Re: beta5 ...
От | Tom Lane |
---|---|
Тема | Re: beta5 ... |
Дата | |
Msg-id | 12614.982684257@sss.pgh.pa.us обсуждение исходный текст |
Ответ на | Re: beta5 ... (Thomas Lockhart <lockhart@alumni.caltech.edu>) |
Список | pgsql-hackers |
>> As a matter of curiosity, is each beta compiled and then regression >> tested against *every* one of the known "supported" platforms before >> release? Who are you expecting to do that, exactly? One of the differences between Postgres and a proprietary commercial database is that there is no vast support machinery behind the scenes. What you see going on on this list is what you get: beta testing consists of the activities performed and reported by list members. Normally, if we are about to push out a beta then two or three people will double-check that the current CVS tip builds and passes regression on their personal machines. But the "supported platforms" coverage depicted in the docs consists of all the platforms that are reported to us as working during the entire beta test period, including many that the key developers have no direct access to. There's no way that we could reverse the process and cause that to happen before a beta release instead of after; certainly no way that we could cause all that effort to be repeated for each beta version. If you are using a beta version then you are part of that testing process, not a beneficiary of something that's happened behind closed doors. regards, tom lane
В списке pgsql-hackers по дате отправления: