import fs from 'node:fs';
import wordlistEnglish from 'wordlist-english';

const validTwo = new Set([
  'am', 'an', 'as', 'at', 'be', 'by', 'do', 'go', 'he', 'hi',
  'if', 'in', 'is', 'it', 'me', 'my', 'no', 'of', 'oh', 'on',
  'or', 'so', 'to', 'up', 'us', 'we'
]);

// Build a clean, high-frequency recognized vocabulary for client-side unlimited/custom puzzles
const vocabulary = new Set();
for (const word of wordlistEnglish.english) {
  const w = word.toLowerCase().trim();
  if (w.length === 2 && validTwo.has(w)) {
    vocabulary.add(w);
  } else if (w.length >= 3 && w.length <= 16 && /^[a-z]+$/.test(w)) {
    vocabulary.add(w);
  }
}

console.log('Total vocabulary words:', vocabulary.size);

// Pack into space-separated string or array
const sortedWords = [...vocabulary].sort();
const lexiconContent = `// Client-side recognized English words for solver & custom puzzles
// Loaded as a Set for O(1) instantaneous lookup
const RAW_LEXICON = ${JSON.stringify(sortedWords)};

export const DICTIONARY_SET = new Set(RAW_LEXICON);

// Prefix set for lightning fast tree-pruning search (1 to 6 chars)
export const PREFIX_SET = new Set();
for (const word of RAW_LEXICON) {
  for (let len = 1; len <= Math.min(word.length, 6); len++) {
    PREFIX_SET.add(word.slice(0, len));
  }
}

export function isWordValid(word) {
  return DICTIONARY_SET.has(word.toLowerCase());
}
`;

fs.writeFileSync('src/data/lexicon.js', lexiconContent, 'utf8');
console.log('Saved src/data/lexicon.js successfully!');