private def trimPhraseChunk()

in src/main/scala/com/twitter/penguin/korean/phrase_extractor/KoreanPhraseExtractor.scala [74:98]

• 23 lines of code
• 8 McCabe index (conditional complexity)

  private def trimPhraseChunk(phrases: KoreanPhraseChunk): KoreanPhraseChunk = {
    def trimNonNouns: Seq[KoreanPhrase] = {
      phrases
        .dropWhile(t => !PhraseHeadPoses.contains(t.pos))
        .reverse
        .dropWhile(t => !PhrasTailPoses.contains(t.pos))
        .reverse
    }

    def trimSpacesFromPhrase(phrases: Seq[KoreanPhrase]): Seq[KoreanPhrase] = {
      phrases.zipWithIndex.map {
        case (phrase, i) if phrases.length == 1 =>
          KoreanPhrase(phrase.tokens
            .dropWhile(_.pos == Space)
            .reverse.dropWhile(_.pos == Space).reverse, phrase.pos)
        case (phrase, i) if i == 0 =>
          KoreanPhrase(phrase.tokens.dropWhile(_.pos == Space), phrase.pos)
        case (phrase, i) if i == phrases.length - 1 =>
          KoreanPhrase(phrase.tokens.reverse.dropWhile(_.pos == Space).reverse, phrase.pos)
        case (phrase, i) => phrase
      }
    }

    trimSpacesFromPhrase(trimNonNouns)
  }