完成作业7第二题的代码

Assignments/Assignment7/src/Problem2/.ipynb_checkpoints/main-checkpoint.ipynb

@@ -0,0 +1,195 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "30712dff-7471-4a6f-a464-aa4305036b9a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pymysql\n",
+    "import random\n",
+    "from tqdm import tqdm\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "\n",
+    "N = 20000"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "480e12b1-02a4-4e89-9a13-1b147b075372",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "db = pymysql.connect(\n",
+    "    host='127.0.0.1', user='kejingfan', \n",
+    "    password='KJF2811879', database='TESTDB'\n",
+    ")\n",
+    "cursor = db.cursor()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "91fb704d-4104-4cf1-86cc-5ef7a2f06601",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "sql_statements = [\n",
+    "    \"SET profiling = 1;\",\n",
+    "    \n",
+    "    \"DROP TABLE IF EXISTS passengers_no_index;\",\n",
+    "    \"DROP TABLE IF EXISTS passengers_with_index;\",\n",
+    "    \"\"\"\n",
+    "    CREATE TABLE passengers_no_index (\n",
+    "        ID BIGINT PRIMARY KEY,\n",
+    "        `Name` VARCHAR (255) NOT NULL,\n",
+    "        Phone_number BIGINT NOT NULL,\n",
+    "        `Password` VARCHAR (255) NOT NULL,\n",
+    "        CHECK (ID REGEXP '^\\\\\\\\d{18}$'),\n",
+    "        CHECK (Phone_number REGEXP '^\\\\\\\\d{11}$')\n",
+    "    );\n",
+    "    \"\"\",\n",
+    "    \"\"\"\n",
+    "    CREATE TABLE passengers_with_index (\n",
+    "        ID BIGINT PRIMARY KEY,\n",
+    "        `Name` VARCHAR (255) NOT NULL,\n",
+    "        Phone_number BIGINT NOT NULL,\n",
+    "        `Password` VARCHAR (255) NOT NULL,\n",
+    "        CHECK (ID REGEXP '^\\\\\\\\d{18}$'),\n",
+    "        CHECK (Phone_number REGEXP '^\\\\\\\\d{11}$')\n",
+    "    );\n",
+    "    \"\"\",\n",
+    "    \"CREATE INDEX idx_phone_number ON passengers_with_index (Phone_number);\",\n",
+    "]\n",
+    "\n",
+    "for sql in sql_statements:\n",
+    "    cursor.execute(sql)\n",
+    "db.commit()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "100a5382",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "操作数据表 passengers_no_index ：\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      " 71%|████████████████████████████████████████████████████████████████████████████████                                 | 14166/20000 [02:56<01:19, 73.12it/s]"
+     ]
+    }
+   ],
+   "source": [
+    "id_list = random.sample(range(100000000000000000, 1000000000000000000), N)\n",
+    "phone_number_list = random.sample(range(10000000000, 20000000000), N)\n",
+    "\n",
+    "insert_times = {\n",
+    "    'passengers_no_index': [],\n",
+    "    'passengers_with_index': []\n",
+    "}\n",
+    "query_times = {\n",
+    "    'passengers_no_index': [],\n",
+    "    'passengers_with_index': []\n",
+    "}\n",
+    "\n",
+    "for table_name in ['passengers_no_index', 'passengers_with_index']:\n",
+    "    print(f\"操作数据表 {table_name} ：\")\n",
+    "    insert_sql = f'''\n",
+    "        INSERT INTO {table_name} (ID, `Name`, Phone_number, `Password`)\n",
+    "        VALUES (%s, %s, %s, %s);\n",
+    "    '''\n",
+    "    query_sql = f'''\n",
+    "        SELECT * FROM {table_name}\n",
+    "        WHERE Phone_number = %s;\n",
+    "    '''\n",
+    "    \n",
+    "    for i in tqdm(range(N)):\n",
+    "        cursor.execute(insert_sql, (id_list[i], 'kejingfan', phone_number_list[i], 'password'))\n",
+    "        db.commit()\n",
+    "        cursor.execute(\"SHOW PROFILES;\")\n",
+    "        profile = cursor.fetchall()[-2]\n",
+    "        if \"INSERT INTO\" in profile[2]:\n",
+    "            insert_times[table_name].append(profile[1])\n",
+    "\n",
+    "        cursor.execute(query_sql, (phone_number_list[random.randint(0, i)],))\n",
+    "        cursor.execute(\"SHOW PROFILES;\")\n",
+    "        profile = cursor.fetchall()[-1]\n",
+    "        if \"SELECT * FROM\" in profile[2]:\n",
+    "            query_times[table_name].append(profile[1])\n",
+    "\n",
+    "cursor.close()\n",
+    "db.close()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3783d8dd",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def get_average_per_n(data, n):\n",
+    "    return [np.mean(data[i:i + n]) for i in range(0, len(data), n)]\n",
+    "\n",
+    "avg_insert_times_no_index = get_average_per_n(insert_times['passengers_no_index'], 100)\n",
+    "avg_insert_times_with_index = get_average_per_n(insert_times['passengers_with_index'], 100)\n",
+    "avg_query_times_no_index = get_average_per_n(query_times['passengers_no_index'], 100)\n",
+    "avg_query_times_with_index = get_average_per_n(query_times['passengers_with_index'], 100)\n",
+    "\n",
+    "plt.figure(figsize=(14, 7))\n",
+    "\n",
+    "plt.subplot(2, 1, 1)\n",
+    "plt.plot(range(len(avg_insert_times_no_index)), avg_insert_times_no_index, label='No Index Insert Time')\n",
+    "plt.plot(range(len(avg_insert_times_with_index)), avg_insert_times_with_index, label='With Index Insert Time')\n",
+    "plt.xlabel('Number of Insertions (in hundreds)')\n",
+    "plt.ylabel('Time (s)')\n",
+    "plt.title('Average Insert Time vs Number of Insertions')\n",
+    "plt.legend()\n",
+    "\n",
+    "plt.subplot(2, 1, 2)\n",
+    "plt.plot(range(len(avg_query_times_no_index)), avg_query_times_no_index, label='No Index Query Time')\n",
+    "plt.plot(range(len(avg_query_times_with_index)), avg_query_times_with_index, label='With Index Query Time')\n",
+    "plt.xlabel('Number of Queries (in hundreds)')\n",
+    "plt.ylabel('Time (s)')\n",
+    "plt.title('Average Query Time vs Number of Queries')\n",
+    "plt.legend()\n",
+    "\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "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": "3.11.9"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

Assignments/Assignment7/src/Problem2/db.sql

@@ -0,0 +1,24 @@
+DROP TABLE IF EXISTS passengers_no_index;
+DROP TABLE IF EXISTS passengers_with_index;
+
+CREATE TABLE passengers_no_index (
+    ID BIGINT PRIMARY KEY,
+    `Name` VARCHAR(255) NOT NULL,
+    Phone_number BIGINT UNIQUE NOT NULL,
+    `Password` VARCHAR(255) NOT NULL,
+    CHECK (ID REGEXP '^\d{18}$'),
+    CHECK (Phone_number REGEXP '^\d{11}$')
+);
+
+CREATE TABLE passengers_with_index (
+    ID BIGINT PRIMARY KEY,
+    `Name` VARCHAR(255) NOT NULL,
+    Phone_number BIGINT UNIQUE NOT NULL,
+    `Password` VARCHAR(255) NOT NULL,
+    CHECK (ID REGEXP '^\d{18}$'),
+    CHECK (Phone_number REGEXP '^\d{11}$')
+);
+
+CREATE INDEX idx_phone_number ON passengers_with_index (Phone_number);
+
+SET profiling = 1;

Assignments/Assignment7/src/Problem2/requirements.txt

@@ -0,0 +1,14 @@
+blinker==1.8.1
+cffi==1.16.0
+click==8.1.7
+configparser==7.0.0
+cryptography==42.0.6
+Flask==3.0.3
+itsdangerous==2.2.0
+Jinja2==3.1.3
+MarkupSafe==2.1.5
+pycparser==2.22
+PyMySQL==1.1.0
+Werkzeug==3.0.2
+jupyterlab
+notebook