[GitHub] [carbondata] Karan980 opened a new pull request #3979: [WIP] Fix insertion from ORC table into carbon table when sort scope is global sort

[GitBox]

Karan980 opened a new pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979


   ### Why is this PR needed?
   Loading records from ORC table having null values into carbon table having sort scope as global sort gives NPE.
   
    ### What changes were proposed in this PR?
   Added null check for arrayType and mapType before writing the data into byteArrays
   
    ### Does this PR introduce any user interface change?
    - No
   
    ### Is any new testcase added?
    - Yes
   
       
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

ajantha-bhat commented on a change in pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#discussion_r503096192



##########
File path: integration/spark/src/test/scala/org/apache/carbondata/spark/testsuite/allqueries/InsertIntoCarbonTableTestCase.scala
##########
@@ -67,6 +67,21 @@ class InsertIntoCarbonTableTestCase extends QueryTest with BeforeAndAfterAll {
 
   }
 
+  test("insert from orc-select columns with columns having null values and sort scope as global sort") {
+    sql("drop table if exists TORCSource")
+    sql("drop table if exists TCarbon")
+    sql("create table TORCSource(name string,col array<String>,fee int) STORED AS orc")
+    sql("insert into TORCSource values('karan',null,2)")
+    sql("create table TCarbon(name string, col array<String>,fee int) STORED AS carbondata TBLPROPERTIES ('SORT_COLUMNS'='name','TABLE_BLOCKSIZE'='128','TABLE_BLOCKLET_SIZE'='128','SORT_SCOPE'='global_SORT')")
+    sql("insert overwrite table TCarbon select name,col,fee from TORCSource")
+    val result = sql("show segments for table TCarbon").collect()(0).get(1).toString()
+    if(!"Success".equalsIgnoreCase(result)) {

Review comment:
       please do a select query and compare orc and carbon table results




----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

ajantha-bhat commented on a change in pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#discussion_r503097655



##########
File path: integration/spark/src/main/scala/org/apache/carbondata/spark/util/CommonUtil.scala
##########
@@ -1011,38 +1011,46 @@ object CommonUtil {
       objectDataType: DataType): AnyRef = {
     objectDataType match {
       case _: ArrayType =>
-        val arrayDataType = objectDataType.asInstanceOf[ArrayType]
-        val arrayData = data.asInstanceOf[UnsafeArrayData]
-        val size = arrayData.numElements()
-        val childDataType = arrayDataType.elementType
-        val arrayChildObjects = new Array[AnyRef](size)
-        var i = 0
-        while (i < size) {
-          arrayChildObjects(i) = convertSparkComplexTypeToCarbonObject(arrayData.get(i,
-            childDataType), childDataType)
-          i = i + 1
+        if (data == null) {

Review comment:
       after line 1011, add a check that if data type is array or struct or map and the data is null, return null to avoid changing many lines.




----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

ajantha-bhat commented on a change in pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#discussion_r503099172



##########
File path: integration/spark/src/test/scala/org/apache/carbondata/spark/testsuite/allqueries/InsertIntoCarbonTableTestCase.scala
##########
@@ -67,6 +67,21 @@ class InsertIntoCarbonTableTestCase extends QueryTest with BeforeAndAfterAll {
 
   }
 
+  test("insert from orc-select columns with columns having null values and sort scope as global sort") {
+    sql("drop table if exists TORCSource")
+    sql("drop table if exists TCarbon")
+    sql("create table TORCSource(name string,col array<String>,fee int) STORED AS orc")
+    sql("insert into TORCSource values('karan',null,2)")
+    sql("create table TCarbon(name string, col array<String>,fee int) STORED AS carbondata TBLPROPERTIES ('SORT_COLUMNS'='name','TABLE_BLOCKSIZE'='128','TABLE_BLOCKLET_SIZE'='128','SORT_SCOPE'='global_SORT')")
+    sql("insert overwrite table TCarbon select name,col,fee from TORCSource")
+    val result = sql("show segments for table TCarbon").collect()(0).get(1).toString()
+    if(!"Success".equalsIgnoreCase(result)) {
+      assert(false)
+    }
+    sql("drop table if exists TORCSource")
+    sql("drop table if exists TCarbon")
+  }
+

Review comment:
       please handle and verify 4 scenarios by comparing with ORC.
   a) local sort insert with complex type null value data
   b) global sort insert with complex type null value data
   c) same point a) with `carbon.enable.bad.record.handling.for.insert `as `true`
   d) same point b) with `carbon.enable.bad.record.handling.for.insert` as `true`




----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

CarbonDataQA1 commented on pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#issuecomment-706972701


   Build Success with Spark 2.4.5, Please check CI http://121.244.95.60:12545/job/ApacheCarbon_PR_Builder_2.4.5/2622/
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

Karan980 commented on a change in pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#discussion_r503139533



##########
File path: integration/spark/src/main/scala/org/apache/carbondata/spark/util/CommonUtil.scala
##########
@@ -1011,38 +1011,46 @@ object CommonUtil {
       objectDataType: DataType): AnyRef = {
     objectDataType match {
       case _: ArrayType =>
-        val arrayDataType = objectDataType.asInstanceOf[ArrayType]
-        val arrayData = data.asInstanceOf[UnsafeArrayData]
-        val size = arrayData.numElements()
-        val childDataType = arrayDataType.elementType
-        val arrayChildObjects = new Array[AnyRef](size)
-        var i = 0
-        while (i < size) {
-          arrayChildObjects(i) = convertSparkComplexTypeToCarbonObject(arrayData.get(i,
-            childDataType), childDataType)
-          i = i + 1
+        if (data == null) {

Review comment:
       For dateType if data is null that is handled by some other way which is not returning null. That's why i didn't put it after line 1011




----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

Karan980 commented on a change in pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#discussion_r503140543



##########
File path: integration/spark/src/main/scala/org/apache/carbondata/spark/util/CommonUtil.scala
##########
@@ -1011,38 +1011,46 @@ object CommonUtil {
       objectDataType: DataType): AnyRef = {
     objectDataType match {
       case _: ArrayType =>
-        val arrayDataType = objectDataType.asInstanceOf[ArrayType]
-        val arrayData = data.asInstanceOf[UnsafeArrayData]
-        val size = arrayData.numElements()
-        val childDataType = arrayDataType.elementType
-        val arrayChildObjects = new Array[AnyRef](size)
-        var i = 0
-        while (i < size) {
-          arrayChildObjects(i) = convertSparkComplexTypeToCarbonObject(arrayData.get(i,
-            childDataType), childDataType)
-          i = i + 1
+        if (data == null) {

Review comment:
       For dateType if data is null that is handled by some other way which is not returning null. That's why i didn't put it after line 1011




----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

CarbonDataQA1 commented on pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#issuecomment-706985772


   Build Success with Spark 2.3.4, Please check CI http://121.244.95.60:12545/job/ApacheCarbonPRBuilder2.3/4372/
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

CarbonDataQA1 commented on pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#issuecomment-708398610


   Build Failed  with Spark 2.3.4, Please check CI http://121.244.95.60:12545/job/ApacheCarbonPRBuilder2.3/4436/
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

CarbonDataQA1 commented on pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#issuecomment-708407338


   Build Failed  with Spark 2.4.5, Please check CI http://121.244.95.60:12545/job/ApacheCarbon_PR_Builder_2.4.5/2683/
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

CarbonDataQA1 commented on pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#issuecomment-712395039






----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

CarbonDataQA1 commented on pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#issuecomment-712748032


   Build Failed  with Spark 2.4.5, Please check CI http://121.244.95.60:12545/job/ApacheCarbon_PR_Builder_2.4.5/2778/
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

CarbonDataQA1 commented on pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#issuecomment-712748210


   Build Failed  with Spark 2.3.4, Please check CI http://121.244.95.60:12545/job/ApacheCarbonPRBuilder2.3/4532/
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

QiangCai commented on pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#issuecomment-712799969


   retest this please


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

CarbonDataQA1 commented on pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#issuecomment-712853433


   Build Failed  with Spark 2.4.5, Please check CI http://121.244.95.60:12545/job/ApacheCarbon_PR_Builder_2.4.5/2782/
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

CarbonDataQA1 commented on pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#issuecomment-712854016


   Build Failed  with Spark 2.3.4, Please check CI http://121.244.95.60:12545/job/ApacheCarbonPRBuilder2.3/4536/
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

ydvpankaj99 commented on pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#issuecomment-713229757


   retest this please


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

CarbonDataQA1 commented on pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#issuecomment-713303716


   Build Success with Spark 2.3.4, Please check CI http://121.244.95.60:12545/job/ApacheCarbonPRBuilder2.3/4558/
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

Karan980 commented on a change in pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#discussion_r509000503



##########
File path: integration/spark/src/test/scala/org/apache/carbondata/spark/testsuite/allqueries/InsertIntoCarbonTableTestCase.scala
##########
@@ -67,6 +67,21 @@ class InsertIntoCarbonTableTestCase extends QueryTest with BeforeAndAfterAll {
 
   }
 
+  test("insert from orc-select columns with columns having null values and sort scope as global sort") {
+    sql("drop table if exists TORCSource")
+    sql("drop table if exists TCarbon")
+    sql("create table TORCSource(name string,col array<String>,fee int) STORED AS orc")
+    sql("insert into TORCSource values('karan',null,2)")
+    sql("create table TCarbon(name string, col array<String>,fee int) STORED AS carbondata TBLPROPERTIES ('SORT_COLUMNS'='name','TABLE_BLOCKSIZE'='128','TABLE_BLOCKLET_SIZE'='128','SORT_SCOPE'='global_SORT')")
+    sql("insert overwrite table TCarbon select name,col,fee from TORCSource")
+    val result = sql("show segments for table TCarbon").collect()(0).get(1).toString()
+    if(!"Success".equalsIgnoreCase(result)) {

Review comment:
       Done

##########
File path: integration/spark/src/main/scala/org/apache/carbondata/spark/util/CommonUtil.scala
##########
@@ -1011,38 +1011,46 @@ object CommonUtil {
       objectDataType: DataType): AnyRef = {
     objectDataType match {
       case _: ArrayType =>
-        val arrayDataType = objectDataType.asInstanceOf[ArrayType]
-        val arrayData = data.asInstanceOf[UnsafeArrayData]
-        val size = arrayData.numElements()
-        val childDataType = arrayDataType.elementType
-        val arrayChildObjects = new Array[AnyRef](size)
-        var i = 0
-        while (i < size) {
-          arrayChildObjects(i) = convertSparkComplexTypeToCarbonObject(arrayData.get(i,
-            childDataType), childDataType)
-          i = i + 1
+        if (data == null) {

Review comment:
       Done

##########
File path: integration/spark/src/test/scala/org/apache/carbondata/spark/testsuite/allqueries/InsertIntoCarbonTableTestCase.scala
##########
@@ -67,6 +67,21 @@ class InsertIntoCarbonTableTestCase extends QueryTest with BeforeAndAfterAll {
 
   }
 
+  test("insert from orc-select columns with columns having null values and sort scope as global sort") {
+    sql("drop table if exists TORCSource")
+    sql("drop table if exists TCarbon")
+    sql("create table TORCSource(name string,col array<String>,fee int) STORED AS orc")
+    sql("insert into TORCSource values('karan',null,2)")
+    sql("create table TCarbon(name string, col array<String>,fee int) STORED AS carbondata TBLPROPERTIES ('SORT_COLUMNS'='name','TABLE_BLOCKSIZE'='128','TABLE_BLOCKLET_SIZE'='128','SORT_SCOPE'='global_SORT')")
+    sql("insert overwrite table TCarbon select name,col,fee from TORCSource")
+    val result = sql("show segments for table TCarbon").collect()(0).get(1).toString()
+    if(!"Success".equalsIgnoreCase(result)) {
+      assert(false)
+    }
+    sql("drop table if exists TORCSource")
+    sql("drop table if exists TCarbon")
+  }
+

Review comment:
       Done




----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]

[GitBox]

ajantha-bhat commented on a change in pull request #3979:
URL: https://github.com/apache/carbondata/pull/3979#discussion_r509026576



##########
File path: integration/spark/src/main/scala/org/apache/carbondata/spark/util/CommonUtil.scala
##########
@@ -1009,6 +1009,10 @@ object CommonUtil {
 
   private def convertSparkComplexTypeToCarbonObject(data: AnyRef,
       objectDataType: DataType): AnyRef = {
+    if (data == null && (objectDataType.isInstanceOf[ArrayType]

Review comment:
       check with  objectDataType.isComplexType




----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[hidden email]