Like the title, what is the HBase RegexStringComparator's efficiency in RowFilter if under this three circumstance: 1: I need to match the beginning of the row , such as "abc*", "abc\d" and so on, I thought it has a good efficiency because it doesn't need to scan the whole table.
2: match at the middle position of the row, such as "\d{3,4}abc\w+" and so on, I think it need to scan all the rows and has a bad efficiency .
3: match at the end of the row, and like the second circumstance, also has a bad efficiency.
Do I get the right understand?
Only use STARTROW and ENDROW for range scan can speed up query efficiency, any FILTER runs after SCAN, so all the 3 conditions you have described have the same efficiency.