[
https://issues.apache.org/jira/browse/FLINK-4111?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15366057#comment-15366057
]
ASF GitHub Bot commented on FLINK-4111:
---------------------------------------
Github user rmetzger commented on the issue:
https://github.com/apache/flink/pull/2209
Are all the dependencies you removed really not needed or are they needed,
but present because they are pulled in transitively?
In general, it makes sense to define all needed dependencies, instead of
relying on transitive dependencies.
The problem with the suggested shading is that you are not relocating
calcite, which means that we have calcite in the org.apache.calcite namespace,
in our flink package.
This can potentially lead to version clashes if our users want to use a
different calcite version in their code (I have to admit that's unlikely.)
But for `com.fasterxml.jackson` this might be an issue, also for
`org.slf4j`.
I would exclude slf4j and jackson from the shading.
Do you know why the relocation of calcite doesn't work? Do you think there
is an issue with relocating class files produced by the scala compiler?
> Flink Table & SQL doesn't work in very simple example
> -----------------------------------------------------
>
> Key: FLINK-4111
> URL: https://issues.apache.org/jira/browse/FLINK-4111
> Project: Flink
> Issue Type: Bug
> Components: Table API & SQL
> Affects Versions: 1.1.0
> Reporter: Jark Wu
> Assignee: Timo Walther
> Fix For: 1.1.0
>
>
> I’m trying to use Flink Table 1.1-SNAPSHOT where I want to use Table API and
> SQL in my project. But when I run the very simple example WordCountTable, I
> encountered the following exception :
> {code}
> Exception in thread "main" java.lang.NoSuchMethodError:
> org.apache.calcite.rel.logical.LogicalAggregate.getGroupSets()Lorg/apache/flink/shaded/com/google/common/collect/ImmutableList;
> at
> org.apache.flink.api.table.plan.rules.dataSet.DataSetAggregateRule.matches(DataSetAggregateRule.scala:47)
> at
> org.apache.calcite.plan.volcano.VolcanoRuleCall.matchRecurse(VolcanoRuleCall.java:269)
> at
> org.apache.calcite.plan.volcano.VolcanoRuleCall.match(VolcanoRuleCall.java:253)
> at
> org.apache.calcite.plan.volcano.VolcanoPlanner.fireRules(VolcanoPlanner.java:1542)
> at
> org.apache.calcite.plan.volcano.VolcanoPlanner.registerImpl(VolcanoPlanner.java:1817)
> at
> org.apache.calcite.plan.volcano.VolcanoPlanner.register(VolcanoPlanner.java:1038)
> at
> org.apache.calcite.plan.volcano.VolcanoPlanner.ensureRegistered(VolcanoPlanner.java:1058)
> at
> org.apache.calcite.plan.volcano.VolcanoPlanner.changeTraits(VolcanoPlanner.java:723)
> at org.apache.calcite.tools.Programs$RuleSetProgram.run(Programs.java:331)
> at
> org.apache.flink.api.table.BatchTableEnvironment.translate(BatchTableEnvironment.scala:250)
> at
> org.apache.flink.api.scala.table.BatchTableEnvironment.toDataSet(BatchTableEnvironment.scala:139)
> at
> org.apache.flink.api.scala.table.TableConversions.toDataSet(TableConversions.scala:41)
> at com.alibaba.flink.examples.WordCountTable$.main(WordCountTable.scala:43)
> at com.alibaba.flink.examples.WordCountTable.main(WordCountTable.scala)
> {code}
> It seems that something wrong with our guava shade. Do you have any ideas?
> My pom file and WordCountTable.scala are
> [here|https://gist.github.com/wuchong/9c1c0df3cb7453502abc4605f5347289].
> And I found someone have the same problem on stack overflow
> [http://stackoverflow.com/questions/37835408/org-apache-flink-api-table-tableexception-alias-on-field-reference-expression-e#comment63160086_37838816]
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
