org.bdgenomics.adam

projections

package projections

Visibility
  1. Public
  2. All

Type Members

  1. abstract class FieldEnumeration extends Enumeration

  2. trait FieldValue extends AnyRef

    The FieldValue trait (and its sister abstract class, FieldEnumeration) are meant to help clean up the use of Projection when dealing with field enumeration classes like ADAMRecordField.

    The FieldValue trait (and its sister abstract class, FieldEnumeration) are meant to help clean up the use of Projection when dealing with field enumeration classes like ADAMRecordField.

    Projection is a class for turning fields from an Enumeration into a projected Avro Schema object.

    In the old way of doing this, we used a "normal" Enumeration. Projection would receive a collection of Read.Value objects, and use their names (plus the Read.SCHEMA$ static field) to turn them into a Schema for ADAMRecords. This worked fine for Read, and not at all for generalizing to other Schemas over other field enumerations.

    In the new system, we embed the Avro Schema object as an argument *in* each enumeration Value. We do this in two steps: (1) instead of ADAMRecordField (e.g.) extending Enumeration, it extends FieldEnumeration and provides the appropriate (static) Schema object as an argument. (2) instead of using the (final, non-overrideable) Value method within Enumeration to provide each enum value, it calls FieldEnumeration.SchemaValue instead, which embeds the corresponding Schema in each value.

    Finally, Projection will extract the Schema value from the first FieldValue that is given to it and produce the corresponding (projected) Schema.

    This means, of course, that Projection can't handle empty field lists -- but that was always going to be an error-filled edge-case anyway (why would you want to project to zero fields?)

Value Members

  1. object AlignmentRecordField extends FieldEnumeration

    This enumeration exist in order to reduce typo errors in the code.

    This enumeration exist in order to reduce typo errors in the code. It needs to be kept in sync with any changes to Read.

    This enumeration is necessary because Parquet needs the field string names for predicates and projections.

  2. object DatabaseVariantAnnotationField extends FieldEnumeration

  3. object FeatureField extends FieldEnumeration

    This enumeration exist in order to reduce typo errors in the code.

    This enumeration exist in order to reduce typo errors in the code. It needs to be kept in sync with any changes to Feature.

    This enumeration is necessary because Parquet needs the field string names for predicates and projections.

  4. object Filter

  5. object GenotypeField extends FieldEnumeration

  6. object NucleotideContigFragmentField extends FieldEnumeration

    This enumeration exist in order to reduce typo errors in the code.

    This enumeration exist in order to reduce typo errors in the code. It needs to be kept in sync with any changes to Read.

    This enumeration is necessary because Parquet needs the field string names for predicates and projections.

  7. object Projection

    Avro utility object to create a projection of a Schema

  8. object VariantCallingAnnotationsField extends FieldEnumeration

  9. object VariantField extends FieldEnumeration

Ungrouped