table-scan-question: more info

Hi,

I played a bit with the problem I described in my previous mail. And I
found that I've simplified the call-listing too much. Here is a better
listing:

ha_myc::info
ha_myc::cond_push
ha_myc::rnd_init
ha_myc::extra_opt
ha_myc::cond_push
ha_myc::rnd_next
........
ha_myc::rnd_end

This is the order of the calls for the select-stmt.

As you can see cond_push() is called twice. 

The fist call happens from the file sql_select.cc, the second one comes
from records.cc.

In the case of the delete- and update-stmts the calls only come from the
file records.cc.

In records.cc this happens in the function init_read_record() in the
last if-block.

I do not know:
- if I can move this block up in init_read_record()

- if this is the correct order: 
	- cond_push()
	- rnd_init()
	- rnd_next()
	- rnd_end()

- why cond_push() is called twice in the case of the select-stmt

Any tips?
Thanks
August



-- 
MySQL Internals Mailing List
For list archives: http://lists.mysql.com/internals
To unsubscribe:    http://lists.mysql.com/internals