Home > mailing lists

Thread: Kind of error-handler in a pgsql function

Kind of error-handler in a pgsql function

From

DI Hasenöhrl

Date:

04 July 2001, 18:37:02

Hi,

When I write in psql:

testdb=# update table1 set tableattribute='any' where table_nr=1;

if a tuple exists, I get this message

testdb=# update 1

if no tuple with table_nr=1 exists, I get this message

testdb=# update 0

Is there a possibility to make a difference in a pgsql function like this:

create function updTable(text,integer) returns int AS

'DECLARE

msg ALIAS FOR $1;

nr ALIAS FOR $2;

BEGIN

update table1 set tableattribute=msg where table_nr=nr;

--pseudocode

if update = 0 then

return 0;

else

return 1;

end if;

END;

'language 'plpgsql';

or for a function, which inserts data:

create function insTable(text,integer) returns int AS

'DECLARE

msg ALIAS FOR $1;

nr ALIAS FOR $2;

BEGIN

insert into table1 values (nr,msg);

--pseudocode

if error= cannot insert duplicate key.....then

return 0;

else

return 1;

end if;

END;

'language 'plpgsql';

I want to know the result of an insert or update, because I call these functions from an Access form and the next steps of the program depend on these results.

I hope, someone can help me, because I didn't find anything in the docu or mailing list.

Thanks in advance

Irina

E-Mail: i.hasenoehrl@aon.at

Re: Kind of error-handler in a pgsql function

From

Tom Lane

Date:

04 July 2001, 23:38:55

DI Hasenöhrl <i.hasenoehrl@aon.at> writes:
> I want to know the result of an insert or update, because I call these func=
> tions from an Access form and the next steps of the program depend on these=
>  results.

See the FOUND variable (if you just want to test zero or not-zero rows
processed) or the SET DIAGNOSTICS statement to get the row count.

There is no way to trap an error inside a plpgsql function, however.
        regards, tom lane

Re: Kind of error-handler in a pgsql function

From

Alex Pilosov

Date:

05 July 2001, 02:01:28

You must do it the other way around: 

First, try update, then see how many records were updated, if there were 0
records updated, then do the insert.

Currently, plpgsql lacks decent exception handling. 

Sample code:

create function ...
...
declare rec_affected int;
begin update ...  get diagnostics rec_affected = ROW_COUNT; if rec_affected = 0 then     insert ... end if;
end;

On Wed, 4 Jul 2001, [iso-8859-1] DI Hasen�hrl wrote:

> Hi,
> 
> When I write in psql:
> testdb=# update table1 set tableattribute='any' where table_nr=1;
>     if a tuple exists, I get this message 
> testdb=# update 1
>     if no tuple with table_nr=1 exists, I get this message
> testdb=# update 0
>     
> 
> Is there a possibility to make a difference in a pgsql function like this:
> create function updTable(text,integer) returns int AS
> 'DECLARE
>    msg ALIAS FOR $1;
>    nr    ALIAS FOR $2;
>  BEGIN
>      update table1 set tableattribute=msg where table_nr=nr;
>      --pseudocode
>        if update = 0 then
>          return 0;
>        else
>          return 1;
>        end if;
>    END;
> 'language 'plpgsql';
> 
> or for a function, which inserts data:
> create function insTable(text,integer) returns int AS
> 'DECLARE
>    msg ALIAS FOR $1;
>    nr    ALIAS FOR $2;
>  BEGIN
>      insert into table1 values (nr,msg);
>      --pseudocode
>        if error= cannot insert duplicate key.....then
>          return 0;
>        else
>          return 1;
>        end if;
>    END;
> 'language 'plpgsql';
> 
> I want to know the result of an insert or update, because I call these functions from an Access form and the next
stepsof the program depend on these results.
 
> 
> I hope, someone can help me, because I didn't find anything in the docu or mailing list.
> Thanks in advance
> Irina
> 
> E-Mail: i.hasenoehrl@aon.at
>

Re: Kind of error-handler in a pgsql function

From

DI Hasenöhrl

Date:

05 July 2001, 07:02:12

Thank you for your example.

I understand, in order to avoid inserting *duplicate keys*, first I try to update, when it fails make an insert. But other errors may occur, is there a possibility to distinguish between errors by an error-number.

Please can you tell me, where I can find system functions, variables like *diagnostics, row_count, found, and so on.....*

I studied the docu in http://www.ca.postgresql.org/devel-corner/docs/postgres/programmer.html, but I didn't find

Many thanks in advance

Irina

You must do it the other way around:

First, try update, then see how many records were updated, if there were 0
records updated, then do the insert.

Currently, plpgsql lacks decent exception handling.

Sample code:

create function ...
...
declare rec_affected int;
begin
update ...
get diagnostics rec_affected = ROW_COUNT;
if rec_affected = 0 then
    insert ...
end if;
end;

On Wed, 4 Jul 2001, [iso-8859-1] DI Hasenöhrl wrote:

> Hi,
>
> When I write in psql:
> testdb=# update table1 set tableattribute='any' where table_nr=1;
>     if a tuple exists, I get this message
> testdb=# update 1
>     if no tuple with table_nr=1 exists, I get this message
> testdb=# update 0
>
>
> Is there a possibility to make a difference in a pgsql function like this:
> create function updTable(text,integer) returns int AS
> 'DECLARE
>    msg ALIAS FOR $1;
>    nr    ALIAS FOR $2;
> BEGIN
>      update table1 set tableattribute=msg where table_nr=nr;
>      --pseudocode
>        if update = 0 then
>          return 0;
>        else
>          return 1;
>        end if;
>    END;
> 'language 'plpgsql';
>
> or for a function, which inserts data:
> create function insTable(text,integer) returns int AS
> 'DECLARE
>    msg ALIAS FOR $1;
>    nr    ALIAS FOR $2;
> BEGIN
>      insert into table1 values (nr,msg);
>      --pseudocode
>        if error= cannot insert duplicate key.....then
>          return 0;
>        else
>          return 1;
>        end if;
>    END;
> 'language 'plpgsql';
>
> I want to know the result of an insert or update, because I call these functions from an Access form and the next steps of the program depend on these results.
>
> I hope, someone can help me, because I didn't find anything in the docu or mailing list.
> Thanks in advance
> Irina
>
> E-Mail: i.hasenoehrl@aon.at
>

---------------------------(end of broadcast)---------------------------
TIP 1: subscribe and unsubscribe commands go to majordomo@postgresql.org

Re: Kind of error-handler in a pgsql function

From

Alex Pilosov

Date:

05 July 2001, 13:18:57

There is _no error handling_ in plpgsql

You can find documentation at:
http://www.postgresql.org/idocs/index.php?plpgsql.html

On Thu, 5 Jul 2001, [iso-8859-1] DI Hasen�hrl wrote:

> Thank you for your example. I understand, in order to avoid inserting
> *duplicate keys*, first I try to update, when it fails make an insert.
> But other errors may occur, is there a possibility to distinguish
> between errors by an error-number. Please can you tell me, where I can
> find system functions, variables like *diagnostics, row_count, found,
> and so on.....* I studied the docu in
> http://www.ca.postgresql.org/devel-corner/docs/postgres/programmer.html,
> but I didn't find
> 
> Many thanks in advance
> Irina
> 
>   You must do it the other way around: 
> 
>   First, try update, then see how many records were updated, if there were 0
>   records updated, then do the insert.
> 
>   Currently, plpgsql lacks decent exception handling. 
> 
>   Sample code:
> 
>   create function ...
>   ...
>   declare rec_affected int;
>   begin
>     update ... 
>     get diagnostics rec_affected = ROW_COUNT;
>     if rec_affected = 0 then  
>       insert ...
>     end if;
>   end;
> 
>   On Wed, 4 Jul 2001, [iso-8859-1] DI Hasen�hrl wrote:
> 
>   > Hi,
>   > 
>   > When I write in psql:
>   > testdb=# update table1 set tableattribute='any' where table_nr=1;
>   >     if a tuple exists, I get this message 
>   > testdb=# update 1
>   >     if no tuple with table_nr=1 exists, I get this message
>   > testdb=# update 0
>   >     
>   > 
>   > Is there a possibility to make a difference in a pgsql function like this:
>   > create function updTable(text,integer) returns int AS
>   > 'DECLARE
>   >    msg ALIAS FOR $1;
>   >    nr    ALIAS FOR $2;
>   >  BEGIN
>   >      update table1 set tableattribute=msg where table_nr=nr;
>   >      --pseudocode
>   >        if update = 0 then
>   >          return 0;
>   >        else
>   >          return 1;
>   >        end if;
>   >    END;
>   > 'language 'plpgsql';
>   > 
>   > or for a function, which inserts data:
>   > create function insTable(text,integer) returns int AS
>   > 'DECLARE
>   >    msg ALIAS FOR $1;
>   >    nr    ALIAS FOR $2;
>   >  BEGIN
>   >      insert into table1 values (nr,msg);
>   >      --pseudocode
>   >        if error= cannot insert duplicate key.....then
>   >          return 0;
>   >        else
>   >          return 1;
>   >        end if;
>   >    END;
>   > 'language 'plpgsql';
>   > 
>   > I want to know the result of an insert or update, because I call these functions from an Access form and the next
stepsof the program depend on these results.
 
>   > 
>   > I hope, someone can help me, because I didn't find anything in the docu or mailing list.
>   > Thanks in advance
>   > Irina
>   > 
>   > E-Mail: i.hasenoehrl@aon.at
>   > 
> 
> 
>   ---------------------------(end of broadcast)---------------------------
>   TIP 1: subscribe and unsubscribe commands go to majordomo@postgresql.org
> 
>