{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import string"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def find_comment(text):\n",
    "    start = text.find('<')\n",
    "    if start > -1:\n",
    "        end = text.find('>', start + 1)\n",
    "        if end > -1:\n",
    "            return start, end\n",
    "    return None, None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def decomment_no_re(text):\n",
    "    s, e = find_comment(text)\n",
    "    while s is not None:\n",
    "        text = text[:s] + text[e+1:]\n",
    "        s, e = find_comment(text)\n",
    "    return text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def find_expansion_mark(text):\n",
    "    start = text.find(':')\n",
    "    if start > -1:\n",
    "        mid = text.find(':', start + 1)\n",
    "        if mid > -1:\n",
    "            end = text.find(':', mid + 1)\n",
    "            if end > -1:\n",
    "                length = int(text[start+1:mid])\n",
    "                times = int(text[mid+1:end])\n",
    "                if length > start:\n",
    "                    length = start\n",
    "                return start, end, length, times\n",
    "    return None, None, None, None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def expand_no_re(text, expansion_limit=None):\n",
    "    i = 1\n",
    "    s, e, l, n = find_expansion_mark(text)\n",
    "    while s and (expansion_limit is None or (expansion_limit is not None and i <= expansion_limit)):\n",
    "        text = text[:s-l] + text[s-l:s] * n + text[e+1:]\n",
    "        s, e, l, n = find_expansion_mark(text)\n",
    "        i += 1\n",
    "    return text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_p_t = [l.strip() for l in open('../../data/05-instructions.txt')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_p = [decomment_no_re(l) for l in data_p_t]\n",
    "data = [expand_no_re(l) for l in data_p]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "149043"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len([c for c in ''.join(data_p) if c not in string.whitespace])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "302266"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len([c for c in ''.join(data) if c not in string.whitespace])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_s_p_t = open('../../data/05-instructions.txt').read()\n",
    "data_s_p = decomment_no_re(data_s_p_t)\n",
    "data_s = expand_no_re(data_s_p)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "149043"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(1 for c in data_s_p if c not in string.whitespace)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "302266"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(1 for c in data_s if c not in string.whitespace)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "23.7 ms ± 1.79 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "data_p = [decomment_no_re(l) for l in data_p_t]\n",
    "data = [expand_no_re(l) for l in data_p]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "360 ms ± 6.67 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "data_s_p = decomment_no_re(data_s_p_t)\n",
    "data_s = expand_no_re(data_s_p)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}