[Moving conversation to sqoop-user@incubator.apache.org. Please subscribe to that list.]
We are working on SQOOP-342 which will allow you to override type mappings. Once it is committed, you will be able to specify your own mappings for individual columns. https://issues.apache.org/jira/browse/SQOOP-342 Thanks, Arvind On Fri, Oct 7, 2011 at 9:53 AM, Walter Chang <weidezhang2...@gmail.com>wrote: > Hi, > > It seems when sqoop trying to import tiny(1) , it goes to boolean > type in hive. However, tiny(1) in my mysql table is really a number. > How can I disable the boolean type conversion and make it a number. > Following is the example: (day_num field). > > here are version info: > mysql Ver 14.14 Distrib 5.1.41, for debian-linux-gnu (i486) using > readline 6.1 > mysql-connector-java-5.1.18 > Sqoop 1.3.0-cdh3u1 > > mysql> select * from xxx limit 5; > +----+---------+------------+---------------------+------------- > +--------+------------+----------+ > | ID | DAY_NUM | ENT_SEQ_NO | CREATE_DATE | UPDATE_DATE | > SOURCE | START_TIME | END_TIME | > +----+---------+------------+---------------------+------------- > +--------+------------+----------+ > | 0 | 0 | 1 | 2011-06-15 06:54:38 | NULL | > 2.0.4 | 00:00:00 | 23:59:59 | > | 0 | 1 | 1 | 2011-06-15 06:54:38 | NULL | > 2.0.4 | 00:00:00 | 23:59:59 | > | 0 | 2 | 1 | 2011-06-15 06:54:38 | NULL | > 2.0.4 | 00:00:00 | 23:59:59 | > | 0 | 3 | 1 | 2011-06-15 06:54:38 | NULL | > 2.0.4 | 00:00:00 | 23:59:59 | > | 0 | 4 | 1 | 2011-06-15 06:54:38 | NULL | > 2.0.4 | 00:00:00 | 23:59:59 | > +----+---------+------------+---------------------+------------- > +--------+------------+----------+ > 5 rows in set (0.00 sec) > > mysql> describe day_parting > -> ; > +-------------+---------------------+------+-----+---------+-------+ > | Field | Type | Null | Key | Default | Extra | > +-------------+---------------------+------+-----+---------+-------+ > | ID | int(10) unsigned | NO | PRI | 0 | | > | DAY_NUM | tinyint(1) unsigned | NO | PRI | 7 | | > | ENT_SEQ_NO | tinyint(2) unsigned | NO | PRI | 1 | | > | CREATE_DATE | datetime | NO | | NULL | | > | UPDATE_DATE | datetime | YES | | NULL | | > | SOURCE | varchar(32) | NO | | | | > | START_TIME | time | NO | | NULL | | > | END_TIME | time | NO | | NULL | | > +-------------+---------------------+------+-----+---------+-------+ > 8 rows in set (0.00 sec) > > However, after sqoop importing, Hive shows : > > id int > day_num boolean > ent_seq_no tinyint > create_date string > update_date string > source string > start_time string > end_time string > > 0 false 1 2011-06-15 06:54:38.0 null 2.0.4 00:00:00 > 23:59:59 > 7 false 1 2011-03-23 07:38:39.0 2011-03-23 07:38:39.0 > IO#DishROI32011 > 12:30:00 15:30:00 > 16 false 1 2011-04-01 09:31:28.0 2011-04-01 09:31:28.0 > IO#KDA032011 > 00:00:00 17:00:00 > 17 false 1 2011-04-01 09:33:20.0 2011-04-01 09:33:20.0 > IO#Direc20110215 17:00:00 23:59:00 > 19 false 1 2011-04-12 04:46:43.0 2011-04-12 04:46:43.0 > IO#buycalls2011 > 12:30:00 15:30:00 > > -- > NOTE: The mailing list sqoop-u...@cloudera.org is deprecated in favor of > Apache Sqoop mailing list sqoop-user@incubator.apache.org. Please > subscribe to it by sending an email to > incubator-sqoop-user-subscr...@apache.org. >
