/**
 * Curriculum ordering utilities: reading the order files that define which
 * characters, vocabulary, and radicals are taught, and in what sequence.
 *
 * Order files live under data/lang/{targetLang}/order/.
 */

import { parse } from '@std/csv/parse'
import { DATA_ROOT } from './fs.ts'
import { readDictByHant, type Definition } from './dict.ts'

/**
 * Reads a vocabulary or radical order CSV (e.g. lang/zh_CN/order/vocabulary.csv).
 * Returns an array of levels, each level being an array of slugs (traditional forms).
 */
export function readLessonOrder(input: string): string[][] {
  return parse(Deno.readTextFileSync(DATA_ROOT + input)) as string[][]
}

/**
 * Reads a character order file (e.g. lang/zh_CN/order/characters.txt).
 * Each line is one level; each character in the line is one slug.
 */
export function readCharacterOrder(targetLang: string): string[][] {
  const text = Deno.readTextFileSync(DATA_ROOT + `lang/${targetLang}/order/characters.txt`)
  return text.split('\n').map((row) => row.split(''))
}

/**
 * Returns the ordered set of definitions for items in the curriculum, deduplicated
 * and in curriculum order. Only includes items that are actually taught — not the
 * full dictionary.
 *
 * @param type - 'character' or 'vocabulary'
 * @param targetLang - target language code (default: 'zh_CN')
 */
export function readOrderedDefs(
  type: 'character' | 'vocabulary',
  targetLang = 'zh_CN',
): Definition[] {
  const [dictPath, slugsList] = type === 'character'
    ? ['lang/characters.tsv', readCharacterOrder(targetLang).flat()]
    : ['lang/vocabulary.tsv', readLessonOrder(`lang/${targetLang}/order/vocabulary.csv`).flat()]

  const byHant = readDictByHant(dictPath)
  const seen = new Set<string>()
  const result: Definition[] = []
  for (const slug of slugsList) {
    if (!slug || seen.has(slug)) continue
    seen.add(slug)
    const def = byHant[slug]
    if (def) result.push(def)
  }
  return result
}