{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2021届杨浦区基础考.tex\n",
      "填空题\n",
      "(011988)\n",
      "(011989)\n",
      "(011990)\n",
      "(011991)\n",
      "(011992)\n",
      "(011993)\n",
      "(011994)\n",
      "(011995)\n",
      "(011996)\n",
      "(011997)\n",
      "(011998)\n",
      "(011999)\n",
      "选择题\n",
      "(012000)\n",
      "(012001)\n",
      "(012002)\n",
      "(012003)\n",
      "解答题\n",
      "(012004)\n",
      "(012005)\n",
      "(012006)\n",
      "(012007)\n",
      "(012008)\n",
      "\n",
      "\n",
      "\n",
      "2022届杨浦区基础考.tex\n",
      "填空题\n",
      "(012009)\n",
      "(012010)\n",
      "(012011)\n",
      "(012012)\n",
      "(012013)\n",
      "(012014)\n",
      "(012015)\n",
      "(012016)\n",
      "(012017)\n",
      "(012018)\n",
      "(012019)\n",
      "(012020)\n",
      "选择题\n",
      "(012021)\n",
      "(012022)\n",
      "(012023)\n",
      "(012024)\n",
      "解答题\n",
      "(012025)\n",
      "(012026)\n",
      "(012027)\n",
      "(012028)\n",
      "(012029)\n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import os,re\n",
    "\"---此处输入文件夹名---\"\n",
    "directory = r\"C:\\Users\\weiye\\Documents\\wwy sync\\23届\\其他讲义与试卷\"\n",
    "\"---文件夹名输入结束---\"\n",
    "\n",
    "filelist = [filename for filename in os.listdir(directory) if \".tex\" in filename]\n",
    "\n",
    "output = \"\"\n",
    "\n",
    "for filename in filelist:\n",
    "    print(filename)\n",
    "    output += filename + \"\\n\"\n",
    "    with open(os.path.join(directory,filename),\"r\",encoding = \"u8\") as f:\n",
    "        try:\n",
    "            data = re.findall(r\"\\\\begin{document}([\\s\\S]*?)\\\\end{document}\",f.read())[0]\n",
    "            data = data.replace(r\"\\section\",\"endsecbeginsec\") + \"endsec\"\n",
    "            sectionlist = re.findall(r\"beginsec([\\s\\S]*?)endsec\",data)\n",
    "            for sec in sectionlist:\n",
    "                secname = re.findall(r\"{([\\S]*)}\",sec)[0]\n",
    "                output += secname + \"\\n\"\n",
    "                print(secname)\n",
    "                for id in re.findall(r\"\\(\\d{6}\\)\",sec):\n",
    "                    print(id)\n",
    "                    output += id + \"\\n\"\n",
    "        except:\n",
    "            pass    \n",
    "    output += \"\\n\\n\"\n",
    "    print(\"\\n\\n\")\n",
    "\n",
    "with open(os.path.join(directory,\"题号清点.txt\"),\"w\",encoding = \"u8\") as f:\n",
    "    f.write(output + \"\\n\\n\\n以下题号不含括号\\n\\n\\n\" + output.replace(\"(\",\"\").replace(\")\",\"\"))\n",
    "\n",
    "                \n",
    "\n",
    "\n",
    "                \n",
    "            \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.8.8 ('base')",
   "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": "3.8.8"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "d311ffef239beb3b8f3764271728f3972d7b090c974f8e972fcdeedf230299ac"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}