[ 
https://issues.apache.org/jira/browse/BEAM-5675?focusedWorklogId=156906&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-156906
 ]

ASF GitHub Bot logged work on BEAM-5675:
----------------------------------------

                Author: ASF GitHub Bot
            Created on: 22/Oct/18 18:40
            Start Date: 22/Oct/18 18:40
    Worklog Time Spent: 10m 
      Work Description: kanterov closed pull request #6773: [BEAM-5675] Fix 
RowCoder#verifyDeterministic
URL: https://github.com/apache/beam/pull/6773
 
 
   

This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:

As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):

diff --git 
a/sdks/java/core/src/main/java/org/apache/beam/sdk/coders/RowCoder.java 
b/sdks/java/core/src/main/java/org/apache/beam/sdk/coders/RowCoder.java
index bcb87adfcee..017e14b6ae3 100644
--- a/sdks/java/core/src/main/java/org/apache/beam/sdk/coders/RowCoder.java
+++ b/sdks/java/core/src/main/java/org/apache/beam/sdk/coders/RowCoder.java
@@ -23,6 +23,7 @@
 import java.io.IOException;
 import java.io.InputStream;
 import java.io.OutputStream;
+import java.util.ArrayList;
 import java.util.List;
 import java.util.Map;
 import java.util.UUID;
@@ -123,28 +124,31 @@ public void verifyDeterministic()
 
   private void verifyDeterministic(Schema schema)
       throws org.apache.beam.sdk.coders.Coder.NonDeterministicException {
+    List<String> reasons = new ArrayList<>();
+    NonDeterministicException firstCause = null;
+
     for (Field field : schema.getFields()) {
-      verifyDeterministic(field.getType());
+      try {
+        verifyDeterministic(field.getType());
+      } catch (NonDeterministicException e) {
+        if (firstCause == null) {
+          firstCause = e;
+        }
+
+        for (String reason : e.getReasons()) {
+          reasons.add(field.getName() + " -> " + reason);
+        }
+      }
+    }
+
+    if (firstCause != null) {
+      throw new NonDeterministicException(this, reasons, firstCause);
     }
   }
 
   private void verifyDeterministic(FieldType fieldType)
       throws org.apache.beam.sdk.coders.Coder.NonDeterministicException {
-    switch (fieldType.getTypeName()) {
-      case MAP:
-        throw new NonDeterministicException(
-            this,
-            "Map-valued fields cannot be used in keys as Beam requires 
deterministic encoding for"
-                + " keys.");
-      case ROW:
-        verifyDeterministic(fieldType.getRowSchema());
-        break;
-      case ARRAY:
-        verifyDeterministic(fieldType.getCollectionElementType());
-        break;
-      default:
-        break;
-    }
+    coderForFieldType(fieldType).verifyDeterministic();
   }
 
   @Override
diff --git 
a/sdks/java/core/src/test/java/org/apache/beam/sdk/coders/org/apache/beam/sdk/coders/RowCoderTest.java
 
b/sdks/java/core/src/test/java/org/apache/beam/sdk/coders/org/apache/beam/sdk/coders/RowCoderTest.java
index a550422f3a3..5bf60f58fb4 100644
--- 
a/sdks/java/core/src/test/java/org/apache/beam/sdk/coders/org/apache/beam/sdk/coders/RowCoderTest.java
+++ 
b/sdks/java/core/src/test/java/org/apache/beam/sdk/coders/org/apache/beam/sdk/coders/RowCoderTest.java
@@ -18,13 +18,19 @@
 
 package org.apache.beam.sdk.coders.org.apache.beam.sdk.coders;
 
+import static org.hamcrest.Matchers.containsInAnyOrder;
 import static org.junit.Assert.assertEquals;
+import static org.junit.Assert.assertThat;
+import static org.junit.Assert.fail;
 
 import com.google.common.collect.Lists;
 import java.io.ByteArrayInputStream;
 import java.io.ByteArrayOutputStream;
 import java.io.IOException;
 import java.math.BigDecimal;
+import java.util.Arrays;
+import java.util.List;
+import org.apache.beam.sdk.coders.Coder.NonDeterministicException;
 import org.apache.beam.sdk.coders.RowCoder;
 import org.apache.beam.sdk.schemas.Schema;
 import org.apache.beam.sdk.schemas.Schema.FieldType;
@@ -115,4 +121,50 @@ public void testArrayOfArray() throws Exception {
             .build();
     checkEncodeDecode(row);
   }
+
+  @Test
+  public void testVerifyDeterministic() {
+    Schema schema =
+        Schema.builder()
+            .addField("f1", FieldType.DOUBLE)
+            .addField("f2", FieldType.FLOAT)
+            .addField("f3", FieldType.INT32)
+            .build();
+    RowCoder coder = RowCoder.of(schema);
+
+    String reason1 = "f1 -> Floating point encodings are not guaranteed to be 
deterministic.";
+    String reason2 = "f2 -> Floating point encodings are not guaranteed to be 
deterministic.";
+
+    assertNonDeterministic(coder, Arrays.asList(reason1, reason2));
+  }
+
+  @Test
+  public void testVerifyDeterministicNestedRow() {
+    Schema schema =
+        Schema.builder()
+            .addField("f1", FieldType.DOUBLE)
+            .addField(
+                "f2",
+                FieldType.row(
+                    Schema.builder()
+                        .addField("a1", FieldType.DOUBLE)
+                        .addField("a2", FieldType.INT64)
+                        .build()))
+            .build();
+    RowCoder coder = RowCoder.of(schema);
+
+    String reason1 = "f1 -> Floating point encodings are not guaranteed to be 
deterministic.";
+    String reason2 = "f2 -> a1 -> Floating point encodings are not guaranteed 
to be deterministic.";
+
+    assertNonDeterministic(coder, Arrays.asList(reason1, reason2));
+  }
+
+  private void assertNonDeterministic(RowCoder coder, List<String> reasons) {
+    try {
+      coder.verifyDeterministic();
+      fail("Expected " + coder + " to be non-deterministic");
+    } catch (NonDeterministicException e) {
+      assertThat(e.getReasons(), containsInAnyOrder(reasons.toArray()));
+    }
+  }
 }


 

----------------------------------------------------------------
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:
[email protected]


Issue Time Tracking
-------------------

    Worklog Id:     (was: 156906)
    Time Spent: 40m  (was: 0.5h)

> RowCoder#verifyDeterministic isn't consistent with Beam coders
> --------------------------------------------------------------
>
>                 Key: BEAM-5675
>                 URL: https://issues.apache.org/jira/browse/BEAM-5675
>             Project: Beam
>          Issue Type: Bug
>          Components: dsl-sql
>    Affects Versions: 2.7.0
>            Reporter: Gleb Kanterov
>            Assignee: Gleb Kanterov
>            Priority: Major
>          Time Spent: 40m
>  Remaining Estimate: 0h
>
> For instance, `DoubleCoder#verifyDeterministic` throws an exception, but 
> `RowCoder#verifyDeterministic` doesn't.
> Expected to throw NonDeterministicException
> {code:java}
> RowCoder.of(
>     Schema
>         .builder()
>         .addField("foo", Schema.FieldType.DOUBLE)
>         .build()
> ).verifyDeterministic();
> {code}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to