Le Samedi 23 Février 2002 21:53, Dave Page a écrit :
> This would be very expensive, even on moderately sized databases wouldn't
> it? Why not just check the ctid of the relevant tuple to see if it has
> changed since it was last read.
What if you create a trigger on a function which does not exi
> -Original Message-
> From: Jean-Michel POURE [mailto:[EMAIL PROTECTED]]
> Sent: 23 February 2002 18:30
> To: Dave Page
> Cc: [EMAIL PROTECTED]
> Subject: Re: [pgadmin-hackers] ALTER VIEW
>
>
>
> > 4) A class library such as this should _never_ display any user
> > interface. That
> -Original Message-
> From: Jean-Michel POURE [mailto:[EMAIL PROTECTED]]
> Sent: 23 February 2002 19:59
> To: [EMAIL PROTECTED]
> Subject: [pgadmin-hackers] Schema unique stamp (OID, MD5?)
>
>
> Dear friends,
>
> Do you see an existing way to stamp an existing schema with a
> uniqu
Dear friends,
Do you see an existing way to stamp an existing schema with a unique value
(OID?). This would be very usefull for pgAdmin2 when working in development
teams to decide whether a schema should be reloaded or not.
Using PLbash, this would look like :
"md5sum < pg_dump --schema-only
Hi Dave,
I commited some changes to CVS. Now, I need more info to understand :
A. VIEWS
> 1) Don't create the view twice, create the fake one then rename it - it's
> less expensive and avoids the potential for the second create to fail.
Done.
> 2) The szOldDefintion stuff can now be removed.
Do
