Refining uMatrix rules

Refining_uMatrix_rules.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Export rule list from uMatrix extension to a file\n",
    "path_to_file = '/Users/username/Downloads/my-umatrix-rules.txt'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# reading the file with rules\n",
    "with open(path_to_file, 'r') as fh:\n",
    "    list_of_rules = [\n",
    "        tuple(line.rstrip('\\n').split(' '))\n",
    "        for line in fh\n",
    "    ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# top 50 rules are default \n",
    "list_of_rules = list_of_rules[50:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# lets put all custom site rules in to DataFrame\n",
    "import pandas as pd\n",
    "df = pd.DataFrame(list_of_rules, columns=('orig', 'source', 'scope', 'rule'))\n",
    "df = df[df.rule=='allow']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# same source domains are whitelised over and over\n",
    "df.source.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Collapsing all rules irregardless of origin\n",
    "set_of_uni_rules = {\n",
    "    ('*', source, scope, rule) if rule!='block' else (orig, source, scope, rule)\n",
    "    for orig, source, scope, rule in list_of_rules\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# lets aggregate base domains for some sites\n",
    "set_of_unique_rules = set()\n",
    "for full_rule in set_of_uni_rules:\n",
    "    orig,domain,cont,rule = full_rule\n",
    "    dms = domain.split('.')\n",
    "    if len(dms) > 2 and dms[0] != 'www' and dms[-1] in ('com', 'net'):\n",
    "        new_domain = '.'.join(dms[-2:])\n",
    "        print(\"Replacing domain '{}' for '{}'\".format(domain,new_domain))\n",
    "        set_of_unique_rules.add((orig, new_domain, cont, rule))\n",
    "    else:\n",
    "        set_of_unique_rules.add(full_rule)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Double check if rules make sense\n",
    "set_of_unique_rules"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# write to file and import into extension\n",
    "with open(\"/tmp/new-umatrix-rules.txt\", 'w+') as wfh:\n",
    "    wfh.write('\\n'.join(' '.join(t) for t in set_of_unique_rules))"
   ]
  }
 ],
 "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}