Client javadoc suggestion: add code examples about obtaining historical values
-------------------------------------------------------------------------------
Key: HBASE-2004
URL: https://issues.apache.org/jira/browse/HBASE-2004
Project: Hadoop HBase
Issue Type: Improvement
Components: client
Affects Versions: 0.20.1
Reporter: Doug Meil
Priority: Minor
The implicit support HBase provides for versioning of values is very powerful,
but it's not all that obvious for application programmers to use it to obtain
the historical values.
I would like to suggest adding some comments and sample code to the Result
class (org.apache.hadoop.hbase.client.Result) Javadoc. I know this seems sort
of obvious to people that regularly use HBase, but I think that for new folks
having code examples available in Javadoc is helpful because it's "one stop
shopping" for documentation (i.e., as opposed to looking to an external
writeup). Arguably, this type of example could also go in the HTable class
javadoc.
e.g.,....
HTable table = new HTable(config, "mytable");
Scan scan = new Scan(); // no arguments indicates will scan all rows
scan.setMaxVersions( 5 ); // setting this to 1 only returns current
version
ResultScanner rs = table.getScanner(scan);
for (Iterator<Result> i = rs.iterator(); i.hasNext(); ) {
Result r = i.next();
// obtains current value from 'family:column'
byte b[] = r.getValue( Bytes.toBytes("family"), Bytes.toBytes("column")
);
KeyValue kv[] = r.raw();
for (int j = 0; j < kv.length; j++) {
byte bv[] = kv[j].getValue();
// this loop returns both current and historical values
byte bc[] = kv[j].getColumn();
// returns 'family:column'
}
}
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
