy()
|
|
|
+ display_df.columns = ['日期', '关键词'] + [FIELD_MAPPING[m] for m in selected_metrics]
|
|
|
+ display_df['日期'] = display_df['日期'].dt.strftime("%Y-%m-%d")
|
|
|
+
|
|
|
+ st.dataframe(
|
|
|
+ display_df.sort_values(['日期', '关键词'], ascending=[False, True]),
|
|
|
+ width='stretch',
|
|
|
+ hide_index=True
|
|
|
+ )
|
