Re: ADD FOREIGN KEY (was Re: [GENERAL] 7.4Beta)
От | Jan Wieck |
---|---|
Тема | Re: ADD FOREIGN KEY (was Re: [GENERAL] 7.4Beta) |
Дата | |
Msg-id | 3F84CADA.6050905@Yahoo.com обсуждение исходный текст |
Ответ на | Re: ADD FOREIGN KEY (was Re: [GENERAL] 7.4Beta) (Bruce Momjian <pgman@candle.pha.pa.us>) |
Список | pgsql-hackers |
Bruce Momjian wrote: > Christopher Kings-Lynne wrote: >> >> >>Well, with the CREATE CONSTRAINT TRIGGER you _can_, but we already have >> >>a consensus that we don't _want_ that. Probably we should declare it >> >>deprecated and remove it in 7.5. And the option currently under >> >>discussion is exactly what will cause ALTER TABLE to let you, but IMHO >> >>that _should_ be restricted. >> >> How can we ever remove it - what about people upgrading from 7.0, 7.1, >> 7.2? Also, people upgrading from 7.3 who've never heard of adddepend... > > Not sure. We can remove documentation about it, at least. > If the idea is to support any 7.n -> 7.m (where n < m) upgrade directly, then it's IMHO time for 8.0 and clearly stating that 7.x -> 8.y only is supported as 7.x -> 7.3 -> 8.0 -> 8.y and "you're on your own with any other attempt". Don't get this wrong, I am a big friend of easy upgrades. But I am not a big friend of making improvements impossible by "backward compatibility forever". It was the "backward compatibility" to CP/M-80 (v2.2) that caused MS-DOS 7.0 to have a maximum commandline length of 127 characters ... that was taking compatibility too far. Well, M$ took it too far the other way from there and is compatible to nothing any more, not even to themself ... but at least they learned from that mistake. Jan -- #======================================================================# # It's easier to get forgiveness for being wrong than for being right. # # Let's break this rule - forgive me. # #================================================== JanWieck@Yahoo.com #
В списке pgsql-hackers по дате отправления: