Re: [HACKERS] Allow GiST opcalsses without compress\decompres functions - Mailing list pgsql-hackers

From Alexander Korotkov
Subject Re: [HACKERS] Allow GiST opcalsses without compress\decompres functions
Date
Msg-id CAPpHfdu7YRDT7pVOp8t-cpjqeO7RtqRdsU40kywOn7v-BvaGFQ@mail.gmail.com
Whole thread Raw
In response to Re: [HACKERS] Allow GiST opcalsses without compress\decompres functions  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
On Wed, Sep 20, 2017 at 5:02 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
Alexander Korotkov <a.korotkov@postgrespro.ru> writes:
> On Wed, Sep 20, 2017 at 4:25 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> Yes.  We don't allow out-of-line values, but we do allow compressed and
>> short-header values.

> BTW, this comment looks still invalid for me...

>> #define SIGLENINT  4 /* >122 => key will *toast*, so very slow!!! */

I haven't done the math to see if 122 is exactly the right value,
but at some point index_form_tuple would decide that the tuple was
uncomfortably big and try to compress it.  So the comment is right
in general terms.

If comment means toast as compression, not out-of-line storage then it's true.
However, it would be good to rephrase this comment to clarify that.

------
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company 

pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: [HACKERS] SCRAM in the PG 10 release notes
Next
From: Robert Haas
Date:
Subject: Re: [HACKERS] Re: [COMMITTERS] pgsql: Expand partitioned table RTEslevel by level, without flattening