CODEWITHSUNDEEP
javahadoophbase
Mr37037
Mr37037

Reputation: 748

Why SingleColumnValueFilter do not returns true answer on < and > operator?

Hi I am new to hbase and in phase of learning it. I created a table using hbase shell and put values into that table uing shell. But now when i want to get thos values using some filter like this: 

HTable table = new HTable(conf, "test");

        List<Filter> filters = new ArrayList<Filter>();
        SingleColumnValueFilter colValFilter = new SingleColumnValueFilter(Bytes.toBytes("col"), Bytes.toBytes("age")
                , CompareFilter.CompareOp.GREATER, new BinaryComparator(Bytes.toBytes("118")));
        colValFilter.setFilterIfMissing(false);
        filters.add(colValFilter); 
        FilterList fl = new FilterList( FilterList.Operator.MUST_PASS_ALL, filters);

        Scan s = new Scan();
        s.setFilter(fl);
        ResultScanner ss = table.getScanner(s);
        JSONArray arr=new JSONArray();
        for(Result r:ss){
            JSONObject obj=new JSONObject();
            for(KeyValue kv : r.raw()){

                String column=new String(kv.getQualifier());
                String value=new String(kv.getValue());
                obj.accumulate(column, value);

            }
            arr.put(obj);
        }

Now when i run this program i get different results. Like i get users whose age are less than 118 but still they get displayed.

My Question:

Why SingleColumnValueFilter do this? Any other way to solve this problem?

Upvotes: 0

Views: 1210

Answers (1)

Averman
Averman

Reputation: 441

If you put data from hbase shell like put 'test','r1','col:age','19' it will be stored as String-bytes so when you compare it with > on BinaryComparator, it will compare lexicographicly and return any string that starts with character > than '1'

you should put it put 'test','r1','col:age','\x13' (hexadecimal value) if you want to put it as integer-bytes

Upvotes: 1

Related Questions