reports/RQ1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# RQ 1 (Frequency)\n",
    "\n",
    "> How often does unsafe code appear explicitly in Rust crates?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: plotly in /opt/conda/lib/python3.7/site-packages (4.9.0)\r\n",
      "Requirement already satisfied: retrying>=1.3.3 in /opt/conda/lib/python3.7/site-packages (from plotly) (1.3.3)\r\n",
      "Requirement already satisfied: six in /opt/conda/lib/python3.7/site-packages (from plotly) (1.14.0)\r\n"
     ]
    }
   ],
   "source": [
    "!pip install plotly"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import libraries\n",
    "import itertools\n",
    "import functools\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn import tree\n",
    "\n",
    "import utils\n",
    "import top_crates\n",
    "from whitelists import *\n",
    "from drawing import *\n",
    "\n",
    "# Configure pyplot: enlarge plots \n",
    "plt.rcParams['figure.figsize'] = [15, 8]\n",
    "# Configure pandas: show all columns when printing a dataframe\n",
    "pd.set_option('display.max_columns', None)\n",
    "# Configure pandas: show more rows when printing a dataframe\n",
    "pd.set_option('display.min_rows', 20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load tables\n",
    "selected_builds = utils.load_selected_builds()\n",
    "unsafe_blocks = utils.load_unsafe_blocks()\n",
    "traits = utils.load_selected_traits()\n",
    "traits_impls = utils.load_selected_impl_definitions()\n",
    "function_definitions = utils.load_selected_function_definitions()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Validation\n",
    "assert len(selected_builds) > 20000\n",
    "assert selected_builds.build.is_unique"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Preprocessing\n",
    "crates = pd.DataFrame({\n",
    "    \"crate\": selected_builds.crate,\n",
    "    \"crate_hash\": selected_builds.crate_hash,\n",
    "    \"package\": selected_builds.package.map(lambda x: str(x).lower().replace(\"-\", \"_\")),\n",
    "})\n",
    "user_written_unsafe_blocks = unsafe_blocks[unsafe_blocks.check_mode == \"UnsafeBlockUserProvided\"]\n",
    "unsafe_function_definitions = function_definitions[function_definitions.unsafety == \"Unsafe\"]\n",
    "unsafe_traits = traits[traits.unsafety == \"Unsafe\"]\n",
    "unsafe_traits_impls = traits_impls[traits_impls.unsafety == \"Unsafe\"]\n",
    "\n",
    "# Validation\n",
    "assert len(user_written_unsafe_blocks) > 0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Query 1\n",
    "\n",
    "> Table 1 shows in both absolute and relative numbers how many crates contain unsafe code, and which unsafe features they use (our first query)\n",
    "\n",
    "#### Results from the paper for comparison (Table 1)\n",
    "\n",
    "| Unsafe Feature| #crates      |    %  |\n",
    "| ------------- |--------------|-------|\n",
    "| None          | 24,360       |  76.4 |\n",
    "| Any           | 7,507        |  23.6 |\n",
    "| Blocks        | 6,414        |  20.1 |\n",
    "| Function Declarations| 4,287        |  13.5 |\n",
    "| Trait Implementations| 1,591        |  5.0 |\n",
    "| Trait Declarations| 280        |  0.9 |"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of crates: 31867\n",
      "has_no_unsafe             24360\n",
      "has_unsafe                 7507\n",
      "has_unsafe_blocks          6414\n",
      "has_unsafe_functions       4287\n",
      "has_unsafe_trait_impls     1591\n",
      "has_unsafe_traits           280\n",
      "dtype: int64\n",
      "has_no_unsafe             76.442715\n",
      "has_unsafe                23.557285\n",
      "has_unsafe_blocks         20.127405\n",
      "has_unsafe_functions      13.452788\n",
      "has_unsafe_trait_impls     4.992626\n",
      "has_unsafe_traits          0.878652\n",
      "dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "join_key = \"crate_hash\"\n",
    "crates[\"has_unsafe_blocks\"] = crates[join_key].isin(set(user_written_unsafe_blocks[join_key].unique()))\n",
    "crates[\"has_unsafe_functions\"] = crates[join_key].isin(set(unsafe_function_definitions[join_key].unique()))\n",
    "crates[\"has_unsafe_traits\"] = crates[join_key].isin(set(unsafe_traits[join_key].unique()))\n",
    "crates[\"has_unsafe_trait_impls\"] = crates[join_key].isin(set(unsafe_traits_impls[join_key].unique()))\n",
    "\n",
    "feature_names = [\n",
    "    \"has_unsafe_blocks\",\n",
    "    \"has_unsafe_functions\",\n",
    "    \"has_unsafe_traits\",\n",
    "    \"has_unsafe_trait_impls\",\n",
    "]\n",
    "\n",
    "crates[\"has_unsafe\"] = functools.reduce(lambda x, y: x | y, [crates[f] for f in feature_names])\n",
    "feature_names.append(\"has_unsafe\")\n",
    "\n",
    "crates[\"has_no_unsafe\"] = ~crates[\"has_unsafe\"]\n",
    "feature_names.append(\"has_no_unsafe\")\n",
    "\n",
    "print(\"Number of crates: {}\".format(len(crates)))\n",
    "\n",
    "feature_freq = crates[feature_names].apply(sum).sort_values(ascending=False)\n",
    "feature_freq.plot.bar(subplots=True, color=\"blue\", title=\"Features per crate\")\n",
    "print(feature_freq)\n",
    "print(feature_freq / len(crates) * 100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Query 2\n",
    "\n",
    "> while Fig. 2 shows the relative amount of statements in unsafe blocks and functions in both all crates and, for readability, crates that contain at least one unsafe statement (our second query).\n",
    "\n",
    "#### Results from the paper for comparison (Fig. 2)\n",
    "\n",
    "![Fig. 2](pics/unsafe_proportion.png)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load tables\n",
    "selected_function_sizes = utils.load_selected_function_sizes()\n",
    "selected_build_sizes = utils.load_selected_build_sizes()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "assert len(selected_build_sizes.query('statement_count == 0')) == 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The number of crates with at least one user written unsafe MIR statement: 6777\n",
      "Percentage: 21.266513948598863\n"
     ]
    }
   ],
   "source": [
    "crates_with_unsafe_statements_count = len(selected_build_sizes.query('user_unsafe_statement_count > 0'))\n",
    "print(\"The number of crates with at least one user written unsafe MIR statement:\",\n",
    "      crates_with_unsafe_statements_count)\n",
    "print(\"Percentage:\", 100 * crates_with_unsafe_statements_count / len(crates))\n",
    "selected_build_sizes[\"ratio\"] = selected_build_sizes.apply(\n",
    "    lambda row: 100 * row['user_unsafe_statement_count']/row['statement_count'], axis=1)\n",
    "build_sizes_with_unsafe = selected_build_sizes.query('user_unsafe_statement_count > 0')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "selected_builds = utils.load_selected_builds()\n",
    "unsafe_selected_build_hashes = set(selected_build_sizes['build_crate_hash'])\n",
    "selected_build_hashes = set(selected_builds['crate_hash'])\n",
    "assert unsafe_selected_build_hashes.issubset(selected_build_hashes)\n",
    "safe_selected_build_hashes = selected_build_hashes - unsafe_selected_build_hashes\n",
    "safe_selected_build_sizes = []\n",
    "for (_, build) in selected_builds.iterrows():\n",
    "    if build['crate_hash'] in unsafe_selected_build_hashes:\n",
    "        continue\n",
    "    safe_selected_build_sizes.append({\n",
    "        'package_name': build['package'],\n",
    "        'package_version': build['version'],\n",
    "        'crate_name': build['crate'],\n",
    "        'build_crate_hash': build['crate_hash'],\n",
    "        'edition': build['edition'],\n",
    "        'statement_count': 1,\n",
    "        'unsafe_statement_count': 0,\n",
    "        'user_unsafe_statement_count': 0,\n",
    "        'ratio': 0.0,\n",
    "    })\n",
    "all_selected_build_sizes = selected_build_sizes.append(safe_selected_build_sizes)\n",
    "assert len(all_selected_build_sizes) == len(selected_build_sizes) + len(safe_selected_build_sizes)\n",
    "assert len(selected_builds) == len(all_selected_build_sizes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Proportion of all crates for which unsafe statement ratio is <= 10%: 92.28982960429285\n",
      "Proportion of crates with at least one unsafe statement for which unsafe statement ratio is <= 20%: 75.38733953076583\n"
     ]
    }
   ],
   "source": [
    "print('Proportion of all crates for which unsafe statement ratio is <= 10%:',\n",
    "      100 * len(all_selected_build_sizes.query('ratio <= 10'))/len(all_selected_build_sizes))\n",
    "print('Proportion of crates with at least one unsafe statement for which unsafe statement ratio is <= 20%:',\n",
    "      100 * len(build_sizes_with_unsafe.query('ratio <= 20'))/len(build_sizes_with_unsafe))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = all_selected_build_sizes.ratio.plot.hist(\n",
    " cumulative=True,\n",
    " density=1,\n",
    " bins=len(build_sizes_with_unsafe),\n",
    " xlim=(0, 100),\n",
    " ylim=(0, 1),\n",
    " histtype=\"step\",\n",
    " linewidth=4,\n",
    " fontsize=26,\n",
    ")\n",
    "ax2 = build_sizes_with_unsafe.ratio.plot.hist(\n",
    " cumulative=True,\n",
    " density=1,\n",
    " bins=len(build_sizes_with_unsafe),\n",
    " xlim=(0, 100),\n",
    " ylim=(0, 1),\n",
    " histtype=\"step\",\n",
    " linewidth=4,\n",
    " fontsize=26,\n",
    ")\n",
    "ax.title.set_size(20)\n",
    "ax.set_xlabel(\"Proportion of unsafe statements in a crate\", fontsize=32)\n",
    "ax.set_ylabel(\"Percentage of crates\", fontsize=32)\n",
    "ax.set_axisbelow(True)\n",
    "ax.set_yticklabels(['{:,.0%}'.format(x) for x in ax.get_yticks()])\n",
    "ax.set_xticklabels(['{:,.0%}'.format(x/100) for x in ax.get_xticks()])\n",
    "ax.set_xticks(range(0, 100, 10), minor=True)\n",
    "ax.set_xticks(range(0, 100, 20))\n",
    "ax.grid(True, linestyle='dotted')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}