http://www.sigaev.ru/misc/indexnulls-0.8.gz
Initially patch was developed by Martijn van Oosterhout. But it's reworked and
support of searching NULLS to GiST too. Patch adds new column named amsearchnull
to pg_am. To recognize IS NULL clause ScanKey->sk_flags contains (SK_ISNULL &
SK_INDEXFINDNULL) and ScanKey->sk_strategy sets to InvalidStrategy. IS NOT NULL
isn't supported.
The patch was already suggested to community
(http://archives.postgresql.org/pgsql-patches/2006-12/msg00019.php
and http://archives.postgresql.org/pgsql-hackers/2007-02/msg01162.php), but the
single objection was:
http://archives.postgresql.org/pgsql-patches/2006-12/msg00028.php
Objection was about representing IS [NOT] NULL clause in SkanKey structure,
because it required to set strategy to BTEqualStrategyNumber, and Tom doubted
about needing of such support.
Now ScanKey->sk_strategy = InvalidStrategy and btree code transforms it to
BTEqualStrategyNumber in _bt_preprocess_keys(), btcostestimate knows that fact too.
GiST doesn't need to transform strategy - it looks at SK_INDEXFINDNULL only.
--
Teodor Sigaev E-mail: teodor@sigaev.ru
WWW: http://www.sigaev.ru/