#!/usr/bin/env python3
"""导出2026年付费用户中5月完成>30课时的103人明细"""

import os
import psycopg2
import pandas as pd
from openpyxl import Workbook
from openpyxl.styles import Font, Alignment, PatternFill, Border, Side
from openpyxl.utils import get_column_letter

PG_PASSWORD = "LdfjdjL83h3h3^$&**YGG*"

SQL = """
WITH user_orders AS (
    SELECT DISTINCT o.account_id, o.goods_id, o.trade_no, o.key_from, o.pay_success_date
    FROM bi_vala_order o
    JOIN bi_vala_app_account a ON o.account_id = a.id AND a.status = 1
    WHERE o.pay_success_date >= '2026-01-01'
      AND o.pay_success_date IS NOT NULL
      AND o.order_status IN (3, 4)
      AND o.goods_id NOT IN (32)
),
refunded_trades AS (
    SELECT DISTINCT r.trade_no
    FROM bi_refund_order r
    JOIN bi_vala_order o ON r.trade_no = o.trade_no AND o.order_status = 4
    WHERE r.status = 3
),
user_all_refunded AS (
    SELECT uo.account_id
    FROM user_orders uo
    LEFT JOIN refunded_trades rt ON uo.trade_no = rt.trade_no
    GROUP BY uo.account_id
    HAVING COUNT(*) = COUNT(rt.trade_no)
),
non_refund_users AS (
    SELECT DISTINCT uo.account_id
    FROM user_orders uo
    WHERE uo.account_id NOT IN (SELECT account_id FROM user_all_refunded)
),
user_level AS (
    SELECT
        uo.account_id,
        CASE
            WHEN bool_or(uo.goods_id = 61) THEN 'L1+L2'
            WHEN bool_or(uo.goods_id IN (57, 60, 63)) AND bool_or(uo.goods_id IN (31, 33, 54)) THEN 'L1+L2'
            WHEN bool_or(uo.goods_id IN (57, 60, 63)) THEN 'L1'
            WHEN bool_or(uo.goods_id IN (31, 33, 54)) THEN 'L2'
            ELSE '其他'
        END AS level
    FROM user_orders uo
    GROUP BY uo.account_id
),
purchase_info AS (
    SELECT account_id, 
           STRING_AGG(DISTINCT key_from, ', ' ORDER BY key_from) AS channels,
           MIN(pay_success_date) AS first_purchase_date
    FROM user_orders
    GROUP BY account_id
),
target_chars AS (
    SELECT c.id AS char_id, c.account_id, c.gender, c.birthday
    FROM bi_vala_app_character c
    JOIN non_refund_users n ON c.account_id = n.account_id
),
chapter_done AS (
    SELECT cp.user_id AS char_id, COUNT(DISTINCT cp.chapter_id) AS done_count
    FROM bi_user_chapter_play_record_0 cp
    JOIN target_chars tc ON cp.user_id = tc.char_id AND tc.char_id % 8 = 0
    WHERE cp.play_status = 1
      AND cp.created_at >= '2026-05-01' AND cp.created_at < '2026-06-01'
    GROUP BY cp.user_id
    UNION ALL
    SELECT cp.user_id AS char_id, COUNT(DISTINCT cp.chapter_id) AS done_count
    FROM bi_user_chapter_play_record_1 cp
    JOIN target_chars tc ON cp.user_id = tc.char_id AND tc.char_id % 8 = 1
    WHERE cp.play_status = 1
      AND cp.created_at >= '2026-05-01' AND cp.created_at < '2026-06-01'
    GROUP BY cp.user_id
    UNION ALL
    SELECT cp.user_id AS char_id, COUNT(DISTINCT cp.chapter_id) AS done_count
    FROM bi_user_chapter_play_record_2 cp
    JOIN target_chars tc ON cp.user_id = tc.char_id AND tc.char_id % 8 = 2
    WHERE cp.play_status = 1
      AND cp.created_at >= '2026-05-01' AND cp.created_at < '2026-06-01'
    GROUP BY cp.user_id
    UNION ALL
    SELECT cp.user_id AS char_id, COUNT(DISTINCT cp.chapter_id) AS done_count
    FROM bi_user_chapter_play_record_3 cp
    JOIN target_chars tc ON cp.user_id = tc.char_id AND tc.char_id % 8 = 3
    WHERE cp.play_status = 1
      AND cp.created_at >= '2026-05-01' AND cp.created_at < '2026-06-01'
    GROUP BY cp.user_id
    UNION ALL
    SELECT cp.user_id AS char_id, COUNT(DISTINCT cp.chapter_id) AS done_count
    FROM bi_user_chapter_play_record_4 cp
    JOIN target_chars tc ON cp.user_id = tc.char_id AND tc.char_id % 8 = 4
    WHERE cp.play_status = 1
      AND cp.created_at >= '2026-05-01' AND cp.created_at < '2026-06-01'
    GROUP BY cp.user_id
    UNION ALL
    SELECT cp.user_id AS char_id, COUNT(DISTINCT cp.chapter_id) AS done_count
    FROM bi_user_chapter_play_record_5 cp
    JOIN target_chars tc ON cp.user_id = tc.char_id AND tc.char_id % 8 = 5
    WHERE cp.play_status = 1
      AND cp.created_at >= '2026-05-01' AND cp.created_at < '2026-06-01'
    GROUP BY cp.user_id
    UNION ALL
    SELECT cp.user_id AS char_id, COUNT(DISTINCT cp.chapter_id) AS done_count
    FROM bi_user_chapter_play_record_6 cp
    JOIN target_chars tc ON cp.user_id = tc.char_id AND tc.char_id % 8 = 6
    WHERE cp.play_status = 1
      AND cp.created_at >= '2026-05-01' AND cp.created_at < '2026-06-01'
    GROUP BY cp.user_id
    UNION ALL
    SELECT cp.user_id AS char_id, COUNT(DISTINCT cp.chapter_id) AS done_count
    FROM bi_user_chapter_play_record_7 cp
    JOIN target_chars tc ON cp.user_id = tc.char_id AND tc.char_id % 8 = 7
    WHERE cp.play_status = 1
      AND cp.created_at >= '2026-05-01' AND cp.created_at < '2026-06-01'
    GROUP BY cp.user_id
),
user_done AS (
    SELECT tc.account_id, SUM(cd.done_count) AS total_done
    FROM target_chars tc
    JOIN chapter_done cd ON tc.char_id = cd.char_id
    GROUP BY tc.account_id
    HAVING SUM(cd.done_count) > 30
),
char_info AS (
    SELECT DISTINCT ON (c.account_id) 
        c.account_id, c.gender, c.birthday
    FROM bi_vala_app_character c
    JOIN user_done ud ON c.account_id = ud.account_id
    WHERE c.deleted_at IS NULL
    ORDER BY c.account_id, c.birthday DESC NULLS LAST
),
addr_info AS (
    SELECT DISTINCT ON (account_id) account_id, login_address
    FROM account_detail_info
    WHERE login_address IS NOT NULL AND login_address != ''
    ORDER BY account_id, created_time DESC
)
SELECT 
    ud.account_id,
    ul.level,
    CASE 
        WHEN ci.gender = 1 THEN '男'
        WHEN ci.gender = 2 THEN '女'
        ELSE '未知'
    END AS gender,
    CASE 
        WHEN ci.birthday IS NOT NULL AND ci.birthday != '' 
        THEN EXTRACT(YEAR FROM AGE(TO_DATE(ci.birthday, 'YYYY-MM-DD')))
        ELSE NULL
    END AS age,
    SPLIT_PART(ai.login_address, '-', 1) AS province,
    SPLIT_PART(ai.login_address, '-', 2) AS city,
    pi.channels AS purchase_channel,
    TO_CHAR(pi.first_purchase_date, 'YYYY-MM-DD') AS first_purchase_date,
    ud.total_done AS may_done_count
FROM user_done ud
JOIN user_level ul ON ud.account_id = ul.account_id
LEFT JOIN char_info ci ON ud.account_id = ci.account_id
LEFT JOIN addr_info ai ON ud.account_id = ai.account_id
LEFT JOIN purchase_info pi ON ud.account_id = pi.account_id
ORDER BY ul.level, ud.account_id;
"""

def main():
    conn = psycopg2.connect(
        host="bj-postgres-16pob4sg.sql.tencentcdb.com",
        port=28591,
        user="ai_member",
        password=PG_PASSWORD,
        database="vala_bi"
    )
    
    df = pd.read_sql(SQL, conn)
    conn.close()
    
    # Rename columns
    df.columns = ['账号ID', '课程级别', '性别', '年龄', '省份', '城市', '购课渠道', '首次购课日期', '5月完成课时数']
    
    # Output path
    output_path = "/root/.openclaw/workspace/output/2026付费用户_5月完成30课时以上_103人.xlsx"
    os.makedirs(os.path.dirname(output_path), exist_ok=True)
    
    # Write with openpyxl for styling
    wb = Workbook()
    ws = wb.active
    ws.title = "5月完成>30课时用户"
    
    # Header style
    header_font = Font(name='微软雅黑', bold=True, size=11, color='FFFFFF')
    header_fill = PatternFill(start_color='4472C4', end_color='4472C4', fill_type='solid')
    header_alignment = Alignment(horizontal='center', vertical='center', wrap_text=True)
    thin_border = Border(
        left=Side(style='thin'),
        right=Side(style='thin'),
        top=Side(style='thin'),
        bottom=Side(style='thin')
    )
    
    # Write headers
    for col_idx, col_name in enumerate(df.columns, 1):
        cell = ws.cell(row=1, column=col_idx, value=col_name)
        cell.font = header_font
        cell.fill = header_fill
        cell.alignment = header_alignment
        cell.border = thin_border
    
    # Write data
    data_font = Font(name='微软雅黑', size=10)
    data_alignment = Alignment(vertical='center')
    
    for row_idx, row in df.iterrows():
        for col_idx, value in enumerate(row, 1):
            cell = ws.cell(row=row_idx + 2, column=col_idx, value=value if pd.notna(value) else '')
            cell.font = data_font
            cell.alignment = data_alignment
            cell.border = thin_border
    
    # Column widths
    col_widths = [12, 10, 8, 8, 16, 12, 60, 16, 16]
    for i, w in enumerate(col_widths, 1):
        ws.column_dimensions[get_column_letter(i)].width = w
    
    # Freeze header
    ws.freeze_panes = 'A2'
    
    # Auto filter
    ws.auto_filter.ref = f"A1:{get_column_letter(len(df.columns))}{len(df)+1}"
    
    # Summary sheet
    ws2 = wb.create_sheet("汇总")
    summary_data = [
        ["指标", "数值"],
        ["总人数", len(df)],
        ["仅L1", len(df[df['课程级别'] == 'L1'])],
        ["仅L2", len(df[df['课程级别'] == 'L2'])],
        ["L1+L2", len(df[df['课程级别'] == 'L1+L2'])],
        ["平均年龄", round(df['年龄'].mean(), 1)],
        ["男性", len(df[df['性别'] == '男'])],
        ["女性", len(df[df['性别'] == '女'])],
        ["性别未知", len(df[df['性别'] == '未知'])],
        ["平均5月完成课时", round(df['5月完成课时数'].mean(), 1)],
        ["最高5月完成课时", int(df['5月完成课时数'].max())],
    ]
    
    for row_idx, row_data in enumerate(summary_data, 1):
        for col_idx, value in enumerate(row_data, 1):
            cell = ws2.cell(row=row_idx, column=col_idx, value=value)
            if row_idx == 1:
                cell.font = header_font
                cell.fill = header_fill
                cell.alignment = header_alignment
            else:
                cell.font = data_font
                cell.alignment = data_alignment
            cell.border = thin_border
    
    ws2.column_dimensions['A'].width = 20
    ws2.column_dimensions['B'].width = 15
    
    wb.save(output_path)
    print(f"✅ 已导出: {output_path}")
    print(f"   共 {len(df)} 条记录")

if __name__ == "__main__":
    main()