Re: ALTER EXTENSION UPGRADE, v3 - Mailing list pgsql-hackers

From Tom Lane
Subject Re: ALTER EXTENSION UPGRADE, v3
Date
Msg-id 4835.1297372929@sss.pgh.pa.us
Whole thread Raw
In response to Re: ALTER EXTENSION UPGRADE, v3  (Robert Haas <robertmhaas@gmail.com>)
Responses Re: ALTER EXTENSION UPGRADE, v3  ("David E. Wheeler" <david@kineticode.com>)
List pgsql-hackers
Robert Haas <robertmhaas@gmail.com> writes:
> On Thu, Feb 10, 2011 at 3:58 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> Hmm. �The problem with that is that once there are large numbers of
>> intermediate versions, the number of potential paths grows
>> exponentially.

> It's certainly not exponential i.e. O(2^n) or something of that form.
> Even a naive application of Dijkstra's algorithm is only going to be
> O(n^2) in the number of versions, which is likely tolerable even if
> upgrades are supported for dozens of old versions.

Well, okay, let's go with that plan then.  If we don't need to assume
anything more than equality of version names being meaningful, I think
chaining update scripts automatically should solve most of the
complaints here.  People who really want to maintain shortcut scripts
still could, but I think it'd be an unusual case.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: ALTER EXTENSION UPGRADE, v3
Next
From: Robert Haas
Date:
Subject: Re: ALTER EXTENSION UPGRADE, v3