On 01/30/2012 09:45 AM, hubert depesz lubaczewski wrote:
> On Mon, Jan 30, 2012 at 09:43:46AM -0800, Adrian Klaver wrote:
>> On 01/30/2012 09:23 AM, hubert depesz lubaczewski wrote:
>>
>>>
>>> I think I explained it in previous mails, and if not - sorry, but
>>> I clearly can't explain good enough - the point is that with the way how
>>> extensions now work, they are useless for providing way to create
>>> tables that will store data, in case you would ever want dump without
>>> this data.
>>
>> So in summary; if an extension creates a user table you want access
>> to that table(schema and data) via pg_dump, outside the extension
>> mechanism, without resorting to marking it as a configuration table.
>> Is that correct ?
>
> no.
> I want to be able to do both:
> 1. dump the data for the table
> 2. dump structure of other tables
> but not in the same file.
Actually that was what I was saying:)
"..via pg_dump, outside the extension mechanism.."
"..without resorting to marking it as a configuration table.."
Currently the extension mechanism is getting in the way of 1 & 2 above.
What you want is for pg_dump to ignore the extension dependency process
when you explicitly name a table and the operation on it.
>
> Best regards,
>
> depesz
>
--
Adrian Klaver
adrian.klaver@gmail.com