[Ruby-DBI-devel] Performance problems with dbd_pg

[James F. H. <jf...@ci...>]

I noticed in the archives that others had reported the performance problem
with dbd_pg. 

It turns out the problem lies in the repeated calling of PGResult#result
by fetchrow. PGResult#result creates the result array each time it's called.
The following patch takes care of the problem in DBD::Pg by caching the
result array, though it may be better to have the postgres driver cache 
the result array after the first call instead.

*** Pg.rb       Sun Dec  2 12:23:59 2001
--- /local/lib/ruby/site_ruby/1.7/DBD/Pg/Pg.rb  Tue Apr 16 23:07:17 2002
***************
*** 452,457 ****
--- 452,458 ----
          @db = db
          @pg_result = pg_result
          @index = -1
+         @result = @pg_result.result
          @row = Array.new
        end
  
***************
*** 461,468 ****
  
        def fetchrow
          @index += 1
!         if @index < @pg_result.result.size 
!           fill_array(@pg_result.result[@index])
          else
            @row = nil
          end
--- 462,469 ----
  
        def fetchrow
          @index += 1
!         if @index < @result.size 
!           fill_array(@result[@index])
          else
            @row = nil
          end

----------------------------------------------------------------------
| Jim Hranicky, Senior SysAdmin                   UF/CISE Department |
| E314D CSE Building                            Phone (352) 392-1499 |
| jf...@ci...                      http://www.cise.ufl.edu/~jfh |
----------------------------------------------------------------------