Thread: misread release notes

misread release notes

From
ohp@pyrenet.fr
Date:
Hi all,

Methinks I made a big mistake when swapping from 8.2.1 to 8.2.2

I read the "fix incorrect permission check in
information_schema_key_column_usage_view" chapter in HISTORY far too fast
and fed psql on each database with share/information_schema.sql.
Too late to stop it!
What did I do wrong, and how can I go backwards (I imagine %I broke
something!)

Unrelated, I have problem with conforming string: adding a E is easy in C
or PHP but SpamAssassin has this plpgsql function  :

CREATE OR REPLACE FUNCTION put_tokens(inuserid INTEGER,                                     intokenary BYTEA[],
                           inspam_count INTEGER,                                     inham_count INTEGER,
                     inatime INTEGER)
 
RETURNS VOID AS '
DECLARE _token BYTEA; new_tokens INTEGER := 0;
BEGIN for i in array_lower(intokenary, 1) .. array_upper(intokenary, 1) LOOP   _token := intokenary[i];   UPDATE
bayes_token     SET spam_count = greatest_int(spam_count + inspam_count, 0),          ham_count =
greatest_int(ham_count+ inham_count, 0),          atime = greatest_int(atime, inatime)    WHERE id = inuserid      AND
token= _token;   IF NOT FOUND THEN     -- we do not insert negative counts, just return true     IF NOT (inspam_count <
0OR inham_count < 0) THEN       INSERT INTO bayes_token (id, token, spam_count, ham_count, atime)       VALUES
(inuserid,_token, inspam_count, inham_count, inatime);       IF FOUND THEN         new_tokens := new_tokens + 1;
ENDIF;     END IF;   END IF; END LOOP;
 
 IF new_tokens > 0 AND inatime > 0 THEN   UPDATE bayes_vars      SET token_count = token_count + new_tokens,
newest_token_age= greatest_int(newest_token_age, inatime),          oldest_token_age = least_int(oldest_token_age,
inatime)   WHERE id = inuserid; ELSEIF new_tokens > 0 AND NOT inatime > 0 THEN   UPDATE bayes_vars      SET token_count
=token_count + new_tokens    WHERE id = inuserid; ELSEIF NOT new_tokens > 0 AND inatime > 0 THEN   UPDATE bayes_vars
 SET newest_token_age = greatest_int(newest_token_age, inatime),          oldest_token_age =
least_int(oldest_token_age,inatime)    WHERE id = inuserid; END IF; RETURN;
 
END;
' LANGUAGE 'plpgsql';

As you see, token is bytea where do I put the E in the insert query?

Keep the good job!
Regards
-- 
Olivier PRENANT                    Tel: +33-5-61-50-97-00 (Work)
15, Chemin des Monges                +33-5-61-50-97-01 (Fax)
31190 AUTERIVE                       +33-6-07-63-80-64 (GSM)
FRANCE                          Email: ohp@pyrenet.fr
------------------------------------------------------------------------------
Make your life a dream, make your dream a reality. (St Exupery)


Re: misread release notes

From
Tom Lane
Date:
ohp@pyrenet.fr writes:
> I read the "fix incorrect permission check in
> information_schema_key_column_usage_view" chapter in HISTORY far too fast
> and fed psql on each database with share/information_schema.sql.
> Too late to stop it!
> What did I do wrong, and how can I go backwards (I imagine %I broke
> something!)

I don't think you changed anything --- none of the commands say OR REPLACE
so they'd just all fail.
        regards, tom lane


Re: misread release notes

From
Andrew Dunstan
Date:
ohp@pyrenet.fr wrote:
> Unrelated, I have problem with conforming string: adding a E is easy in C
> or PHP but SpamAssassin has this plpgsql function  :
>
>
>   
[snip]
> As you see, token is bytea where do I put the E in the insert query?
>
>   

Since you aren't using a string literal in this function for token or 
anything else, AFAICS, I don't understand why you would need to.

cheers

andrew


Re: misread release notes

From
ohp@pyrenet.fr
Date:
Hi Tom,
On Tue, 6 Feb 2007, Tom Lane wrote:

> Date: Tue, 06 Feb 2007 11:50:56 -0500
> From: Tom Lane <tgl@sss.pgh.pa.us>
> To: ohp@pyrenet.fr
> Cc: pgsql-hackers list <pgsql-hackers@postgresql.org>
> Subject: Re: [HACKERS] misread release notes
>
> ohp@pyrenet.fr writes:
> > I read the "fix incorrect permission check in
> > information_schema_key_column_usage_view" chapter in HISTORY far too fast
> > and fed psql on each database with share/information_schema.sql.
> > Too late to stop it!
> > What did I do wrong, and how can I go backwards (I imagine %I broke
> > something!)
>
> I don't think you changed anything --- none of the commands say OR REPLACE
> so they'd just all fail.
Yes but I saw a lot of inserts...

>
>             regards, tom lane
>

-- 
Olivier PRENANT                    Tel: +33-5-61-50-97-00 (Work)
15, Chemin des Monges                +33-5-61-50-97-01 (Fax)
31190 AUTERIVE                       +33-6-07-63-80-64 (GSM)
FRANCE                          Email: ohp@pyrenet.fr
------------------------------------------------------------------------------
Make your life a dream, make your dream a reality. (St Exupery)