HBase - Java Client Limit Scan Results Explanation?

Question

Is there a way to limit the total number of rows of sequential data when performing a scan of the data?

Notes:

• I'm working with 500,000 total rows
• I've tried both setMaxResultSize and setMaxResultsPerColumnFamily. This proves to be ineffectual (there does seem to be some behavior when both are set to low numbers or setMaxResultSize is larger. What is the relationship between these two functions?)
• I've worked with setting a PageFilter (size 10), and the behavior displays 5 different sequence data sets of 10.
• I actually got it sudo-working while typing this out by setting the PageFilter size and the setMaxResultSize equal. When I change either, it conforms to the PageFilter. It will also jump to another large subset of PageFilter size if I make setMaxResultSize significantly larger.
• HBase version is 1.1.1

Can someone better explain what is happening here and how to get the results I want?

Answer 1

you can use either hbase shell or hbase java client.

1- hbase shell: use this command and pipe the results to a file and do "wc -l ..."

count 'table_name',1

2- java hbase client api

long count=0;
String row="";

for (Result res : scanner)
{
    for (Cell cell : res.listCells())
    {
        row = new String(CellUtil.cloneRow(cell));
        if(!row.equals(""))
            count++;
    }
}