You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Jonathan Hsieh (JIRA)" <ji...@apache.org> on 2013/12/17 17:11:14 UTC
[jira] [Comment Edited] (HBASE-9970) HBase BulkLoad, table is
creating with the timestamp key also as a column to the table.
[ https://issues.apache.org/jira/browse/HBASE-9970?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13850612#comment-13850612 ]
Jonathan Hsieh edited comment on HBASE-9970 at 12/17/13 4:09 PM:
-----------------------------------------------------------------
The Magic column names in ImportTsv are not documented http://hbase.apache.org/book/ops_mgt.html#importtsv
There is a small section where the usage instructions from an older version is present.
Mind filing a follow-on issue and submitting a patch to document this so that users know they can do this and that it is expected behavior?
was (Author: jmhsieh):
The Magic column names in importtvs are not documented http://hbase.apache.org/book/ops_mgt.html#importtsv
There is a small section where the usage instructions from an older version is present.
Mind filing a follow-on issue and submitting a patch to document this so that users know they can do this and that it is expected behavior?
> HBase BulkLoad, table is creating with the timestamp key also as a column to the table.
> ----------------------------------------------------------------------------------------
>
> Key: HBASE-9970
> URL: https://issues.apache.org/jira/browse/HBASE-9970
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.94.11
> Reporter: Y. SREENIVASULU REDDY
> Assignee: Y. SREENIVASULU REDDY
> Fix For: 0.98.0, 0.96.1, 0.94.14
>
> Attachments: HBASE-9970.java.patch, HBASE-9970_94.patch
>
>
> If BulkLoad job is running with out creating a table.
> job itself will create the table if table is not found.
> {code}
> if (!doesTableExist(tableName)) {
> createTable(conf, tableName);
> }
> {code}
> if columns contains timestamp also then table is creating with defined columns and timestamp key.
> {quote}
> eg: -Dimporttsv.columns=HBASE_ROW_KEY,HBASE_TS_KEY,d:num
> {quote}
> table is creating with the following columnFamilies.
> 'HBASE_TS_KEY' and 'd'
> while iterating timestamp key also need to avoid while describing the column descriptors.
> {code}
> private static void createTable(HBaseAdmin admin, String tableName, String[] columns)
> throws IOException {
> HTableDescriptor htd = new HTableDescriptor(TableName.valueOf(tableName));
> Set<String> cfSet = new HashSet<String>();
> for (String aColumn : columns) {
> if (TsvParser.ROWKEY_COLUMN_SPEC.equals(aColumn)) continue;
> // we are only concerned with the first one (in case this is a cf:cq)
> cfSet.add(aColumn.split(":", 2)[0]);
> }
> for (String cf : cfSet) {
> HColumnDescriptor hcd = new HColumnDescriptor(Bytes.toBytes(cf));
> htd.addFamily(hcd);
> }
> LOG.warn(format("Creating table '%s' with '%s' columns and default descriptors.",
> tableName, cfSet));
> admin.createTable(htd);
> }
> {code}
> {code}
> Index: hbase-server/src/main/java/org/apache/hadoop/hbase/mapreduce/ImportTsv.java
> ===================================================================
> --- hbase-server/src/main/java/org/apache/hadoop/hbase/mapreduce/ImportTsv.java (revision 1539967)
> +++ hbase-server/src/main/java/org/apache/hadoop/hbase/mapreduce/ImportTsv.java (working copy)
> @@ -413,7 +413,8 @@
> HTableDescriptor htd = new HTableDescriptor(TableName.valueOf(tableName));
> Set<String> cfSet = new HashSet<String>();
> for (String aColumn : columns) {
> - if (TsvParser.ROWKEY_COLUMN_SPEC.equals(aColumn)) continue;
> + if (TsvParser.ROWKEY_COLUMN_SPEC.equals(aColumn)
> + || TsvParser.TIMESTAMPKEY_COLUMN_SPEC.equals(aColumn)) continue;
> // we are only concerned with the first one (in case this is a cf:cq)
> cfSet.add(aColumn.split(":", 2)[0]);
> }
> {code}
--
This message was sent by Atlassian JIRA
(v6.1.4#6159)