optimizing impossible matches - Mailing list pgsql-hackers

From Merlin Moncure
Subject optimizing impossible matches
Date
Msg-id 6EE64EF3AB31D5448D0007DD34EEB34101AD84@Herge.rcsinc.local
Whole thread Raw
Responses Re: optimizing impossible matches
List pgsql-hackers
Dear hackers:
Is it reasonable or possible to have the optimizer filter out impossible
string/numeric matches because of length/overflow conditions?

For example: (on 7.4.1)
select * from t where f = '1234567' -- f = char(6)
explain says index/seq scan, depending on presence of index, when
obviously there can be no matches.

also, for the typical case for indexed field f,select * from t where f = f' or false
generates an index scan

but select * from t where f = f' or f = f''
generates a seq. scan with default settings.  If f'' is not in the
domain of f, the first case should apply.  Is this really as simple as
it seems?

Merlin





pgsql-hackers by date:

Previous
From: Matthew Kirkwood
Date:
Subject: Re: Scalable postgresql using sys_epoll
Next
From: "Merlin Moncure"
Date:
Subject: Re: Timing of 'SELECT 1'