On Thu, 3 Feb 2005, Karl Denninger wrote:
> On Thu, Feb 03, 2005 at 10:20:47PM -0500, Tom Lane wrote:
>> Karl Denninger <karl@denninger.net> writes:
>>> I agree with this - what would be even better would be a way to create
>>> 'subclasses' for things like this, which could then be 'included' easily.
>>
>> We could decree that a contrib module's script should create a schema
>> and shove everything it makes into that schema. Then "DROP SCHEMA CASCADE"
>> is all you need to get rid of it. However, you'd probably end up having
>> to add this schema to your search path to use the module conveniently.
>>
>> regards, tom lane
>
> I would prefer that vastly over what I had to deal with this time. Until I
> discovered the "untsearch2.sql" script I was seriously considering the
> trouble of backing this out <BY HAND> either in a dump or in the online
> database.
As I already pointed you should apply regprocedure_update.sql to your
7.4.1 database before dumping to get OIDs free tsearch2 functions !
This is the way most people dump their db without any problem. Unfortunately,
this was not documented in 7.4 readme, but it does in current documentations,
web site and discussed in mailing list.
>
> Either would have been a stupendous amount of work.
>
> Please consider this change in approach - someone else has to have been bit
> in the butt by this one other than me by now.
>
> --
>
Regards,
Oleg
_____________________________________________________________
Oleg Bartunov, sci.researcher, hostmaster of AstroNet,
Sternberg Astronomical Institute, Moscow University (Russia)
Internet: oleg@sai.msu.su, http://www.sai.msu.su/~megera/
phone: +007(095)939-16-83, +007(095)939-23-83