HBase shell scan 过滤器用法总结

比较器：
前面例子中的regexstring:2014-11-08.*、binary:x00x00x00x05，这都是比较器。HBase的filter有四种比较器：
（1）二进制比较器：如’binary:abc’，按字典排序跟’abc’进行比较
（2）二进制前缀比较器：如’binaryprefix:abc’，按字典顺序只跟’abc’比较前3个字符
（3）正则表达式比较器：如’regexstring:ab*yz’，按正则表达式匹配以ab开头，以yz结尾的值。这个比较器只能使用=、!=两个比较运算符。
（4）子串比较器：如’substring:abc123’，匹配以abc123开头的值。这个比较顺也只能使用=、!=两个比较运算符。

比较运算符：
HBase的filter中有7个比较运算符

1. LESS (<)

2. LESS_OR_EQUAL (<=)

3. EQUAL (=)

4. NOT_EQUAL (!=)

5. GREATER_OR_EQUAL (>=)

6. GREATER (>)

7. NO_OP (no operation)（不知道这个怎么用）

如果需要再bash直接执行查询，可以使用一下方式

echo "scan 'testByCrq', FILTER=>"ValueFilter(=,'substring:111')"" | hbase shell

几个例子

scan 'obd:journey', LIMIT => 10, FILTER=>"(ValueFilter(>,'binary:x00x00x00x00x00x00x00x0F') )"  

scan 'obd:journey', LIMIT => 10, FILTER=>"(ValueFilter(>,'binary:x00x00x02x81') AND ValueFilter(<,'binary:x00x00x02x82'))"

scan 'obd:journey', LIMIT => 10, FILTER=>"(RowFilter(>,'binary:x00x00x02x81') AND RowFilter(<,'binary:x00x00x02x82'))"

echo "scan 'obd:journey', FILTER=>"(RowFilter(>,'binary:x00x00x02x81'))"" | ./hbase shell > ~/journey.txt


echo "scan 'obd:journey', FILTER=>"(RowFilter(=,'binaryprefix:x00x00x02x81'))"" | ./hbase shell > ~/journey.txt


echo "scan 'obd:journey', FILTER=>"(RowFilter(>=,'binaryprefix:x00x00x02x81x00x00x00x00x00x00x00x00'))"" | ./hbase shell > ~/journey.txt

echo "scan 'obd:journey', FILTER=>"(RowFilter(>=,'binaryprefix:x00x00x00x81x00x00x00x00x00x00x00x00'))"" | ./hbase shell > ~/journey.txt

scan 'obd:journey', FILTER=>"(RowFilter(>,'binary:x00x00x02x81x00x00x00x00x00x00x00x00'))"


echo "scan 'obd:journey', FILTER=>"(RowFilter(>,'binaryprefix:x00x00x02x81x00x00x00x00x00x00x00x00') AND RowFilter(<,'binaryprefix:x00x00x02x82x00x00x00x00x00x00x00x00'))"" | ./hbase shell > ~/journey.txt