Added page_range to records_in_range() to improve range statistics

Prototype change:
-  virtual ha_rows records_in_range(uint inx, key_range *min_key,
-                                   key_range *max_key)
+  virtual ha_rows records_in_range(uint inx, const key_range *min_key,
+                                   const key_range *max_key,
+                                   page_range *res)

The handler can ignore the page_range parameter. In the case the handler
updates the parameter, the optimizer can deduce the following:
- If previous range's last key is on the same block as next range's first
  key
- If the current key range is in one block
- We can also assume that the first and last block read are cached!
  This can be used for a better calculation of IO seeks when we
  estimate the cost of a range index scan.

The parameter is fully implemented for MyISAM, Aria and InnoDB.
A separate patch will update handler::multi_range_read_info_const() to
take the benefits of this change and also remove the double
records_in_range() calls that are not anymore needed.

1 files changed, 4 insertions, 2 deletions

diff --git a/sql/ha_partition.h b/sql/ha_partition.h
index 85cb736b5bd..eb10cf84e76 100644
--- a/sql/ha_partition.h
+++ b/sql/ha_partition.h
@@ -1048,8 +1048,10 @@ public:
     For the given range how many records are estimated to be in this range.
     Used by optimiser to calculate cost of using a particular index.
   */
-  ha_rows records_in_range(uint inx, key_range * min_key, key_range * max_key)
-    override;
+  ha_rows records_in_range(uint inx,
+                           const key_range * min_key,
+                           const key_range * max_key,
+                           page_range *pages) override;
 
   /*
     Upper bound of number records returned in scan is sum of all