Disallow CREATE INDEX on system catalogs, non-tables (views, sequences, etc).
Disallow CREATE/DROP TRIGGER on system catalogs, non-tables.
Disallow ALTER TABLE ADD/DROP CONSTRAINT on system catalogs.
Disallow FOREIGN KEY reference to non-table.
None of these things can actually work in the present system structure,
but the code was letting them pass without complaint.

start up, wait until a psql connection attempt succeeds.  Per gripe
from Jason Tishler.

< 	  allow error codes to be specified; requires nested transactions
---
> 	  allow error codes to be specified; requires savepoints
321c321
< * Allow nested transactions / savepoints [transactions]
---
> * Allow savepoints / nested transactions [transactions]

< 	  allow error codes to be specified
---
> 	  allow error codes to be specified; requires nested transactions

portal's memory context, so that they will live as long as the portal does.

> 	o Allow COPY to report error lines and continue;  optionally
> 	  allow error codes to be specified

> * Make equals sign optional in CREATE DATABASE WITH param = 'val'

> * Prevent create/drop scripts from allowing extra args (Bruce)

> * Allow only owner to create indexes

< * Add config file check for $ODBCINI, $HOME/.odbc.ini, and /etc/odbc.ini
---
> * Add config file check for $ODBCINI, $HOME/.odbc.ini, installpath/etc/odbc.ini

Took test from createlang.

> * Add config file check for $ODBCINI, $HOME/.odbc.ini, and /etc/odbc.ini

> * Make it easier to create a database owned by someone who can't createdb,
>   perhaps CREATE DATABASE dbname WITH USER = "user"

wrote:
> > > Just testing pgcrypto on freebsd/alpha.  I get some warnings:
> > They should be harmless, although I should fix them.
>
> The actual code is:
>
>     if ((dlen & 15) || (((unsigned) res) & 3))
>         return -1;

> Hard to imagine how (uint *) & 3 makes any sense, unless res isn't
> always a (uint8 *).  Is that true?

At some point it was casted to (uint32*) so I wanted to be sure its ok.
ATM its pointless.  Please apply the following patch.

--
marko

tables.

not work.  Fixed to match createlang.

< 	o Move to using 'make' rather than 'ant'(?)
---
> 	o Support both 'make' and 'ant'

existing code of dbname optional actually work.

< * Change 'createlang [langname] dbname' to 'createlang langname [dbname]'

> * Change 'createlang [langname] dbname' to 'createlang langname [dbname]'

per bug report from Laurette Cisneros.

formats will be taken as 2000, not year zero.  Per bug report from
Aasmund Midttun Godal.  Fix from Karel Zak.

(IMHO) the code should be fixed to do what the documentation formerly
claimed, and then revert this change.

macros, but only at explicit CHECK_FOR_INTERRUPTS() calls.  Not clear
whether overenthusiastic acceptance of interrupts accounts for any real
bugs, but it definitely seems risky and unnecessary.

to insert the same key into a supposedly unique index.  The bug is of
low probability, and may not explain any of the recent reports of
duplicated rows; but a bug is a bug.

 token. Seems to be isolated to datetime.c and datetime.h.

values; it's not portable to call them with signed chars.  I recall doing
this for the last release, but a few more uncasted calls have snuck in.

> * Make locking of shared data structures more fine-grained