{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "from collections import defaultdict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z']\n"
     ]
    }
   ],
   "source": [
    "letters = [chr(letter) for letter in range(ord(\"a\"), ord(\"z\") + 1)]\n",
    "print(letters)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "370105\n"
     ]
    }
   ],
   "source": [
    "# The file `words_alpha.txt` has been obtained from https://github.com/dwyl/english-words \n",
    "with open(\"./reasoning_gym/data/words_alpha.txt\") as f:\n",
    "    words = f.read().splitlines()\n",
    "print(len(words))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "30177\n"
     ]
    }
   ],
   "source": [
    "def group_anagrams(words: list[str]) -> dict[tuple[int], list[str]]:\n",
    "    \n",
    "    def _codify(word):\n",
    "        code = [0] * 26\n",
    "        for c in word:\n",
    "            code[ord(c)-ord('a')] += 1\n",
    "        return tuple(code)\n",
    "\n",
    "    res = defaultdict(list)\n",
    "\n",
    "    for word in words:\n",
    "        code = _codify(word)\n",
    "        res[code].append(word)\n",
    "    return res\n",
    "\n",
    "anagrams = group_anagrams(words)\n",
    "anagrams = {k: v for k, v in anagrams.items() if len(v) > 1} # only keep anagrams with more than 1 word\n",
    "print(len(anagrams))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"./reasoning_gym/data/anagrams.jsonl\", \"w\") as f:\n",
    "    for counts, words in anagrams.items():\n",
    "        letter_counts = {letter: count for letter, count in zip(letters, counts)}\n",
    "        f.write(json.dumps({\"letter_counts\": letter_counts, \"words\": words}) + \"\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "reasoning_gym",
   "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": "3.11.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}