97 lines
2.9 KiB
Plaintext
97 lines
2.9 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 1,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"相同题目已标注: 000100 031239\n",
|
|
"关联题目已标注: 000466 030610\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"import os,re,json\n",
|
|
"\n",
|
|
"filename = \"临时文件/相似题目.txt\"\n",
|
|
"\n",
|
|
"# 读取题库数据并转换为字典\n",
|
|
"with open(r\"../题库0.3/Problems.json\",\"r\",encoding = \"utf8\") as f:\n",
|
|
" database = f.read()\n",
|
|
"pro_dict = json.loads(database)\n",
|
|
"\n",
|
|
"# 读取已分类的相似文件列表\n",
|
|
"with open(filename,\"r\",encoding = \"utf8\") as f:\n",
|
|
" similar_text = \"\\n\"+f.read()\n",
|
|
"\n",
|
|
"similar_types = re.findall(r\"\\n[\\d]\\.[\\d]{4}[\\s]*([srSRnN ])[\\s]*\\n\",similar_text)\n",
|
|
"similar_problems = re.findall(r\"\\n([\\d]{6}) \",similar_text)\n",
|
|
"\n",
|
|
"if len(similar_types) * 2 == len(similar_problems):\n",
|
|
" for i in similar_types:\n",
|
|
" id1 = similar_problems.pop(0)\n",
|
|
" id2 = similar_problems.pop(0)\n",
|
|
" if i.upper() == \"S\":\n",
|
|
" if not id2 in pro_dict[id1][\"same\"]:\n",
|
|
" pro_dict[id1][\"same\"].append(id2)\n",
|
|
" if not id1 in pro_dict[id2][\"same\"]:\n",
|
|
" pro_dict[id2][\"same\"].append(id1)\n",
|
|
" print(\"相同题目已标注:\",id1,id2)\n",
|
|
" elif i.upper() == \"R\":\n",
|
|
" if not id2 in pro_dict[id1][\"related\"]:\n",
|
|
" pro_dict[id1][\"related\"].append(id2)\n",
|
|
" if not id1 in pro_dict[id2][\"related\"]:\n",
|
|
" pro_dict[id2][\"related\"].append(id1)\n",
|
|
" print(\"关联题目已标注:\",id1,id2)\n",
|
|
"\n",
|
|
"else:\n",
|
|
" print(\"相似程度数据:\",len(similar_types),\"个, 相似题目:\",len(similar_problems),\"题. 数据有问题, 请检查.\")\n",
|
|
"\n",
|
|
"\n",
|
|
"# 将题库字典转换为json文件并保存至原位\n",
|
|
"database = json.dumps(pro_dict,indent=4,ensure_ascii=False)\n",
|
|
"with open(r\"../题库0.3/Problems.json\",\"w\",encoding = \"utf8\") as f:\n",
|
|
" f.write(database)"
|
|
]
|
|
},
|
|
{
|
|
"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": "3.8.15"
|
|
},
|
|
"orig_nbformat": 4,
|
|
"vscode": {
|
|
"interpreter": {
|
|
"hash": "d311ffef239beb3b8f3764271728f3972d7b090c974f8e972fcdeedf230299ac"
|
|
}
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 2
|
|
}
|