RowChangeDataProducer (OpenRefine 4.0-SNAPSHOT API)

Type Parameters:

T -

All Superinterfaces:

Serializable

All Known Implementing Classes:

ColumnAdditionByFetchingURLsOperation.URLFetchingChangeProducer, PerformWikibaseEditsOperation.RowEditingResultsProducer, ReconOperation.ReconChangeDataProducer, RowInRecordChangeDataProducer
```
public interface RowChangeDataProducer<T>
extends Serializable
```
A function which computes change data to be persisted to disk, to be later joined back to the project to produce the new grid. This data might be serialized because it is volatile or expensive to compute.
The calls to the external resource can be batched by overriding getBatchSize() to specify the size of batches and callRowBatch(List) for the batch processing itself. In that case, the call(long, Row) method's implementation can be omitted (by throwing NotImplementedException for instance).
It is also possible to limit the number of concurrent calls to the producer (for instance for rate-limited resources) by overriding getMaxConcurrency().

Method Summary

All Methods Instance Methods Abstract Methods Default Methods
Modifier and Type	Method	Description
`T`	`call(long rowId, Row row)`	Compute the change data on a given row.
`default List<T>`	`callRowBatch(List<IndexedRow> rows)`	Compute the change data on a batch of consecutive rows.
`default int`	`getBatchSize()`	The size of batches this producer would like to be called on.
`default int`	`getMaxConcurrency()`	The maximum number of concurrent calls to this change data producer.

- Method Detail
  - call
```
T call(long rowId,
       Row row)
```
    Compute the change data on a given row.
  - callRowBatch
```
default List<T> callRowBatch(List<IndexedRow> rows)
```
    Compute the change data on a batch of consecutive rows. This defaults to individual calls if the method is not overridden.
    
    Parameters:
    
    rows - the list of rows to fetch change data on
    
    Returns:
    
    a list of the same size
  - getBatchSize
```
default int getBatchSize()
```
    The size of batches this producer would like to be called on. Smaller batches can be submitted (for instance at the end of a partition). Defaults to 1.
  - getMaxConcurrency
```
default int getMaxConcurrency()
```
    The maximum number of concurrent calls to this change data producer. If 0, there is no limit to the concurrency.