Re: Relax transactional restrictions on ALTER ENUM ... ADD TYPE (redux)
От | Thomas Munro |
---|---|
Тема | Re: Relax transactional restrictions on ALTER ENUM ... ADD TYPE (redux) |
Дата | |
Msg-id | CAEepm=0cQ4wdhzTyH6nefCg1wApcPtLu+bNrXFC6JY_zjZcnew@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Relax transactional restrictions on ALTER ENUM ... ADD TYPE(redux) (Andrew Dunstan <andrew.dunstan@2ndquadrant.com>) |
Ответы |
Re: Relax transactional restrictions on ALTER ENUM ... ADD TYPE(redux)
Re: Relax transactional restrictions on ALTER ENUM ... ADD TYPE (redux) |
Список | pgsql-hackers |
On Sat, Oct 6, 2018 at 2:31 AM Andrew Dunstan <andrew.dunstan@2ndquadrant.com> wrote: > >> On Wed, Oct 3, 2018 at 2:24 PM Thomas Munro > >> <thomas.munro@enterprisedb.com> wrote: > >>> Over the thread for bug #14825 I posted some draft code to show one > >>> way to save/restore the enum blacklist for parallel workers. Here's a > >>> better version, and a new thread. 0001 is the code by Andrew Dustan > >>> and Tom Lane that was reverted in 93a1af0b, unchanged by me except for > >>> resolving trivial conflicts on current master. 0002 fixes the failure > >>> seen with make installcheck when postgresql.conf says > >>> force_parallel_mode = regress. > > Many thanks for doing this. Your solution seems simpler and cleaner that > what was previously proposed. > > I have tested it, and confirm that without your 0002 patch there is an > error with force_parallel_mode=regress and with 0002 that error goes away. Thanks. Here is a version squashed into one commit, with a decent commit message and a small improvement: the code to create the hash table is moved into a static function, to avoid repetition. I will push this to master early next week, unless there is more feedback or one of you would prefer to do that. -- Thomas Munro http://www.enterprisedb.com
Вложения
В списке pgsql-hackers по дате отправления: