Re: [GENERAL] pg_dump -s dumps data?!
От | Tom Lane |
---|---|
Тема | Re: [GENERAL] pg_dump -s dumps data?! |
Дата | |
Msg-id | 22582.1328903667@sss.pgh.pa.us обсуждение исходный текст |
Ответ на | Re: [GENERAL] pg_dump -s dumps data?! (Tom Lane <tgl@sss.pgh.pa.us>) |
Список | pgsql-hackers |
I wrote: > Now, back to the original subject of this thread: both HEAD and 9.1 are > now operating as designed, in that they will dump the (user-provided > portion of the) contents of an extension config table whenever that > extension is dumped, even if --schema is specified. Or so I thought, anyway. Further experimentation with despez's example shows that in HEAD, --schema is still able to block dumping of extension config table contents, and the reason appears to be commit a4cd6abcc901c1a8009c62a27f78696717bb8fe1, which added yet another set of filtering conditions in a poorly chosen place; or possibly I should say it made arbitrary changes in the definition of the --schema switch. That patch needs some rethinking too, though I'm not sure what yet. I also note that his example shows that if you have a selective dump (say, with a -t switch), config table contents will be dumped even when the owning extension is not. This seems like a pretty clear bug: getExtensionMembership should not be creating TableDataInfo objects for extension config tables if the owning extension is not to be dumped. Barring objections, I'll go fix and back-patch that. regards, tom lane
В списке pgsql-hackers по дате отправления: