Re: SQL-Invoked Procedures for 8.1 - Mailing list pgsql-hackers

From Peter Mount
Subject Re: SQL-Invoked Procedures for 8.1
Date
Msg-id 4152C57F.20604@retep.org.uk
Whole thread Raw
In response to Re: SQL-Invoked Procedures for 8.1  (Hannu Krosing <hannu@tm.ee>)
List pgsql-hackers
Hannu Krosing wrote:

>On N, 2004-09-23 at 15:22, Peter Mount wrote:
>  
>
>>Hannu Krosing wrote:
>>
>>    
>>
>>>On N, 2004-09-23 at 11:34, Grant Finnemore wrote:
>>> 
>>>
>>>      
>>>
>>>>Hi Magnus,
>>>>
>>>>Yes, this is the situation that I have been thinking about. Specifically
>>>>when a single stored procedure returns many recordsets.
>>>>   
>>>>
>>>>        
>>>>
>>>IIRC support for returning multiple recordsets was removed from
>>>postgresql fe-be protocol years ago as "nobody ever needs it" ;)
>>> 
>>>
>>>      
>>>
>>Until recently I would have said the same thing.
>>
>>A couple of months ago I started a new job where they are MS-SQL based, 
>>and they do have several procedures that actually do return multiple 
>>recordsets from a single procedure. However this is the only time that 
>>I've ever seen any procedure actually require it.
>>    
>>
>
>Actually the original support was inhetited from Postgres4.2 where
>PostQuel had native support for it.
>
>It was probably decided that SQL (which replaced PostQuel) would not
>generate such things.
>  
>
I'm not so sure as both JDBC & ODBC have implicit support for it and 
they are younger than SQL.

ie: with Prepared/Callable statements in JDBC you are supposed to check 
for the existence of any other ResultSets when reading the results... 
this was what I had to do with CallableStatement last week with MSSQL.

Peter



pgsql-hackers by date:

Previous
From: Hannu Krosing
Date:
Subject: Re: SQL-Invoked Procedures for 8.1
Next
From: Andrew Dunstan
Date:
Subject: Re: Use of zlib