Method parse

google-drive-embeddings/src/parser.js:89–167

* Main parsing routine. Splits and validates a filename. * Supports strict delimiters (spaces around the dash) or loose delimiters.

(filename, options = {})

   * Supports strict delimiters (spaces around the dash) or loose delimiters.
   */
  static parse(filename, options = {}) {
    const { strictDelimiter = true } = options;
    const cleaned = this.cleanString(filename);
    const result = {
      original: filename,
      cleaned: cleaned,
      approved: false,
      artist: null,
      title: null,
      reasons: [],
      checklist: {
        separatorFound: false,
        artistValid: false,
        titleValid: false
      }
    };

    if (!cleaned) {
      result.reasons.push('Empty filename after cleaning');
      return result;
    }

    // Standard separators: space-dash-space, en-dash, em-dash, colon, tilde
    // e.g., "Artist - Title", "Artist: Title", "Artist ~ Title"
    let separatorRegex = /\s+[-–—~:]\s+/;
    if (!strictDelimiter) {
      // In loose mode, support any dash bordered by word characters, or just colon/tilde
      separatorRegex = /\s+[-–—~:]\s+|(?<=\w)-(?=\w)|[-–—~:]/;
    }

    let parts = cleaned.split(separatorRegex).map(p => p.trim()).filter(Boolean);

    // If no standard separator is found but we split on raw dash/colon
    if (parts.length === 1 && strictDelimiter) {
      result.reasons.push('No standard space-separated delimiter (e.g. " - ") found');
      return result;
    }

    result.checklist.separatorFound = true;

    // Check if we have standard "01 - Artist - Title" (3 parts) where first is index
    if (parts.length >= 3) {
      const firstPartCheck = this.validatePart(parts[0]);
      if (!firstPartCheck.valid && (this.NUMERIC_ONLY_REGEX.test(parts[0]) || this.TRACK_INDEX_ONLY_REGEX.test(parts[0]))) {
        // Discard the leading track number part and take the second and third parts!
        parts = [parts[1], parts[2]];
      }
    }

    if (parts.length < 2) {
      result.reasons.push('Could not split filename into both artist and title candidates');
      return result;
    }

    let artistCandidate = parts[0];
    let titleCandidate = parts[1];

    // If the first part is purely numbers/indexes, it's just index and title (no artist!)