Re: gsoc, text search selectivity and dllist enhancments

Tom Lane wrote:
> Alvaro Herrera <alvherre@[EMAIL PROTECTED]
> writes:
> > Should we have a CHECK_FOR_INTERRUPTS() call in the inner loop of
> > compute_tsvector_stats?
> 
> Isn't the vacuum_delay_point() good enough?

But that's in the outer loop ... I mean here:

Index: src/backend/tsearch/ts_typanalyze.c
===================================================================
RCS file:
/home/alvherre/Code/cvs/pgsql/src/backend/tsearch/ts_typanalyze.c,v
retrieving revision 1.1
diff -c -p -r1.1 ts_typanalyze.c
*** src/backend/tsearch/ts_typanalyze.c	14 Jul 2008 00:51:45 -0000	1.1
--- src/backend/tsearch/ts_typanalyze.c	14 Jul 2008 04:59:59 -0000
*************** compute_tsvector_stats(VacAttrStats *sta
*** 206,211 ****
--- 206,213 ----
  		{
  			bool			found;
  
+ 			CHECK_FOR_INTERRUPTS();
+ 
  			/* Construct a hash key */
  			hash_key.lexeme = lexemesptr + curentryptr->pos;
  			hash_key.length = curentryptr->len;

-- 
Alvaro Herrera                               
http://www.CommandPrompt.com/
PostgreSQL Replication, Consulting, Custom Development, 24x7 sup****t

-- 
Sent via pgsql-hackers mailing list (pgsql-hackers@[EMAIL PROTECTED]
)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers