[
https://issues.apache.org/jira/browse/BEAM-5441?focusedWorklogId=146414&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-146414
]
ASF GitHub Bot logged work on BEAM-5441:
----------------------------------------
Author: ASF GitHub Bot
Created on: 21/Sep/18 15:28
Start Date: 21/Sep/18 15:28
Worklog Time Spent: 10m
Work Description: ryan-williams commented on a change in pull request
#6461: [BEAM-5441] extend primitive transform check to find subtransform leaves
URL: https://github.com/apache/beam/pull/6461#discussion_r219537254
##########
File path:
runners/core-construction-java/src/main/java/org/apache/beam/runners/core/construction/graph/QueryablePipeline.java
##########
@@ -123,15 +125,24 @@ private QueryablePipeline(Collection<String>
transformIds, Components components
@VisibleForTesting
static Collection<String> getPrimitiveTransformIds(RunnerApi.Components
components) {
Collection<String> ids = new LinkedHashSet<>();
+
for (Map.Entry<String, PTransform> transformEntry :
components.getTransformsMap().entrySet()) {
PTransform transform = transformEntry.getValue();
boolean isPrimitive = isPrimitiveTransform(transform);
if (isPrimitive) {
- List<String> subtransforms = transform.getSubtransformsList();
- if (subtransforms.isEmpty()) {
- ids.add(transformEntry.getKey());
- } else {
- ids.addAll(subtransforms);
+ // for a "primitive" transform, we want to keep only "leaf"
sub-transform descendents, if any (in the common
Review comment:
[This write-up on
BEAM-5441](https://issues.apache.org/jira/browse/BEAM-5441?focusedCommentId=16622983&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16622983)
gives the best context:
> DirectRunner [replaces a
GBK](https://github.com/apache/beam/blob/master/runners/direct-java/src/main/java/org/apache/beam/runners/direct/DirectRunner.java#L264-L266)
(normally a primitive transform) with [a wrapper for two runner-specific
transforms ("group by key only" and "group also by
window")](https://github.com/apache/beam/blob/master/runners/direct-java/src/main/java/org/apache/beam/runners/direct/portable/DirectGroupByKey.java#L63-L70),
and the expectation is that the wrapped transforms are considered
"primitives", but the wrapper is not
>
> … in the wordcount case, the `Reshuffle` gets added as a primitive
transform (being a child of the otherwise-primitive `Read`) and, separately,
its child `RemoveRandomKeys` (URN `urn:beam:transform:pardo:v1`) gets marked
primitive, so there are two "primitive" transforms that have the same output
PCollection, failing [the
assertion](https://github.com/apache/beam/blob/9d0b03ad736797c33fbd6a4eb13155fa91367779/runners/core-construction-java/src/main/java/org/apache/beam/runners/core/construction/graph/QueryablePipeline.java#L176-L182).
Basically I've been collecting cases where "primitive" transforms end up
with descendents, and trying to do the right thing here as each comes to light!
Definitely interested in guidance from folks like yourself that have a more
complete "big picture" understanding of whether "primitives" should never have
children (let alone deeper descendents; this was the original definition of
"primitive" pre-#6328, but seemed to have false-positives and -negatives), or
how to handle the cases where they do (if not how I've prototyped here).
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
-------------------
Worklog Id: (was: 146414)
Time Spent: 1h 10m (was: 1h)
> Portable Wordcount fails in GreedyPipelineFuser
> -----------------------------------------------
>
> Key: BEAM-5441
> URL: https://issues.apache.org/jira/browse/BEAM-5441
> Project: Beam
> Issue Type: New Feature
> Components: examples-python, sdk-java-core
> Affects Versions: 2.8.0
> Reporter: Maximilian Michels
> Assignee: Ryan Williams
> Priority: Major
> Labels: portability, portability-flink
> Fix For: 2.8.0
>
> Time Spent: 1h 10m
> Remaining Estimate: 0h
>
> The Python SDK wordcount with the PortableRunner throws the following
> exception:
> {noformat}
> java.lang.IllegalArgumentException: A PCollectionNode should have exactly one
> producing PTransformNode, PCollectionNode{id=ref_PCollection_PCollection_26,
> PCollection=unique_name:
> "60write/Write/WriteImpl/DoOnce/Read/Reshuffle/RemoveRandomKeys.None"
> coder_id: "ref_Coder_FastPrimitivesCoder_2"
> is_bounded: BOUNDED
> windowing_strategy_id: "ref_Windowing_Windowing_1"
> } has
> [PTransformNode{id=ref_AppliedPTransform_write/Write/WriteImpl/DoOnce/Read/Reshuffle_32,
> transform=spec {
> urn: "beam:transform:generic_composite:v1"
> payload: "<Reshuffle(PTransform) label=[Reshuffle]>"
> }
> subtransforms:
> "ref_AppliedPTransform_write/Write/WriteImpl/DoOnce/Read/Reshuffle/AddRandomKeys_33"
> subtransforms:
> "ref_AppliedPTransform_write/Write/WriteImpl/DoOnce/Read/Reshuffle/ReshufflePerKey_34"
> subtransforms:
> "ref_AppliedPTransform_write/Write/WriteImpl/DoOnce/Read/Reshuffle/RemoveRandomKeys_41"
> inputs {
> key: "0"
> value: "ref_PCollection_PCollection_19"
> }
> outputs {
> key: "None"
> value: "ref_PCollection_PCollection_26"
> }
> unique_name: "write/Write/WriteImpl/DoOnce/Read/Reshuffle"
> },
> PTransformNode{id=ref_AppliedPTransform_write/Write/WriteImpl/DoOnce/Read/Reshuffle/RemoveRandomKeys_41,
> transform=spec {
> urn: "urn:beam:transform:pardo:v1"
> payload: "\n\317\006\n\255\006\n
> beam:dofn:pickled_python_info:v1\032\210\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\022\035ref_Environment_Environment_1"
> }
> inputs {
> key: "0"
> value: "ref_PCollection_PCollection_25"
> }
> outputs {
> key: "None"
> value: "ref_PCollection_PCollection_26"
> }
> unique_name: "write/Write/WriteImpl/DoOnce/Read/Reshuffle/RemoveRandomKeys"
> }]
> at
> org.apache.beam.repackaged.beam_runners_core_construction_java.com.google.common.base.Preconditions.checkArgument(Preconditions.java:416)
> at
> org.apache.beam.runners.core.construction.graph.QueryablePipeline.buildNetwork(QueryablePipeline.java:176)
> at
> org.apache.beam.runners.core.construction.graph.QueryablePipeline.<init>(QueryablePipeline.java:119)
> at
> org.apache.beam.runners.core.construction.graph.QueryablePipeline.forPrimitivesIn(QueryablePipeline.java:82)
> at
> org.apache.beam.runners.core.construction.graph.GreedyPipelineFuser.<init>(GreedyPipelineFuser.java:67)
> at
> org.apache.beam.runners.core.construction.graph.GreedyPipelineFuser.fuse(GreedyPipelineFuser.java:89)
> at
> org.apache.beam.runners.flink.FlinkJobInvocation.runPipeline(FlinkJobInvocation.java:96)
> at
> org.apache.beam.repackaged.beam_runners_flink_2.11.com.google.common.util.concurrent.TrustedListenableFutureTask$TrustedFutureInterruptibleTask.runInterruptibly(TrustedListenableFutureTask.java:111)
> at
> org.apache.beam.repackaged.beam_runners_flink_2.11.com.google.common.util.concurrent.InterruptibleTask.run(InterruptibleTask.java:58)
> at
> org.apache.beam.repackaged.beam_runners_flink_2.11.com.google.common.util.concurrent.TrustedListenableFutureTask.run(TrustedListenableFutureTask.java:75)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
> at java.lang.Thread.run(Thread.java:745)
> {noformat}
> Looks like it was caused by https://github.com/apache/beam/pull/6328
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
