scripts/analysis/rq1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 23,
   "outputs": [
    {
     "data": {
      "text/plain": "        tool       program_set  seed   campaign  violation  direct  \\\n0  flowdroid  droidbench-small  NONE  campaign0  VIOLATION  DIRECT   \n1  flowdroid  droidbench-small  NONE  campaign0  VIOLATION  DIRECT   \n2  flowdroid         fossdroid  NONE  campaign0  VIOLATION  DIRECT   \n\n                            config1                           config2  \\\n0  9d2e230d5859daa359c57aeb0fd090a8  3c2c6c9d6e896c028463ca607b7fce7a   \n1  4f9b34cf1904a0ff769463243bdcfe18  3c2c6c9d6e896c028463ca607b7fce7a   \n2  8f0cf1475ff0a90d31e8c21a00731724  3c2c6c9d6e896c028463ca607b7fce7a   \n\n                            partial_order  \\\n0        analyzeframeworks/TRUE/MST/FALSE   \n1  codeelimination/REMOVECODE/MST/DEFAULT   \n2             aliasflowins/TRUE/MPT/FALSE   \n\n                                             program  \n0                        ActivityLifecycle1.apk.json  \n1                        ActivityLifecycle1.apk.json  \n2  com.angrydoughnuts.android.alarmclock_15-relea...  ",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>tool</th>\n      <th>program_set</th>\n      <th>seed</th>\n      <th>campaign</th>\n      <th>violation</th>\n      <th>direct</th>\n      <th>config1</th>\n      <th>config2</th>\n      <th>partial_order</th>\n      <th>program</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>flowdroid</td>\n      <td>droidbench-small</td>\n      <td>NONE</td>\n      <td>campaign0</td>\n      <td>VIOLATION</td>\n      <td>DIRECT</td>\n      <td>9d2e230d5859daa359c57aeb0fd090a8</td>\n      <td>3c2c6c9d6e896c028463ca607b7fce7a</td>\n      <td>analyzeframeworks/TRUE/MST/FALSE</td>\n      <td>ActivityLifecycle1.apk.json</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>flowdroid</td>\n      <td>droidbench-small</td>\n      <td>NONE</td>\n      <td>campaign0</td>\n      <td>VIOLATION</td>\n      <td>DIRECT</td>\n      <td>4f9b34cf1904a0ff769463243bdcfe18</td>\n      <td>3c2c6c9d6e896c028463ca607b7fce7a</td>\n      <td>codeelimination/REMOVECODE/MST/DEFAULT</td>\n      <td>ActivityLifecycle1.apk.json</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>flowdroid</td>\n      <td>fossdroid</td>\n      <td>NONE</td>\n      <td>campaign0</td>\n      <td>VIOLATION</td>\n      <td>DIRECT</td>\n      <td>8f0cf1475ff0a90d31e8c21a00731724</td>\n      <td>3c2c6c9d6e896c028463ca607b7fce7a</td>\n      <td>aliasflowins/TRUE/MPT/FALSE</td>\n      <td>com.angrydoughnuts.android.alarmclock_15-relea...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "# CHANGE TO THE LOCATION OF YOUR CSV!\n",
    "df = pd.read_csv(\"/home/austin/sample.csv\", header=None)\n",
    "df = df.set_axis(['tool', 'program_set', 'seed', 'campaign', 'violation', 'direct', 'config1', 'config2', 'partial_order','program'], axis='columns')\n",
    "df"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Summary of Results"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "outputs": [
    {
     "data": {
      "text/plain": "tool            flowdroid  All\nprogram_set                   \nmicrobenchmark          2    2\nreal-world              1    1\nAll                     3    3",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th>tool</th>\n      <th>flowdroid</th>\n      <th>All</th>\n    </tr>\n    <tr>\n      <th>program_set</th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>microbenchmark</th>\n      <td>2</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>real-world</th>\n      <td>1</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>All</th>\n      <td>3</td>\n      <td>3</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transform = df.replace({\n",
    "    \"cats-microbenchmark\":\"microbenchmark\",\n",
    "    \"fossdroid\":\"real_world\",\n",
    "    \"droidbench\":\"microbenchmark\",\n",
    "    \"dacapo-2006\":\"real_world\",\n",
    "    \"droidbench-small\":\"microbenchmark\",\n",
    "    \"cats-small\": \"microbenchmark\",\n",
    "    \"dacapo-small\": \"real-world\"})\n",
    "pvt = pd.pivot_table(transform, index='program_set', values='partial_order', columns=['tool'], aggfunc=lambda x: len(x.unique()), margins=True)\n",
    "pvt"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Only Base Configuration Testing Results"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "outputs": [
    {
     "data": {
      "text/plain": "tool            flowdroid  All\nprogram_set                   \nmicrobenchmark          2    2\nreal-world              1    1\nAll                     3    3",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th>tool</th>\n      <th>flowdroid</th>\n      <th>All</th>\n    </tr>\n    <tr>\n      <th>program_set</th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>microbenchmark</th>\n      <td>2</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>real-world</th>\n      <td>1</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>All</th>\n      <td>3</td>\n      <td>3</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pvt = pd.pivot_table(transform[transform.seed == 'NONE'], index='program_set', values='partial_order', columns=['tool'],\n",
    "                     aggfunc=lambda x: len(x.unique()), margins=True)\n",
    "pvt"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Figure 6"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 700x300 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.close(\"all\")\n",
    "\n",
    "plt.rc(\"font\", size=14)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(7,3))\n",
    "\n",
    "to_plot = df[[\"partial_order\",\"program_set\"]]\n",
    "to_plot = to_plot.replace({\n",
    "    \"cats-microbenchmark\":\"microbenchmark\",\n",
    "    \"fossdroid\":\"real_world\",\n",
    "    \"droidbench\":\"microbenchmark\",\n",
    "    \"dacapo-2006\":\"real_world\",\n",
    "    \"droidbench-small\":\"microbenchmark\",\n",
    "    \"cats-small\": \"microbenchmark\",\n",
    "    \"dacapo-small\": \"real-world\"})\n",
    "pvt = to_plot.pivot_table(index=\"partial_order\", columns=\"program_set\", values=\"partial_order\", aggfunc=lambda x: len(x)).fillna(0)\n",
    "pvt['total'] = pvt.microbenchmark + pvt.real_world\n",
    "pvt = pvt.sort_values('total')\n",
    "pvt\n",
    "\n",
    "ax.bar(pvt.index, pvt.microbenchmark, label=\"Microbenchmark\")\n",
    "ax.bar(pvt.index, pvt.real_world, bottom=pvt.microbenchmark, label=\"Real-World\")\n",
    "\n",
    "ax.set_xticklabels(\"\")\n",
    "ax.set_yticks(range(0, 28, 4))\n",
    "ax.set_xlabel(\"Partial Order Bugs\")\n",
    "ax.set_ylabel(\"Number of Programs\")\n",
    "ax.legend()\n",
    "plt.margins(x=0.01)\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false
   }
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}