[GitHub] carbondata pull request #2876: [CARBONDATA-3054] Fix Dictionary file cannot ...

[qiuchenjian-2]

GitHub user ajantha-bhat opened a pull request:

    https://github.com/apache/carbondata/pull/2876

    [CARBONDATA-3054] Fix Dictionary file cannot be read in S3a with CarbonDictionaryDecoder.doConsume() codeGen

    [CARBONDATA-3054] Fix Dictionary file cannot be read in S3a with CarbonDictionaryDecoder.doConsume() codeGen
   
    **problem:** In S3a environment, when queried the data which has dictionary files,
    Dictionary file cannot be read in S3a with CarbonDictionaryDecoder.doConsume() codeGen even though file is present.
   
    **cause:** CarbonDictionaryDecoder.doConsume() codeGen doesn't set hadoop conf in thread local variable, only doExecute() sets it.
   
    Hence, when getDictionaryWrapper() called from doConsume() codeGen,
   
    AbstractDictionaryCache.getDictionaryMetaCarbonFile() returns false for fileExists() operation.
   
    **solution:**
    In CarbonDictionaryDecoder.doConsume() codeGen, set hadoop conf in thread local variable
   
    Be sure to do all of the following checklist to help us incorporate
    your contribution quickly and easily:
   
     - [ ] Any interfaces changed? NA
     
     - [ ] Any backward compatibility impacted? NA
     
     - [ ] Document update required? NA
   
     - [ ] Testing done
    done      
     - [ ] For large changes, please consider breaking it into sub-tasks under an umbrella JIRA.  NA
   


You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/ajantha-bhat/carbondata master_new

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/carbondata/pull/2876.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #2876
   
----
commit 492823671d0aead8393a55f147d5e11c21e0bdfa
Author: ajantha-bhat <ajanthabhat@...>
Date:   2018-10-29T12:26:29Z

    [CARBONDATA-3054] Fix Dictionary file cannot be read in S3a with CarbonDictionaryDecoder.doConsume() codeGen

----


---

[qiuchenjian-2]

Github user ajantha-bhat commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    @ravipesala , @kunal642 : please review


---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    Build Success with Spark 2.1.0, Please check CI http://136.243.101.176:8080/job/ApacheCarbonPRBuilder2.1/1122/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    Build Failed with Spark 2.2.1, Please check CI http://95.216.28.178:8080/job/ApacheCarbonPRBuilder1/1334/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    Build Success with Spark 2.3.1, Please check CI http://136.243.101.176:8080/job/carbondataprbuilder2.3/9386/



---

[qiuchenjian-2]

Github user ajantha-bhat commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    retest this please


---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    Build Success with Spark 2.1.0, Please check CI http://136.243.101.176:8080/job/ApacheCarbonPRBuilder2.1/1126/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    Build Failed with Spark 2.2.1, Please check CI http://95.216.28.178:8080/job/ApacheCarbonPRBuilder1/1339/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    Build Success with Spark 2.3.1, Please check CI http://136.243.101.176:8080/job/carbondataprbuilder2.3/9390/



---

[qiuchenjian-2]

Github user ajantha-bhat commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    retest this please


---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    Build Success with Spark 2.1.0, Please check CI http://136.243.101.176:8080/job/ApacheCarbonPRBuilder2.1/1135/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    Build Failed with Spark 2.2.1, Please check CI http://95.216.28.178:8080/job/ApacheCarbonPRBuilder1/1347/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    Build Success with Spark 2.3.1, Please check CI http://136.243.101.176:8080/job/carbondataprbuilder2.3/9399/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    Build Success with Spark 2.1.0, Please check CI http://136.243.101.176:8080/job/ApacheCarbonPRBuilder2.1/1140/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    Build Failed  with Spark 2.3.1, Please check CI http://136.243.101.176:8080/job/carbondataprbuilder2.3/9405/



---

[qiuchenjian-2]

Github user CarbonDataQA commented on the issue:

    https://github.com/apache/carbondata/pull/2876
 
    Build Success with Spark 2.2.1, Please check CI http://95.216.28.178:8080/job/ApacheCarbonPRBuilder1/1352/



---

[qiuchenjian-2]

Github user manishgupta88 commented on a diff in the pull request:

    https://github.com/apache/carbondata/pull/2876#discussion_r229208346
 
    --- Diff: integration/spark2/src/main/scala/org/apache/spark/sql/CarbonDictionaryDecoder.scala ---
    @@ -137,14 +139,15 @@ case class CarbonDictionaryDecoder(
           val forwardDictionaryCache: Cache[DictionaryColumnUniqueIdentifier, Dictionary] =
             cacheProvider.createCache(CacheType.FORWARD_DICTIONARY)
           val dicts: Seq[ForwardDictionaryWrapper] = getDictionaryWrapper(tableNameToCarbonTableMapping,
    -        forwardDictionaryCache)
    +        forwardDictionaryCache, conf)
     
           val exprs = child.output.map { exp =>
             ExpressionCanonicalizer.execute(BindReferences.bindReference(exp, child.output))
           }
           ctx.currentVars = input
           val resultVars = exprs.zipWithIndex.map { case (expr, index) =>
             if (dicts(index) != null) {
    +          ThreadLocalSessionInfo.setConfigurationToCurrentThread(conf.value.value)
    --- End diff --
   
    Remove its usage from this place


---

[qiuchenjian-2]

Github user manishgupta88 commented on a diff in the pull request:

    https://github.com/apache/carbondata/pull/2876#discussion_r229208742
 
    --- Diff: integration/spark2/src/main/scala/org/apache/spark/sql/CarbonDictionaryDecoder.scala ---
    @@ -557,16 +561,20 @@ class CarbonDecoderRDD(
      * It is a wrapper around Dictionary, it is a work around to keep the dictionary serializable in
      * case of codegen
      * @param dictIdentifier Dictionary column unique identifier
    + * @param broadcastConf hadoop broadcast conf for serialization, that contains carbon conf.
      */
     class ForwardDictionaryWrapper(
    -    dictIdentifier: DictionaryColumnUniqueIdentifier) extends Serializable {
    +    dictIdentifier: DictionaryColumnUniqueIdentifier,
    +    broadcastConf: Broadcast[SerializableConfiguration]) extends Serializable {
     
       var dictionary: Dictionary = null
     
       var dictionaryLoader: DictionaryLoader = _
     
       def getDictionaryValueForKeyInBytes (surrogateKey: Int): Array[Byte] = {
         if (dictionary == null) {
    +      ThreadLocalSessionInfo.getOrCreateCarbonSessionInfo().getNonSerializableExtraInfo
    --- End diff --
   
    Move this in doConsume method inside `if (CarbonDictionaryDecoder.isRequiredToDecode(getDictionaryColumnIds)) {` as wrapper is not the correct place for setting ThreadLocalSessionInfo


---

[qiuchenjian-2]

Github user ajantha-bhat commented on a diff in the pull request:

    https://github.com/apache/carbondata/pull/2876#discussion_r229237604
 
    --- Diff: integration/spark2/src/main/scala/org/apache/spark/sql/CarbonDictionaryDecoder.scala ---
    @@ -137,14 +139,15 @@ case class CarbonDictionaryDecoder(
           val forwardDictionaryCache: Cache[DictionaryColumnUniqueIdentifier, Dictionary] =
             cacheProvider.createCache(CacheType.FORWARD_DICTIONARY)
           val dicts: Seq[ForwardDictionaryWrapper] = getDictionaryWrapper(tableNameToCarbonTableMapping,
    -        forwardDictionaryCache)
    +        forwardDictionaryCache, conf)
     
           val exprs = child.output.map { exp =>
             ExpressionCanonicalizer.execute(BindReferences.bindReference(exp, child.output))
           }
           ctx.currentVars = input
           val resultVars = exprs.zipWithIndex.map { case (expr, index) =>
             if (dicts(index) != null) {
    +          ThreadLocalSessionInfo.setConfigurationToCurrentThread(conf.value.value)
    --- End diff --
   
    done. Setted in codeGen now.


---

[qiuchenjian-2]

Github user ajantha-bhat commented on a diff in the pull request:

    https://github.com/apache/carbondata/pull/2876#discussion_r229237654
 
    --- Diff: integration/spark2/src/main/scala/org/apache/spark/sql/CarbonDictionaryDecoder.scala ---
    @@ -557,16 +561,20 @@ class CarbonDecoderRDD(
      * It is a wrapper around Dictionary, it is a work around to keep the dictionary serializable in
      * case of codegen
      * @param dictIdentifier Dictionary column unique identifier
    + * @param broadcastConf hadoop broadcast conf for serialization, that contains carbon conf.
      */
     class ForwardDictionaryWrapper(
    -    dictIdentifier: DictionaryColumnUniqueIdentifier) extends Serializable {
    +    dictIdentifier: DictionaryColumnUniqueIdentifier,
    +    broadcastConf: Broadcast[SerializableConfiguration]) extends Serializable {
     
       var dictionary: Dictionary = null
     
       var dictionaryLoader: DictionaryLoader = _
     
       def getDictionaryValueForKeyInBytes (surrogateKey: Int): Array[Byte] = {
         if (dictionary == null) {
    +      ThreadLocalSessionInfo.getOrCreateCarbonSessionInfo().getNonSerializableExtraInfo
    --- End diff --
   
    done. Setted in codeGen now


---