Re: preserving forensic information when we freeze

Hi,

On 2014-01-02 05:26:26 +0000, Greg Stark wrote:
> 2) refetching a row could conceivably end up retrieving different data than
> was present when the row was originally read. (In some cases that might
> actually be the intended behaviour)

That's possible with system columns as well. In the normal cases we'll
only have copied the HeapTuple, not the HeapTupleHeader, so it will be
re-fetched from the (pinned) buffer.

> If this came up earlier I'm sorry but I suppose it's too hard to have a
> function like foo(tab.*) which magically can tell that the record is a heap
> tuple and look in the header? And presumably throw an error if passed a non
> heap tuple.

I don't see how that could be a good API. What happens if you have two
relations in a query?
Even if that wouldn't be a query, why would this be a helpful? Seems
like a poor reinvention of system columns.

Andres

PS: Could you not always include the full quoted message below your --
signature?

Greetings,

Andres Freund

Greetings,

Andres Freund

--
Andres Freund http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services