[HACKERS] Regression test plpgsql vs. rangefuncs conflict

2005-07-01 Thread Peter Eisentraut
The regression test files plpgsql and rangefuncs both create a dup() 
function, and as they are run in parallel this just caused an error for 
me, as attached.  This just happened once for me, but it still ought to 
be corrected.

-- 
Peter Eisentraut
http://developer.postgresql.org/~petere/
*** ./expected/plpgsql.out	Thu Jun 30 18:32:22 2005
--- ./results/plpgsql.out	Fri Jul  1 19:39:51 2005
***
*** 1855,1860 
--- 1855,1861 
  (1 row)
  
  drop function dup(anyelement);
+ ERROR:  function dup(anyelement) does not exist
  --
  -- test PERFORM
  --

==

*** ./expected/rangefuncs.out	Mon May 30 17:35:15 2005
--- ./results/rangefuncs.out	Fri Jul  1 19:39:48 2005
***
*** 493,498 
--- 493,499 
  --
  CREATE FUNCTION dup (f1 anyelement, f2 out anyelement, f3 out anyarray)
  AS 'select $1, array[$1,$1]' LANGUAGE sql;
+ ERROR:  function "dup" already exists with same argument types
  SELECT dup(22);
dup   
  
***
*** 508,514 
  (1 row)
  
  SELECT * FROM dup('xyz'::text);
!  f2  |f3 
  -+---
   xyz | {xyz,xyz}
  (1 row)
--- 509,515 
  (1 row)
  
  SELECT * FROM dup('xyz'::text);
!   j  | k 
  -+---
   xyz | {xyz,xyz}
  (1 row)
***
*** 516,521 
--- 517,525 
  -- equivalent specification
  CREATE OR REPLACE FUNCTION dup (inout f2 anyelement, out f3 anyarray)
  AS 'select $1, array[$1,$1]' LANGUAGE sql;
+ ERROR:  cannot change return type of existing function
+ DETAIL:  Row type defined by OUT parameters is different.
+ HINT:  Use DROP FUNCTION first.
  SELECT dup(22);
dup   
  

==


---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])


Re: [HACKERS] Regression test plpgsql vs. rangefuncs conflict

2005-07-01 Thread Tom Lane
Peter Eisentraut <[EMAIL PROTECTED]> writes:
> The regression test files plpgsql and rangefuncs both create a dup() 
> function, and as they are run in parallel this just caused an error for 
> me, as attached.  This just happened once for me, but it still ought to 
> be corrected.

Wups, that's my fault.  Fixed.

regards, tom lane

---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])