[ https://issues.apache.org/jira/browse/HIVE-7471?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14070488#comment-14070488 ]
Xuefu Zhang commented on HIVE-7471: ----------------------------------- [~rthapar] Decimal data type doesn't really work well in Hive 0.11. It has since been enhanced quite lot. Thus, you should looking into Hive 0.13 instead. By the way, JIRA isn't a place to ask questions, but to report problem or request features. For usage questions, hive user list is the best option. > testing for equality of decimal columns does not work > ----------------------------------------------------- > > Key: HIVE-7471 > URL: https://issues.apache.org/jira/browse/HIVE-7471 > Project: Hive > Issue Type: Bug > Components: Database/Schema > Environment: x86_64 x86_64 x86_64 GNU/Linux > Reporter: Raj Thapar > > I am using Hive version 0.11 and am facing following problem, > I have 2 tables: venus_all_min_prices and venus_all_prices. They have a > column bp_price_paid having type of decimal. When I try to query for equality > on this column between 2 tables, I don't get any results. However if I use > one specific value and use 2 conditions against this value (one for each > column anded together), it does return results. > ie > 1. venus_all_min_prices.bp_price_paid = venus_all_prices.bp_price_paid does > not return any values > 2. venus_all_min_prices.bp_price_paid = 59.99 and > venus_all_prices.bp_price_paid = 59.99: returns results > What should I do to make (1) work? > My table definitions are below: > CREATE TABLE venus_all_min_prices( > bp_price_paid decimal, > opr_sty_clr_cd string) > PARTITIONED BY ( > partition_timestamp string) > ROW FORMAT DELIMITED > FIELDS TERMINATED BY ',' > STORED AS INPUTFORMAT > 'org.apache.hadoop.mapred.TextInputFormat' > OUTPUTFORMAT > 'org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat' > LOCATION > > 'hdfs://host:8020/user/user/database/temp_data_location/venus_all_min_prices' > TBLPROPERTIES ( > 'numPartitions'='1', > 'numFiles'='1', > 'transient_lastDdlTime'='1406040417', > 'numRows'='0', > 'totalSize'='2507', > 'rawDataSize'='0') > CREATE TABLE venus_all_prices( > bp_price_paid decimal, > ord_key bigint, > oms_ord_ln_key string, > opr_sty_clr_cd string) > PARTITIONED BY ( > partition_timestamp string) > ROW FORMAT DELIMITED > FIELDS TERMINATED BY ',' > STORED AS INPUTFORMAT > 'org.apache.hadoop.mapred.TextInputFormat' > OUTPUTFORMAT > 'org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat' > LOCATION > 'hdfs://host:8020/user/user/database/temp_data_location/venus_all_prices' > TBLPROPERTIES ( > 'numPartitions'='3', > 'numFiles'='11', > 'transient_lastDdlTime'='1405979150', > 'numRows'='0', > 'totalSize'='4845600', > 'rawDataSize'='0') -- This message was sent by Atlassian JIRA (v6.2#6252)