org.apache.beam.sdk.schemas.utils

Class SelectHelpers

java.lang.Object

org.apache.beam.sdk.schemas.utils.SelectHelpers

@Experimental(value=SCHEMAS)
public class SelectHelpers
extends java.lang.Object

Helper methods to select subrows out of rows.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	SelectHelpers.RowSelectorContainer

Field Summary

Fields

Modifier and Type	Field and Description
static SerializableFunction<java.util.List<java.lang.String>,java.lang.String>	CONCAT_FIELD_NAMES This policy keeps all levels of a name.
static SerializableFunction<java.util.List<java.lang.String>,java.lang.String>	KEEP_NESTED_NAME This policy keeps the raw nested field name.

Constructor Summary

Constructors

Constructor and Description
SelectHelpers()

Method Summary

Modifier and Type	Method and Description
static FieldAccessDescriptor	allLeavesDescriptor(Schema schema, SerializableFunction<java.util.List<java.lang.String>,java.lang.String> nameFn)
static Schema	getOutputSchema(Schema inputSchema, FieldAccessDescriptor fieldAccessDescriptor) Get the output schema resulting from selecting the given FieldAccessDescriptor from the given schema.
static RowSelector	getRowSelector(Schema inputSchema, FieldAccessDescriptor fieldAccessDescriptor)
static RowSelector	getRowSelectorOptimized(Schema inputSchema, FieldAccessDescriptor fieldAccessDescriptor)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

CONCAT_FIELD_NAMES

public static final SerializableFunction<java.util.List<java.lang.String>,java.lang.String> CONCAT_FIELD_NAMES

This policy keeps all levels of a name. Every field name in the path to a given field is concated with _ characters.

KEEP_NESTED_NAME

public static final SerializableFunction<java.util.List<java.lang.String>,java.lang.String> KEEP_NESTED_NAME

This policy keeps the raw nested field name. If two differently-nested fields have the same name, flattening will fail with this policy.

Constructor Detail

SelectHelpers

public SelectHelpers()

Method Detail

getOutputSchema

public static Schema getOutputSchema(Schema inputSchema,
                                     FieldAccessDescriptor fieldAccessDescriptor)

Get the output schema resulting from selecting the given FieldAccessDescriptor from the given schema.

Fields are always extracted and then stored in a new Row. For example, consider the following Java POJOs:

  class UserEvent {
    String userId;
    String eventId;
    int eventType;
    Location location;
 }
 

 class Location {
   double latitude;
   double longitude;
 }
 

If selecting just the location field, then the returned schema will wrap that of the singular field being selected; in this case the returned schema will be a Row containing a single Location field. If location.latitude is selected, then the returned Schema will be a Row containing a double latitude field.

The same holds true when selecting from lists or maps. For example:

 class EventList {
   List<UserEvent> events;
 }
 

If selecting events.location.latitude, the returned schema will contain a single array of Row, where that Row contains a single double latitude field; it will not contain an array of double.

getRowSelectorOptimized

public static RowSelector getRowSelectorOptimized(Schema inputSchema,
                                                  FieldAccessDescriptor fieldAccessDescriptor)

getRowSelector

public static RowSelector getRowSelector(Schema inputSchema,
                                         FieldAccessDescriptor fieldAccessDescriptor)

allLeavesDescriptor

public static FieldAccessDescriptor allLeavesDescriptor(Schema schema,
                                                        SerializableFunction<java.util.List<java.lang.String>,java.lang.String> nameFn)