>>I intend to make new archives created with 7.5 pg_dump have the fix, and
>> restoring pre 7.5 binary dumps will have exactly the previous
>>behaviour. The reason for this is that extracting the acls and owners
>>to the end requires scanning the entire archive twice - not necessarily
>>something we want to do (is it?)
> I might be wrong about this, but I had the idea that the entire archive
> TOC gets loaded into memory, and so there'd be no noticeable penalty in
> scanning it twice.
Well, it probably gets scanned twice to do -c mode?
But...it seems kind of hacky to scan it again for owners and privs - are
you sure you want me to go that way?
In response to
pgsql-hackers by date
|Next:||From: Tom Lane||Date: 2004-06-25 03:30:55|
|Subject: Re: Fixing pg_dump |
|Previous:||From: Greg Sabino Mullane||Date: 2004-06-25 02:00:12|
|Subject: Re: [Re] Re: PREPARE and transactions |