Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add IngestionId & EventTimestamp to FeatureRowBatch to calculate lag metric correctly #874

Merged
merged 3 commits into from
Jul 13, 2020
Merged

Add IngestionId & EventTimestamp to FeatureRowBatch to calculate lag metric correctly #874

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
395204b
ingestionId & eventTimestamp in FeatureRowBatch
pyalex Jul 9, 2020
050e3b5
refactor idx operations using schema
pyalex Jul 13, 2020
71a23e9
dummy
Jul 13, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
90 changes: 70 additions & 20 deletions ...igquery/src/main/java/feast/storage/connectors/bigquery/compression/FeatureRowsBatch.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,8 @@
import static feast.proto.types.ValueProto.Value.ValCase.*;
import static feast.storage.connectors.bigquery.common.TypeUtil.*;

import com.google.common.collect.ImmutableList;
import com.google.protobuf.Timestamp;
import feast.proto.types.FeatureRowProto;
import feast.proto.types.FieldProto;
import feast.proto.types.ValueProto;
Expand All @@ -40,6 +42,8 @@
* <p>getFeatureRows provides reverse transformation
*/
public class FeatureRowsBatch implements Serializable {
public static final ImmutableList<String> SERVICE_FIELDS =
ImmutableList.of("eventTimestamp", "ingestionId");
private final Schema schema;
private String featureSetReference;
private List<Object> values = new ArrayList<>();
Expand Down Expand Up @@ -118,6 +122,12 @@ private Schema inferCommonSchema(Iterable<FeatureRowProto.FeatureRow> featureRow
featureSetReference = row.getFeatureSet();
}
}));

fieldsInOrder.add(
Schema.Field.of("eventTimestamp", Schema.FieldType.array(Schema.FieldType.INT64)));
fieldsInOrder.add(
Schema.Field.of("ingestionId", Schema.FieldType.array(Schema.FieldType.STRING)));

Schema schema = Schema.builder().addFields(fieldsInOrder).build();
schema.setUUID(UUID.randomUUID());
return schema;
Expand All @@ -132,16 +142,34 @@ private void initValues() {
}

private void toColumnar(Iterable<FeatureRowProto.FeatureRow> featureRows) {
int timestampColumnIdx = schema.indexOf("eventTimestamp");
int ingestionIdColumnIdx = schema.indexOf("ingestionId");

featureRows.forEach(
row -> {
Map<String, ValueProto.Value> rowValues =
row.getFieldsList().stream()
.collect(Collectors.toMap(FieldProto.Field::getName, FieldProto.Field::getValue));
Map<String, ValueProto.Value> rowValues;
try {
rowValues =
row.getFieldsList().stream()
.collect(
Collectors.toMap(FieldProto.Field::getName, FieldProto.Field::getValue));
} catch (IllegalStateException e) {
// row contains feature duplicates
// omitting for now
return;
}

IntStream.range(0, schema.getFieldCount())
schema
.getFieldNames()
.forEach(
idx -> {
Schema.Field field = schema.getField(idx);
fieldName -> {
if (SERVICE_FIELDS.contains(fieldName)) {
return;
}

Schema.Field field = schema.getField(fieldName);
int idx = schema.indexOf(fieldName);

if (rowValues.containsKey(field.getName())) {
Object o = protoValueToObject(rowValues.get(field.getName()));
if (o != null) {
Expand All @@ -152,6 +180,10 @@ private void toColumnar(Iterable<FeatureRowProto.FeatureRow> featureRows) {

((List<Object>) values.get(idx)).add(defaultValues.get(field.getName()));
});

// adding service fields
((List<Object>) values.get(timestampColumnIdx)).add(row.getEventTimestamp().getSeconds());
((List<Object>) values.get(ingestionIdColumnIdx)).add(row.getIngestionId());
});
}

Expand All @@ -177,27 +209,45 @@ public static FeatureRowsBatch fromRow(Row row) {
}

public Iterator<FeatureRowProto.FeatureRow> getFeatureRows() {
int timestampColumnIdx = schema.indexOf("eventTimestamp");
int ingestionIdColumnIdx = schema.indexOf("ingestionId");

return IntStream.range(0, ((List<Object>) values.get(0)).size())
.parallel()
.mapToObj(
rowIdx ->
FeatureRowProto.FeatureRow.newBuilder()
.setFeatureSet(getFeatureSetReference())
.setEventTimestamp(
Timestamp.newBuilder()
.setSeconds(
(long)
(((List<Object>) values.get(timestampColumnIdx)).get(rowIdx)))
.build())
.setIngestionId(
(String) (((List<Object>) values.get(ingestionIdColumnIdx)).get(rowIdx)))
.addAllFields(
IntStream.range(0, schema.getFieldCount())
.mapToObj(
fieldIdx ->
FieldProto.Field.newBuilder()
.setName(schema.getField(fieldIdx).getName())
.setValue(
objectToProtoValue(
((List<Object>) values.get(fieldIdx)).get(rowIdx),
schemaToProtoTypes.get(
schema
.getField(fieldIdx)
.getType()
.getCollectionElementType())))
.build())
schema.getFieldNames().stream()
.map(
fieldName -> {
if (SERVICE_FIELDS.contains(fieldName)) {
return null;
}
int fieldIdx = schema.indexOf(fieldName);

return FieldProto.Field.newBuilder()
.setName(schema.getField(fieldIdx).getName())
.setValue(
objectToProtoValue(
((List<Object>) values.get(fieldIdx)).get(rowIdx),
schemaToProtoTypes.get(
schema
.getField(fieldIdx)
.getType()
.getCollectionElementType())))
.build();
})
.filter(Objects::nonNull)
.collect(Collectors.toList()))
.build())
.iterator();
Expand Down
9 changes: 9 additions & 0 deletions ...ors/bigquery/src/test/java/feast/storage/connectors/bigquery/writer/BigQuerySinkTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -124,6 +124,11 @@ private FeatureRow generateRow(String featureSet) {
FeatureRow.Builder row =
FeatureRow.newBuilder()
.setFeatureSet(featureSet)
.setEventTimestamp(
com.google.protobuf.Timestamp.newBuilder()
.setSeconds(System.currentTimeMillis() / 1000)
.build())
.setIngestionId("ingestion-id")
.addFields(field("entity", rd.nextInt(), ValueProto.ValueType.Enum.INT64))
.addFields(FieldProto.Field.newBuilder().setName("null_value").build());

Expand Down Expand Up @@ -499,6 +504,8 @@ private List<FeatureRow> dropNullFeature(List<FeatureRow> input) {
r ->
FeatureRow.newBuilder()
.setFeatureSet(r.getFeatureSet())
.setIngestionId(r.getIngestionId())
.setEventTimestamp(r.getEventTimestamp())
.addAllFields(copyFieldsWithout(r, "null_value"))
.build())
.collect(Collectors.toList());
Expand All @@ -520,6 +527,8 @@ public static List<FeatureRow> sortFeaturesByName(List<FeatureRow> rows) {

return FeatureRow.newBuilder()
.setFeatureSet(row.getFeatureSet())
.setEventTimestamp(row.getEventTimestamp())
.setIngestionId(row.getIngestionId())
.addAllFields(fieldsList)
.build();
})
Expand Down