{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 使用说明：\n",
    "    1.需要修改chdir到当前目录\n",
    "    2.需要修改最后输出的文件名称\n",
    "    3.依据情况需要修改保留的列数"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "文件中所有CSV文件: ['ag888_2019.csv', 'ag888_2020.csv', 'ag888_2021.csv', 'ag888_2022.csv', 'ag888_2022_2023.csv', 'ag888_2023.csv']\n",
      "需要筛选的文件名关键字: ['_2022']\n",
      "使用新年份格式采集!!!\n",
      "筛选结果后的CSV文件: ['ag888_2022.csv', 'ag888_2022_2023.csv']\n"
     ]
    }
   ],
   "source": [
    "os.chdir('E:/data/ag')\n",
    "all_csv_files = [file for file in os.listdir('.') if file.endswith('.csv')]\n",
    "all_csv_files = sorted(all_csv_files)\n",
    "print(\"文件中所有CSV文件:\",all_csv_files)\n",
    "\n",
    "sp_chars = ['_2022']\n",
    "sp_chars = sorted(sp_chars)\n",
    "print(\"需要筛选的文件名关键字:\",sp_chars)\n",
    "\n",
    "# 设置后面数据的采集对于的行数# 用 \"old_type\" 或者 \"new_type\" 区分\n",
    "if all(char in ['_2019','_2020','_2021'] for char in sp_chars):\n",
    "    year_type = 'old_type'\n",
    "    print(\"使用旧年份格式采集!!!\")\n",
    "elif all(char in ['_2022','_2023'] for char in sp_chars):\n",
    "    year_type = 'new_type' \n",
    "    print(\"使用新年份格式采集!!!\")\n",
    "else:\n",
    "    print(\"文件夹中CSV没有相关年份的数据或者新旧年份混用!!!\")\n",
    "\n",
    "csv_files = [file for file in all_csv_files if any(sp_char in file for sp_char in sp_chars)]\n",
    "print(\"筛选结果后的CSV文件:\",csv_files)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame()\n",
    "for f in csv_files:\n",
    "    if year_type == 'old_type':\n",
    "        df_temp = pd.read_csv(\n",
    "            f,\n",
    "            usecols=[1, 2, 3, 4, 8, 13, 14, 15, 16],\n",
    "            names=[\n",
    "                \"统一代码\",\n",
    "                \"合约代码\",\n",
    "                \"时间\",\n",
    "                \"最新\",\n",
    "                \"成交量\",\n",
    "                \"买一价\",\n",
    "                \"卖一价\",\n",
    "                \"买一量\",\n",
    "                \"卖一量\",\n",
    "            ],\n",
    "            skiprows=1,\n",
    "            encoding=\"utf-8\",\n",
    "        )\n",
    "    elif year_type == 'new_type':\n",
    "            df_temp = pd.read_csv(\n",
    "        f,\n",
    "        usecols=[0, 1, 2, 5, 12, 21, 22, 23, 24, 25, 26, 44],\n",
    "        names=[\n",
    "            \"交易日\",\n",
    "            \"统一代码\",\n",
    "            \"合约代码\",\n",
    "            \"最新价\",\n",
    "            \"数量\",\n",
    "            \"最后修改时间\",\n",
    "            \"最后修改毫秒\",\n",
    "            \"申买价一\",\n",
    "            \"申买量一\",\n",
    "            \"申卖价一\",\n",
    "            \"申卖量一\",\n",
    "            \"业务日期\",\n",
    "        ],\n",
    "        skiprows=1,\n",
    "        encoding=\"utf-8\",\n",
    "    )\n",
    "\n",
    "    # df_temp = pd.read_csv(f, usecols=[0,5], names=[\n",
    "    #                  'datetime',  'volume'])\n",
    "    df = pd.concat([df, df_temp])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 查看数据的头部和尾部：head()、tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.reset_index(drop=True, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 查看dataframe的基本情况\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 等比复权,先不考虑\n",
    "# df['复权因子'] = df['卖一价'].shift() / df['买一价']\n",
    "df['复权因子'] = np.where(df['合约代码'] != df['合约代码'].shift(), df['卖一价'].shift() / df['买一价'], 1)\n",
    "df['复权因子'] = df['复权因子'].fillna(1)\n",
    "# df['复权因子'].loc[0] = 1\n",
    "df['买一价_adj'] = df['买一价'] * df['复权因子'].cumprod()\n",
    "df['卖一价_adj'] = df['卖一价'] * df['复权因子'].cumprod()\n",
    "df['最新_adj'] = df['最新'] * df['复权因子'].cumprod()\n",
    "# df['low_adj'] = df['low'] * adjust.cumprod()\n",
    "# df['high_adj'] = df['high'] * adjust.cumprod()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 等差复权\n",
    "df['复权因子'] = np.where(df['合约代码'] != df['合约代码'].shift(), df['申卖价一'].shift() - df['申买价一'], 0)\n",
    "df['复权因子'] = df['复权因子'].fillna(0)\n",
    "# df['复权因子'].loc[0] = 1\n",
    "df['申买价一_adj'] = df['申买价一'] + df['复权因子'].cumsum()\n",
    "df['申卖价一_adj'] = df['申卖价一'] + df['复权因子'].cumsum()\n",
    "df['最新价_adj'] = df['最新价'] + df['复权因子'].cumsum()\n",
    "# df['low_adj'] = df['low'] + df['复权因子'].cumsum()\n",
    "# df['high_adj'] = df['high'] + df['复权因子'].cumsum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 查找换期需要复权的索引\n",
    "non_zero_indices = df[df['复权因子'] != 0].index\n",
    "print(non_zero_indices)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 查看未调整买价、卖价和最新价的数据\n",
    "df.loc[non_zero_indices[0]-5:non_zero_indices[0]+5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 将调整后的数值替换原来的值\n",
    "df['申买价一'] = df['申买价一_adj']\n",
    "df['申卖价一'] = df['申卖价一_adj']\n",
    "df['最新价'] = df['最新价_adj']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 查看调整买价、卖价和最新价的数据\n",
    "df.loc[non_zero_indices[0]-5:non_zero_indices[0]+5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 删除多余的值\n",
    "del df['复权因子']\n",
    "del df['申买价一_adj']\n",
    "del df['申卖价一_adj']\n",
    "del df['最新价_adj']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[non_zero_indices[0]-5:non_zero_indices[0]+5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv('./ag888_2022_2023.csv', index=False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "orderflow",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.9"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}