lostluck commented on code in PR #25808: URL: https://github.com/apache/beam/pull/25808#discussion_r1142401126
########## sdks/go/pkg/beam/transforms/periodic/periodic.go: ########## @@ -0,0 +1,211 @@ +// Licensed to the Apache Software Foundation (ASF) under one or more +// contributor license agreements. See the NOTICE file distributed with +// this work for additional information regarding copyright ownership. +// The ASF licenses this file to You under the Apache License, Version 2.0 +// (the "License"); you may not use this file except in compliance with +// the License. You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, software +// distributed under the License is distributed on an "AS IS" BASIS, +// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +// See the License for the specific language governing permissions and +// limitations under the License. + +// Package periodic contains transformations for generating periodic sequences. +package periodic + +import ( + "context" + "fmt" + "math" + "reflect" + "time" + + "github.com/apache/beam/sdks/v2/go/pkg/beam" + "github.com/apache/beam/sdks/v2/go/pkg/beam/core/graph/mtime" + "github.com/apache/beam/sdks/v2/go/pkg/beam/core/graph/window" + "github.com/apache/beam/sdks/v2/go/pkg/beam/core/sdf" + "github.com/apache/beam/sdks/v2/go/pkg/beam/io/rtrackers/offsetrange" + "github.com/apache/beam/sdks/v2/go/pkg/beam/register" +) + +func init() { + register.DoFn5x2[context.Context, *sdf.ManualWatermarkEstimator, *sdf.LockRTracker, SequenceDefinition, + func(beam.EventTime, int64), + sdf.ProcessContinuation, error](&sequenceGenDoFn{}) + register.Emitter2[beam.EventTime, int64]() + beam.RegisterType(reflect.TypeOf(SequenceDefinition{})) +} + +// SequenceDefinition holds the configuration for generating a sequence of +// timestamped elements at an interval. +type SequenceDefinition struct { + Interval time.Duration + Start time.Time + End time.Time +} + +type sequenceGenDoFn struct { + now func() time.Time +} + +func (fn *sequenceGenDoFn) Setup() { + if fn.now == nil { + fn.now = time.Now + } +} + +func (fn *sequenceGenDoFn) CreateInitialRestriction(sd SequenceDefinition) offsetrange.Restriction { + totalOutputs := math.Ceil(float64(sd.End.Sub(sd.Start) / sd.Interval)) + return offsetrange.Restriction{ + Start: int64(0), + End: int64(totalOutputs), + } +} + +func (fn *sequenceGenDoFn) CreateTracker(rest offsetrange.Restriction) *sdf.LockRTracker { + return sdf.NewLockRTracker(offsetrange.NewTracker(rest)) +} + +func (fn *sequenceGenDoFn) RestrictionSize(_ SequenceDefinition, rest offsetrange.Restriction) float64 { + return rest.Size() +} + +func (fn *sequenceGenDoFn) SplitRestriction(_ SequenceDefinition, rest offsetrange.Restriction) []offsetrange.Restriction { + return []offsetrange.Restriction{rest} +} + +// TruncateRestriction immediately truncates the entire restrication. +func (fn *sequenceGenDoFn) TruncateRestriction(_ *sdf.LockRTracker, _ SequenceDefinition) offsetrange.Restriction { + return offsetrange.Restriction{} +} + +func (fn *sequenceGenDoFn) CreateWatermarkEstimator() *sdf.ManualWatermarkEstimator { + return &sdf.ManualWatermarkEstimator{} Review Comment: While I agree in principle, if we were to take that route we should add a constructor function to the sdf package that does that instead `NewManualWatermarkEstimator()`, so it's clearer where to start, and avoid others needing to think about it as well. On the the other hand, the Unix Epoch was 50+ years ago, and users that would be trying to reprocess data dated older than that would know they're probably going to need some special handling around it, and as you've said, it can be worked around already. I'd say, we do nothing at this juncture. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
