[BUGS] Bug #670: netmask displayed for a /32

2002-05-21 Thread pgsql-bugs
Edwin Groothuis ([EMAIL PROTECTED]) reports a bug with a severity of 3 The lower the number the more severe it is. Short Description netmask displayed for a /32 Long Description The documentation says (chapter 3.8.1): The input format for this type is x.x.x.x/y where x.x.x.x is an IP address and

[BUGS] pg_dump && aggregate bug

2002-05-21 Thread Mathieu Arnold
Hi I have : CREATE FUNCTION "first_cat" (text,text) RETURNS text AS 'SELECT CASE WHEN $1 IS NULL THEN $2 ELSE $1 END' LANGUAGE 'sql'; and : CREATE AGGREGATE first ( BASETYPE = text, SFUNC = first_cat, STYPE = text); when I dump my database, in the dump file, the aggregate becomes : CREATE A

Re: [BUGS] [SQL] Bug with Daylight Savings Time & Interval

2002-05-21 Thread Josh Berkus
Tom and Tom, > This isn't a bug per the existing definition of INTERVAL. '250 days' > is > defined as '250*24 hours', exactly, no more no less. When you move > across a DST boundary you get behavior like the above. > I've opined several times that interval should account for three > separate

Re: [BUGS] Bug #670: netmask displayed for a /32

2002-05-21 Thread Tom Lane
[EMAIL PROTECTED] writes: > But the /y portion is shown in this case: > pmoanalysis=> select ipaddress||'a' from dhcpservers; > ?column? > - > 1.2.3.4/32a > (1 row) This implicitly coerces the inet value to text datatype, and the text(inet) function always displays netmask. Per

Re: [BUGS] pg_dump && aggregate bug

2002-05-21 Thread Tom Lane
Mathieu Arnold <[EMAIL PROTECTED]> writes: > when I dump my database, in the dump file, the aggregate becomes : > CREATE AGGREGATE first ( BASETYPE = text, SFUNC = first_cat, STYPE = text, > INITCOND = '' ); Ooops. This seems to be fixed already in current sources, but I think a back-patch to

Re: [BUGS] Bug #669: gawk: cmd. line:2: (END OF FILE)

2002-05-21 Thread Tom Lane
[EMAIL PROTECTED] writes: > gawk complains of "parse error" & "END OF FILE" when called from > genbki.sh in the src/backend/catalog directory. What platform is this? What is your /bin/sh? Offhand this seems like it is probably the shell's fault, and not directly gawk's.

Re: [BUGS] Bug #669: gawk: cmd. line:2: (END OF FILE)

2002-05-21 Thread Tom Lane
"Lecessi, Ralph" <[EMAIL PROTECTED]> writes: > Sun OS 5.6 > /bin/sh is a Bourne shell. Hmph. Nothing particularly unusual-seeming about that. Do you have another flavor of awk available to try? I still doubt that it's gawk's fault, but we need to eliminate possibilities. Another thing to chec

Re: [BUGS] [SQL] Bug with Daylight Savings Time & Interval

2002-05-21 Thread Thomas Lockhart
> > I've opined several times that interval should account for three > > separate units: months, days, and seconds. But our time-meister > > Tom Lockhart doesn't seem to have taken any interest in the idea. I have taken an interest in the idea. But have not implemented it and have not concluded

Re: [BUGS] [SQL] Bug with Daylight Savings Time & Interval

2002-05-21 Thread Thomas Lockhart
... > Ok, so how should things work, then? While I agree that SQL92's spec > is awkward and limited, we'd need a pretty good argument for breaking > standards. Oliver is already wearing me down in this regard. Well, the standard sucks ;) My reference on this is Date and Darwen (I think that D

Re: [BUGS] Bug #669: gawk: cmd. line:2: (END OF FILE)

2002-05-21 Thread Tom Lane
"Lecessi, Ralph" <[EMAIL PROTECTED]> writes: >>Do you have another flavor of awk available to try? I still doubt that >>it's gawk's fault, but we need to eliminate possibilities. > I ran with nawk (/usr/bin/nawk) & the parse error went away. Interesting. What version of gawk have you got? Peo

Re: [BUGS] Bug #668: cursors with params: mismatched parentheses;

2002-05-21 Thread Tom Lane
[EMAIL PROTECTED] writes: > cursors with parameters do not seems to work. Is that implemented? Sigh. They used to work, but I seem to have broken 'em with a last-minute 7.2 fix. (Wish we had better regression tests for plpgsql.) A fix against 7.2.* is attached. regard