Method splitGt

src/main/java/com/rtg/vcf/VcfUtils.java:221–257

Utility method for splitting a VCF genotype subfield into an array of numeric allele identifiers. @param gt a VCF genotype subfield (the GT value). @return a new int[] containing one int per allele. Any missing values ('.') will be assigned index -1. @throws NumberFormatEx

(String gt)

   * @throws NumberFormatException if the subfield is malformed.
   */
  public static int[] splitGt(String gt) {
    try {
      final int gtlen = gt.length();
      if (gtlen == 1) { // Typical haploid call
        return new int[]{alleleId(gt.charAt(0))};
      } else {
        int[] result = new int[2]; // Initialize assuming the most common case, diploid, and resize if needed
        int ploid = 0;
        int allelestart = 0;
        for (int i = 0; i < gtlen; ++i) {
          final char c = gt.charAt(i);
          if (c == PHASED_SEPARATOR || c == UNPHASED_SEPARATOR) {
            if (ploid == result.length) { // More than diploid call!
              result = Arrays.copyOf(result, result.length + 1);
            }
            result[ploid++] = alleleId(gt, allelestart, i - allelestart);
            allelestart = i + 1;
          }
        }
        if (allelestart < gtlen) {
          if (ploid == result.length) { // More than diploid call!
            result = Arrays.copyOf(result, result.length + 1);
          }
          result[ploid++] = alleleId(gt, allelestart, gtlen - allelestart);
        }
        if (ploid < result.length) { // Can only happen if a haploid genotype with allele id > 9
          result = Arrays.copyOf(result, ploid);
        }
        if (ploid == 0) {
          throw new NumberFormatException();
        }
        return result;
      }
    } catch (NumberFormatException e) {
      throw new VcfFormatException("Malformed VCF GT value \"" + gt + "\"");
    }
  }

  /**
   * Utility method for creating a VCF genotype subfield from an array of