Re: Reducing overhead for repeat de-TOASTing

* Tom Lane (tgl(at)sss(dot)pgh(dot)pa(dot)us) wrote:
> One unsolved problem is that this scheme doesn't provide any way to cache
> the result of decompressing an inline-compressed datum, because those have
> no unique ID that could be used for a lookup key.

That's pretty unfortunate.

> Ideas?

Not at the moment, but given the situation it really does strike me as
something we want to solve. Inventing an ID would likely be overkill
or wouldn't solve the problem anyway, I'm guessing...

> Comments, better ideas? Anyone think this is too much trouble to take
> for the problem?

I definitely think it's worth it, even if it doesn't handle an
inline-compressed datum. PostGIS is certainly a good use case for why,
but I doubt it's the only one.

Thanks!

Stephen