Re: Further thoughts about warning for costly FK checks
От | Fabien COELHO |
---|---|
Тема | Re: Further thoughts about warning for costly FK checks |
Дата | |
Msg-id | Pine.LNX.4.58.0403182353070.2899@sablons.cri.ensmp.fr обсуждение исходный текст |
Ответ на | Re: Further thoughts about warning for costly FK checks (Tom Lane <tgl@sss.pgh.pa.us>) |
Список | pgsql-hackers |
Dear Tom, I thought about it... how to solve the contradiction:- backend vs external tool?- new interface? new command? unix? windows?-compatibility with old/existing interfaces?- plugins, any one can contribute?- don't bother DBA's- communicate aboutit? My 2 pence idea of the day (for free): CREATE SCHEMA pg_advisor; CREATE VIEW pg_advisor.table_without_primary_keys AS SELECT ... FROM pg_catalog...; COMMENT ON VIEW pg_advisor.table_without_primary_keys IS 'hey man, it is considered better to have a primary key on tables...'; CREATE VIEW pg_advisor.costly_ri_checks AS ... CREATE VIEW pg_advisor.summary AS SELECT 'missing primary key declarations', COUNT(*) FROM pg_advisor.table_without_primary_keys UNION SELECT 'costly referencial integrity checks...', COUNT(*) FROM pg_advisor.costly_ri_checks UNION ... ; Then:- it is in the backend, somehow;-)- easy plugin: CREATE VIEW ...;;-)- anyone fluent in SQL and in pg_catalog can contribute!-ALL existing interfaces are already compatible;-) I can use psql;-);-)- no one has to look at the views if hedoes not want to.- you can communicate about it in the new release...- well, we're in a relationnal database, so let usstay there;-) Good night, -- Fabien Coelho - coelho@cri.ensmp.fr
В списке pgsql-hackers по дате отправления: