Thread: 8.1-compatible xlogdump

8.1-compatible xlogdump

From
"Jonah H. Harris"
Date:
Does anyone have a copy of xlogdump updated for 8.1?  If so, please send me a copy.  Otherwise I'll update it and forward the diffs.
 
-Jonah

Re: 8.1-compatible xlogdump

From
Alvaro Herrera
Date:
Jonah H. Harris wrote:
> Does anyone have a copy of xlogdump updated for 8.1? If so, please send me a
> copy. Otherwise I'll update it and forward the diffs.

We should definitely consider completing it and including it in contrib.

-- 
Alvaro Herrera       Valdivia, Chile   ICBM: S 39º 49' 17.7", W 73º 14' 26.8"
"El sentido de las cosas no viene de las cosas, sino de
las inteligencias que las aplican a sus problemas diarios
en busca del progreso." (Ernesto Hernández-Novich)


Re: 8.1-compatible xlogdump

From
"Jonah H. Harris"
Date:
I agree.

On 11/1/05, Alvaro Herrera <alvherre@alvh.no-ip.org> wrote:
Jonah H. Harris wrote:
> Does anyone have a copy of xlogdump updated for 8.1? If so, please send me a
> copy. Otherwise I'll update it and forward the diffs.

We should definitely consider completing it and including it in contrib.

--
Alvaro Herrera       Valdivia, Chile   ICBM: S 39º 49' 17.7", W 73º 14' 26.8"
"El sentido de las cosas no viene de las cosas, sino de
las inteligencias que las aplican a sus problemas diarios
en busca del progreso." (Ernesto Hernández-Novich)

Re: 8.1-compatible xlogdump

From
Tom Lane
Date:
Alvaro Herrera <alvherre@alvh.no-ip.org> writes:
> Jonah H. Harris wrote:
>> Does anyone have a copy of xlogdump updated for 8.1? If so, please send me a
>> copy. Otherwise I'll update it and forward the diffs.

> We should definitely consider completing it and including it in contrib.

There is already code in the main backend to convert all the xlog record
types into text.  We should think of some way to reuse that code, rather
than maintaining identical functionality elsewhere.

Mind you, the backend code pretty much sucks from a robustness point of
view (unchecked string buffer use) so rewriting it along the way isn't
a bad idea.  I'm just saying let's think about how to have one copy not
two.
        regards, tom lane