[GitHub] [carbondata] nihal0107 commented on a change in pull request #3865: [CARBONDATA-3928] Handled the Strings which length is greater than 32000 as a bad record.

[GitBox]

nihal0107 commented on a change in pull request #3865:
URL: https://github.com/apache/carbondata/pull/3865#discussion_r472122959



##########
File path: integration/spark/src/test/scala/org/apache/carbondata/integration/spark/testsuite/dataload/TestLoadDataGeneral.scala
##########
@@ -145,47 +150,162 @@ class TestLoadDataGeneral extends QueryTest with BeforeAndAfterEach {
     sql("drop table if exists carbon_table")
   }
 
-  test("test insert / update with data more than 32000 characters") {
+  test("test load / insert / update with data more than 32000 characters and bad record action as Redirect") {
+    val testdata =s"$resourcesPath/MoreThan32KChar.csv"
+    FileFactory.deleteAllFilesOfDir(new File(CarbonProperties.getInstance()
+      .getProperty(CarbonCommonConstants.CARBON_BADRECORDS_LOC)))
+    sql("CREATE TABLE longerthan32kchar(dim1 String, dim2 String, mes1 int) STORED AS carbondata")
+    sql(s"LOAD DATA LOCAL INPATH '$testdata' into table longerThan32kChar OPTIONS('FILEHEADER'='dim1,dim2,mes1', " +
+      s"'BAD_RECORDS_ACTION'='REDIRECT','BAD_RECORDS_LOGGER_ENABLE'='TRUE')")
+    var redirectCsvPath = getRedirectCsvPath("default", "longerthan32kchar", "0", "0")
+    assert(checkRedirectedCsvContentAvailableInSource(testdata, redirectCsvPath))
+    val longChar: String = RandomStringUtils.randomAlphabetic(33000)
+
     CarbonProperties.getInstance()
       .addProperty(CarbonCommonConstants.CARBON_ENABLE_BAD_RECORD_HANDLING_FOR_INSERT, "true")
-    val testdata =s"$resourcesPath/32000char.csv"
-    sql("drop table if exists load32000chardata")
-    sql("drop table if exists load32000chardata_dup")
-    sql("CREATE TABLE load32000chardata(dim1 String, dim2 String, mes1 int) STORED AS carbondata")
-    sql("CREATE TABLE load32000chardata_dup(dim1 String, dim2 String, mes1 int) STORED AS carbondata")
-    sql(s"LOAD DATA LOCAL INPATH '$testdata' into table load32000chardata OPTIONS('FILEHEADER'='dim1,dim2,mes1')")
+    CarbonProperties.getInstance()
+      .addProperty(CarbonCommonConstants.CARBON_BAD_RECORDS_ACTION, "REDIRECT");
+    sql(s"insert into longerthan32kchar values('33000', '$longChar', 4)")
+    checkAnswer(sql("select * from longerthan32kchar"), Seq(Row("ok", "hi", 1), Row("itsok", "hello", 2)))
+    redirectCsvPath = getRedirectCsvPath("default", "longerthan32kchar", "1", "0")
+    var redirectedFileLineList = FileUtils.readLines(redirectCsvPath)
+    var iterator = redirectedFileLineList.iterator()
+    while (iterator.hasNext) {
+      assert(iterator.next().equals("33000,"+longChar+",4"))
+    }
+
+    // Update strings of length greater than 32000
+    sql(s"update longerthan32kchar set(longerthan32kchar.dim2)=('$longChar') " +
+      "where longerthan32kchar.mes1=1").show()
+    checkAnswer(sql("select * from longerthan32kchar"), Seq(Row("itsok", "hello", 2)))
+    redirectCsvPath = getRedirectCsvPath("default", "longerthan32kchar", "0", "1")
+    redirectedFileLineList = FileUtils.readLines(redirectCsvPath)
+    iterator = redirectedFileLineList.iterator()
+    while (iterator.hasNext) {
+      assert(iterator.next().equals("ok,"+longChar+",1"))
+    }
+    CarbonProperties.getInstance()
+      .addProperty(CarbonCommonConstants.CARBON_ENABLE_BAD_RECORD_HANDLING_FOR_INSERT, "false")
+
+    // Insert longer string without converter step will throw exception
     intercept[Exception] {
-      sql("insert into load32000chardata_dup select dim1,concat(load32000chardata.dim2,'aaaa'),mes1 from load32000chardata").show()
+      sql(s"insert into longerthan32kchar values('32000', '$longChar', 3)")
     }
-    sql(s"LOAD DATA LOCAL INPATH '$testdata' into table load32000chardata_dup OPTIONS('FILEHEADER'='dim1,dim2,mes1')")
+
+    FileFactory.deleteAllFilesOfDir(new File(CarbonProperties.getInstance()
+      .getProperty(CarbonCommonConstants.CARBON_BADRECORDS_LOC)))
+  }
+
+  test("test load / insert / update with data more than 32000 characters and bad record action as Force") {
+    val testdata =s"$resourcesPath/MoreThan32KChar.csv"
+    sql("CREATE TABLE longerthan32kchar(dim1 String, dim2 String, mes1 int) STORED AS carbondata")
+    sql(s"LOAD DATA LOCAL INPATH '$testdata' into table longerThan32kChar OPTIONS('FILEHEADER'='dim1,dim2,mes1', " +
+      s"'BAD_RECORDS_ACTION'='FORCE','BAD_RECORDS_LOGGER_ENABLE'='TRUE')")
+    checkAnswer(sql("select * from longerthan32kchar"), Seq(Row("ok", "hi", 1), Row("itsok", "hello", 2), Row("32123", null, 3)))
+    val longChar: String = RandomStringUtils.randomAlphabetic(33000)
+
+    CarbonProperties.getInstance()
+      .addProperty(CarbonCommonConstants.CARBON_ENABLE_BAD_RECORD_HANDLING_FOR_INSERT, "true")
+    CarbonProperties.getInstance()
+      .addProperty(CarbonCommonConstants.CARBON_BAD_RECORDS_ACTION, "FORCE");
+    sql(s"insert into longerthan32kchar values('33000', '$longChar', 4)")
+    checkAnswer(sql("select * from longerthan32kchar"),
+      Seq(Row("ok", "hi", 1), Row("itsok", "hello", 2), Row("32123", null, 3), Row("33000", null, 4)))
+
+    // Update strings of length greater than 32000
+    sql(s"update longerthan32kchar set(longerthan32kchar.dim2)=('$longChar') " +
+      "where longerthan32kchar.mes1=1").show()
+    checkAnswer(sql("select * from longerthan32kchar"),
+      Seq(Row("ok", null, 1), Row("itsok", "hello", 2), Row("32123", null, 3), Row("33000", null, 4)))
+    CarbonProperties.getInstance()
+      .addProperty(CarbonCommonConstants.CARBON_ENABLE_BAD_RECORD_HANDLING_FOR_INSERT, "false")
+
+    // Insert longer string without converter step will throw exception
     intercept[Exception] {

Review comment:
       same as above.




----------------------------------------------------------------
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]