Internally, we call a UDF to generate the TupleId and based on that tupleId we decide if the row is duplicated or not. You can run a slightly tweaked query
" select a.remark,a.id, getTupleId() as TupleId from c_indextest1 a, c_indextest2 b where a.id=b.id; " In order to prompt Multiple Input Rows most probably the getTupleId() will give duplicate entries. In between your given query runs fine with the data i have generated in house. So this is data dependent. So it will be good if we get *origin_data* or the script which has build the data for origin_data. On Fri, Oct 20, 2017 at 7:28 AM, yixu2001 <[email protected]> wrote: > dev > one row record > > > yixu2001 > > From: Liang Chen > Date: 2017-10-19 22:26 > To: dev > Subject: Re: Re: Update statement failed with "Multiple input rows matched > for same row" in version 1.2.0, > Hi > > Execute the below query, return one row record or multiple row records ? > --------------------------------- > select a.remark from c_indextest1 a where a.id=b.id > > Regards > Liang > > > yixu2001 wrote > > dev > > You can follow the steps below to reproduce the problem. > > tables c_indextest2 has 1700w records and table c_indextest1 has about > 30w > > records. > > > > step 1: > > cc.sql("CREATE TABLE IF NOT EXISTS c_indextest2 (id string ,remark > string) > > STORED BY 'carbondata'").show; > > > > step 2: origin_data is a existing table with 1700w records, and fields in > > the table do not matter. > > cc.sql("insert into c_indextest2 select row_number() over(partition by > > a.PKid order by a.pkid) id,a.remark from (SELECT '1' PKID,'dfsdd' remark > > from origin_data limit 17000000) a").show; > > > > step 3: > > cc.sql("CREATE TABLE IF NOT EXISTS c_indextest1 (id string ,remark > string) > > STORED BY 'carbondata'").show; > > > > step 4: > > cc.sql("insert into c_indextest1 select * from c_indextest2 where > > pmod(cast(ID as int),50)=43").show; > > > > setp 5: > > cc.sql("update c_indextest2 b set (b.remark)=(select a.remark from > > c_indextest1 a where a.id=b.id)").show; > > > > > > yixu2001 > > > > From: sounak > > Date: 2017-10-19 18:26 > > To: dev > > Subject: Re: Re: Update statement failed with "Multiple input rows > matched > > for same row" in version 1.2.0, > > Hi Yixu, > > > > Thanks for the ddl. But i am particularly interested on the data in order > > to reproduce the problem. Will it be feasible to share to data or if it > is > > huge then the scripts that generates it? I cannot access the data > location > > from public network. > > > > > > On Thu, Oct 19, 2017 at 3:00 PM, yixu2001 < > > > yixu2001@ > > > > wrote: > > > >> dev > >> > >> Step 1:I make a hive table qqdata2.h_indextest1: > >> CREATE EXTERNAL TABLE `qqdata2.h_indextest1`( > >> `id` INT, > >> `CUST_ORDER_ID` STRING, > >> `ORDER_ITEM_IDATTR_ID` STRING, > >> `ATTR_VALUE_IDATTR_VALUE` STRING, > >> `CREATE_DATE` STRING, > >> `UPDATE_DATE` STRING, > >> `STATUS_CD` STRING, > >> `STATUS_DATE` STRING, > >> `AREA_ID` STRING, > >> `REGION_CD` STRING, > >> `UPDATE_STAFF` STRING, > >> `CREATE_STAFF` STRING, > >> `SHARDING_ID` STRING, > >> `ORDER_ATTR_ID` STRING ) > >> ROW FORMAT DELIMITED > >> FIELDS TERMINATED BY '\u0001' > >> LINES TERMINATED BY '\n' > >> NULL DEFINED AS '' > >> STORED AS INPUTFORMAT > >> 'org.apache.hadoop.mapred.TextInputFormat' > >> OUTPUTFORMAT > >> 'org.apache.hadoop.hive.ql.io > >> > >> .HiveIgnoreKeyTextOutputFormat' > >> LOCATION > >> 'hdfs://hdp78.ffcs.cn:8020/user/bigdata/streamcql/dist1 > >> > >> '; > >> > >> there are csv files in hdfs dir user/bigdata/streamcql/dist1, the data > >> format in the csv file just as following > >> > >> 1939505130,171483932,287305502,813463930,20160709134396669, > >> 201607101469099594,1299,20160711996575390,10,73, > 302063,302064,127859875, > >> 9999999 > >> > >> Step 2:I make a carbon table qqdata2.c_indextest1: > >> cc.sql("CREATE TABLE IF NOT EXISTS qqdata2.c_indextest1 (id STRING, > >> CUST_ORDER_ID STRING,ORDER_ITEM_IDATTR_ID STRING,ATTR_VALUE_IDATTR_VALUE > >> STRING,CREATE_DATE STRING,UPDATE_DATE STRING,STATUS_CD > STRING,STATUS_DATE > >> STRING,AREA_ID STRING,REGION_CD STRING,UPDATE_STAFF STRING,CREATE_STAFF > >> STRING,SHARDING_ID STRING,ORDER_ATTR_ID STRING) STORED BY 'carbondata' > ") > >> > >> Step 3:Insert data: > >> cc.sql("insert into qqdata2.c_indextest1 select * from > >> qqdata2.h_indextest1").show(100,false); > >> > >> Step 4: Repeat from step1 to step 3, I make another carbon table > >> qqdata2.c_indextest2 > >> The record number of qqdata2.c_indextest1 is 30w, the record number of > >> qqdata2.c_indextest2 is 1700w. > >> > >> > >> > >> yixu2001 > >> > >> From: sounak > >> Date: 2017-10-19 14:13 > >> To: dev > >> Subject: Re: Update statement failed with "Multiple input rows matched > >> for > >> same row" in version 1.2.0, > >> Hi Yixu, > >> > >> Can you please share the DDLs and the Data for the above problem with > us? > >> > >> Thanks > >> Sounak > >> > >> On Wed, Oct 18, 2017 at 12:44 PM, yixu2001 < > > > yixu2001@ > > > > wrote: > >> > >> > dev > >> > > >> > > >> > In carbondata version 1.2.0, I execute "update" statement with > >> sub-query, > >> > it failed. > >> > All the rows in the 2 tables are not duplicated, and the same > statement > >> > will succeed in carbondata version 1.1.1. > >> > > >> > The test log as following: > >> > scala> cc.sql("select count(*), count(distinct id) from > >> > qqdata2.c_indextest1").show(100,false); > >> > +--------+------------------+ > >> > |count(1)|count(DISTINCT id)| > >> > +--------+------------------+ > >> > |300000 |300000 | > >> > +--------+------------------+ > >> > > >> > scala> cc.sql("select count(*), count(distinct id) from > >> > qqdata2.c_indextest2").show(100,false); > >> > +--------+------------------+ > >> > |count(1)|count(DISTINCT id)| > >> > +--------+------------------+ > >> > |71223220|71223220 | > >> > +--------+------------------+ > >> > > >> > scala> cc.sql("update qqdata2.c_indextest2 a > >> set(a.CUST_ORDER_ID,a.ORDER_ > >> > ITEM_IDATTR_ID,a.ATTR_VALUE_IDATTR_VALUE,a.CREATE_DATE,a. > >> > UPDATE_DATE,a.STATUS_CD,a.STATUS_DATE,a.AREA_ID,a. > >> > REGION_CD,a.UPDATE_STAFF,a.CREATE_STAFF,a.SHARDING_ID,a. > ORDER_ATTR_ID) > >> = > >> > (select b.CUST_ORDER_ID,b.ORDER_ITEM_IDATTR_ID,b.ATTR_VALUE_IDATTR_ > >> > VALUE,b.CREATE_DATE,b.UPDATE_DATE,b.STATUS_CD,b.STATUS_ > >> > DATE,b.AREA_ID,b.REGION_CD,b.UPDATE_STAFF,b.CREATE_STAFF,b. > >> SHARDING_ID,b.ORDER_ATTR_ID > >> > from qqdata2.c_indextest1 b where a.id = b.id)").show(100,false); > >> > 17/10/18 11:32:46 WARN Utils: Truncated the string representation of a > >> > plan since it was too large. This behavior can be adjusted by setting > >> > 'spark.debug.maxToStringFields' in SparkEnv.conf. > >> > 17/10/18 11:33:20 AUDIT deleteExecution$: [hdp84.ffcs.cn > >> > > >> > ][bigdata][Thread-1]Delete data operation is failed for > >> > qqdata2.c_indextest2 > >> > 17/10/18 11:33:20 ERROR deleteExecution$: main Delete data operation > is > >> > failed due to failure in creating delete delta file for segment : null > >> > block : null > >> > 17/10/18 11:33:20 ERROR ProjectForUpdateCommand$: main Exception in > >> update > >> > operationjava.lang.Exception: Multiple input rows matched for same > row. > >> > java.lang.RuntimeException: Update operation failed. Multiple input > >> rows > >> > matched for same row. > >> > at scala.sys.package$.error(package.scala:27) > >> > at org.apache.spark.sql.execution.command.ProjectForUpdateCommand. > >> > processData(IUDCommands.scala:239) > >> > at > >> org.apache.spark.sql.execution.command.ProjectForUpdateCommand.run( > >> > IUDCommands.scala:141) > >> > at org.apache.spark.sql.execution.command.ExecutedCommandExec. > >> > sideEffectResult$lzycompute(commands.scala:58) > >> > at org.apache.spark.sql.execution.command.ExecutedCommandExec. > >> > sideEffectResult(commands.scala:56) > >> > at org.apache.spark.sql.execution.command.ExecutedCommandExec. > >> > executeTake(commands.scala:71) > >> > at org.apache.spark.sql.execution.CollectLimitExec. > >> > executeCollect(limit.scala:38) > >> > at org.apache.spark.sql.Dataset$$anonfun$org$apache$spark$sql$ > >> > Dataset$$execute$1$1.apply(Dataset.scala:2378) > >> > at org.apache.spark.sql.execution.SQLExecution$.withNewExecutionId( > >> > SQLExecution.scala:57) > >> > at > >> org.apache.spark.sql.Dataset.withNewExecutionId(Dataset.scala:2780) > >> > at org.apache.spark.sql.Dataset.org > >> > > >> > $apache$spark$sql$Dataset$$execute$1(Dataset.scala:2377) > >> > at org.apache.spark.sql.Dataset.org > >> > > >> > $apache$spark$sql$Dataset$$collect(Dataset.scala:2384) > >> > at org.apache.spark.sql.Dataset$$anonfun$head$1.apply(Dataset. > >> > scala:2120) > >> > at org.apache.spark.sql.Dataset$$anonfun$head$1.apply(Dataset. > >> > scala:2119) > >> > at org.apache.spark.sql.Dataset.withTypedCallback(Dataset. > scala:2810) > >> > at org.apache.spark.sql.Dataset.head(Dataset.scala:2119) > >> > at org.apache.spark.sql.Dataset.take(Dataset.scala:2334) > >> > at org.apache.spark.sql.Dataset.showString(Dataset.scala:248) > >> > at org.apache.spark.sql.Dataset.show(Dataset.scala:640) > >> > ... 50 elided > >> > > >> > > >> > > >> > yixu2001 > >> > > >> > >> > >> > >> -- > >> Thanks > >> Sounak > >> > > > > > > > > -- > > Thanks > > Sounak > > > > > > -- > Sent from: http://apache-carbondata-dev-mailing-list-archive.1130556. > n5.nabble.com/ > -- Thanks Sounak
