OSDN Git Service
Bruce Momjian [Fri, 17 Dec 1999 01:16:03 +0000 (01:16 +0000)]
initdb.sh fix from Peter.
Bruce Momjian [Fri, 17 Dec 1999 01:05:31 +0000 (01:05 +0000)]
This is my -- hopefully sufficiently portable -- attempt at cleaning out
initdb. No more obscure dependencies on environment variables or paths.
It
now finds the templates and the right postgres itself (with cmd line
options as fallback). It also no longer depends on $USER (su safe), and
doesn't advertise that --username allows you to install the db as a
different user, since that doesn't work anyway. Also, recovery and
cleanup
on all errors. Consistent options, clearer documentation.
Please take a look at this and adopt it if you feel it's safe enough. I
have simulated all the stupid circumstances I could think of, but you
never know with shell scripts.
Oh yeah, you can give the postgres user a default password now.
--
Peter Eisentraut Sernanders väg 10:115
Bruce Momjian [Thu, 16 Dec 1999 23:54:41 +0000 (23:54 +0000)]
Clear paren level flag on \r or any backslash command, rather than
keeping parenlevel unchanged.
Jan Wieck [Thu, 16 Dec 1999 22:20:03 +0000 (22:20 +0000)]
Some changes to prepare for LONG attributes.
Jan
Bruce Momjian [Thu, 16 Dec 1999 20:10:02 +0000 (20:10 +0000)]
Hi,
I sending promised patch with:
* getopt_long() - for pg_dump (portable)
* and "Usage: " changes in scripts in src/bin/
- this changes are cosmetic only, not change any
feature ...etc.
All PostgreSQL routines (scripts) support now long options and
help's output is alike for all scripts and all support -? or --help.
Karel
Karel Zak <zakkr@zf.jcu.cz> http://home.zf.jcu.cz/~zakkr/
Bruce Momjian [Thu, 16 Dec 1999 20:07:41 +0000 (20:07 +0000)]
>Turning nextval and currval into keywords is not an acceptable way to
>go about this. That will risk breaking existing applications that use
>those names as column names.
>
>It should actually almost work to write sq.nextval as things stand,
>because Postgres has for a long time considered table.function and
>function(table) to be interchangeable notations for certain kinds of
>functions. nextval doesn't seem to be one of that kind of function,
>at the moment. I'd suggest leaving the grammar as it was, and taking a
>look at ParseFuncOrColumn in parse_func.c to see if you can't persuade
>it to accept the sequence functions in that style.
OK, good point. I tried to implement it somewhere else and ended up
extending transformAttr. Attached you'll find the patch.
Jeroen van Vianen
Bruce Momjian [Thu, 16 Dec 1999 17:24:19 +0000 (17:24 +0000)]
Here's the Create/Alter/Drop Group stuff that's been really overdue. I
didn't have time for documentation yet, but I'll write some. There are
still some things to work out what happens when you alter or drop users,
but the group stuff in and by itself is done.
--
Peter Eisentraut Sernanders väg 10:115
Bruce Momjian [Thu, 16 Dec 1999 16:52:53 +0000 (16:52 +0000)]
Update for QNX.
Bruce Momjian [Thu, 16 Dec 1999 16:34:06 +0000 (16:34 +0000)]
autoconf
Michael Meskes [Thu, 16 Dec 1999 06:53:12 +0000 (06:53 +0000)]
*** empty log message ***
Bruce Momjian [Thu, 16 Dec 1999 01:30:49 +0000 (01:30 +0000)]
Ethernet MAC addresses (macaddr type) are not compared correctly for
equality. The lobits macro is wrong and extracts the wrong set of
bits out of the structure.
To exhibit the problem:
select '000000:000000'::macaddr = '000000:110000'::macaddr ;
?column?
--------
t
(1 row)
Daniel Boyd
Bruce Momjian [Thu, 16 Dec 1999 01:25:23 +0000 (01:25 +0000)]
I have done the QNX4 port with the current source tree. The number of
backend/Makefiles to be patched could significantly be reduced since
they
have been adopted to the QNX4 needs.
Andreas Kardos
Bruce Momjian [Tue, 14 Dec 1999 22:03:48 +0000 (22:03 +0000)]
This patch solves a couple of memory leaks in ecpglib.c. The patch is
ok for both the
development tree (CVS) and for 6.5.3.
Stephen Birch
Tom Lane [Tue, 14 Dec 1999 03:35:28 +0000 (03:35 +0000)]
fix_parsetree_attnums was not nearly smart enough about walking parse
trees. Also rewrite find_all_inheritors() in a more intelligible style.
Bruce Momjian [Tue, 14 Dec 1999 00:17:33 +0000 (00:17 +0000)]
> From what I gather, this should be a little cleaner because the
triggered
> function now returns the right datatype.
Oops, I got crossed up with Jan's improvements. Ignore this.
--
Peter Eisentraut Sernanders väg 10:115
peter_e@gmx.net 75262 Uppsala
Bruce Momjian [Tue, 14 Dec 1999 00:12:06 +0000 (00:12 +0000)]
>From what I gather, this should be a little cleaner because the
triggered
function now returns the right datatype.
--
Peter Eisentraut Sernanders väg 10:115
Bruce Momjian [Tue, 14 Dec 1999 00:08:21 +0000 (00:08 +0000)]
Depending on my interpreting (and programming) skills, this might solve
anywhere from zero to two TODO items.
* Allow flag to control COPY input/output of NULLs
I got this:
COPY table .... [ WITH NULL AS 'string' ]
which does what you'd expect. The default is \N, otherwise you can use
empty strings, etc. On Copy In this acts like a filter: every data item
that looks like 'string' becomes a NULL. Pretty straightforward.
This also seems to be related to
* Make postgres user have a password by default
If I recall this discussion correctly, the problem was actually that the
default password for the postgres (or any) user is in fact "\N", because
of the way copy is used. With this change, the file pg_pwd is copied out
with nulls as empty strings, so if someone doesn't have a password, the
password is just '', which one would expect from a new account. I don't
think anyone really wants a hard-coded default password.
Peter Eisentraut Sernanders väg 10:115
Bruce Momjian [Mon, 13 Dec 1999 22:35:27 +0000 (22:35 +0000)]
New LDOUT makefile variable for QNX os.
Tom Lane [Mon, 13 Dec 1999 17:39:38 +0000 (17:39 +0000)]
Update documentation to reflect availability of aggregate(DISTINCT).
Try to provide a more lucid discussion in 'Using Aggregate Functions'
tutorial section.
Bruce Momjian [Mon, 13 Dec 1999 04:54:01 +0000 (04:54 +0000)]
Prevent _deadcode from showing in ctags and mkid
Tom Lane [Mon, 13 Dec 1999 01:27:21 +0000 (01:27 +0000)]
aggregate(DISTINCT ...) works, per SQL spec.
Note this forces initdb because of change of Aggref node in stored rules.
Tom Lane [Sun, 12 Dec 1999 20:51:29 +0000 (20:51 +0000)]
any_ordering_op()'s argument should be declared Oid not int.
Bruce Momjian [Sun, 12 Dec 1999 05:57:36 +0000 (05:57 +0000)]
I'm in TODO mood today ...
* Document/trigger/rule so changes to pg_shadow recreate pg_pwd
I did it with a trigger and it seems to work like a charm. The function
that already updates the file for create and alter user has been made a
built-in "SQL" function and a trigger is created at initdb time.
Comments around the pg_pwd updating function seem to be worried about
this
routine being called concurrently, but I really don't see a reason to
worry about this. Verify for yourself. I guess we never had a system
trigger before, so treat this with care, and feel free to adjust the
nomenclature as well.
--
Peter Eisentraut Sernanders väg 10:115
Bruce Momjian [Sun, 12 Dec 1999 05:15:10 +0000 (05:15 +0000)]
Meanwhile, database names with single quotes in names don't work very well
at all, and because of shell quoting rules this can't be fixed, so I put
in error messages to that end.
Also, calling create or drop database in a transaction block is not so
good either, because the file system mysteriously refuses to roll back rm
calls on transaction aborts. :) So I put in checks to see if a transaction
is in progress and signal an error.
Also I put the whole call in a transaction of its own to be able to roll
back changes to pg_database in case the file system operations fail.
The alternative location issues I posted recently were untouched, awaiting
the outcome of that discussion. Other than that, this should be much more
fool-proof now.
The docs I cleaned up as well.
Peter Eisentraut Sernanders väg 10:115
Bruce Momjian [Sat, 11 Dec 1999 21:35:49 +0000 (21:35 +0000)]
Fix for psql from Peter.
Bruce Momjian [Sat, 11 Dec 1999 01:03:36 +0000 (01:03 +0000)]
Allow ; in () in psql.
Bruce Momjian [Sat, 11 Dec 1999 00:31:37 +0000 (00:31 +0000)]
Update TODO list.
Bruce Momjian [Sat, 11 Dec 1999 00:31:05 +0000 (00:31 +0000)]
This takes care of TODO item
* pg_dump should preserve primary key information
Also a couple of warnings removed.
--
Peter Eisentraut Sernanders väg 10:115
Bruce Momjian [Fri, 10 Dec 1999 17:46:41 +0000 (17:46 +0000)]
Remove old file.
Bruce Momjian [Fri, 10 Dec 1999 17:32:57 +0000 (17:32 +0000)]
Update drop.sql
Bruce Momjian [Fri, 10 Dec 1999 17:31:30 +0000 (17:31 +0000)]
Rename destroy.sql to drop.sql.
Jan Wieck [Fri, 10 Dec 1999 12:34:15 +0000 (12:34 +0000)]
Added global variable to have RI triggers override
time qualification of HeapTupleSatisfiesSnapshot()
Jan
Tatsuo Ishii [Fri, 10 Dec 1999 10:29:01 +0000 (10:29 +0000)]
Fix memory overrun while setting ps status
Tom Lane [Fri, 10 Dec 1999 07:37:35 +0000 (07:37 +0000)]
Teach grammar and parser about aggregate(DISTINCT ...). No implementation
yet, but at least we can give a better error message:
regression=> select count(distinct f1) from int4_tbl;
ERROR: aggregate(DISTINCT ...) is not implemented yet
instead of 'parser: parse error at or near distinct'.
Bruce Momjian [Fri, 10 Dec 1999 05:17:13 +0000 (05:17 +0000)]
Remove unneeded action.
Bruce Momjian [Fri, 10 Dec 1999 05:00:36 +0000 (05:00 +0000)]
Fix LDREL.
Bruce Momjian [Fri, 10 Dec 1999 03:59:30 +0000 (03:59 +0000)]
This should fix the \e (\p, \g, ...) behaviour on an empty query buffer.
Also, minor tweakage of tab completion, and I hope the output is flushed
on time now.
--
Peter Eisentraut Sernanders väg 10:115
Bruce Momjian [Fri, 10 Dec 1999 03:56:14 +0000 (03:56 +0000)]
Rename several destroy* functions/tags to drop*.
Tom Lane [Fri, 10 Dec 1999 03:01:05 +0000 (03:01 +0000)]
Correct coredump in ALTER TABLE foo ADD(). Accept explicit NULL in
typecasts, eg 'NULL::text'. Later parts of the parser don't like this
yet, but I'll work on that next.
Bruce Momjian [Fri, 10 Dec 1999 00:48:32 +0000 (00:48 +0000)]
Update regression for new dropdb command.
Jan Wieck [Fri, 10 Dec 1999 00:26:35 +0000 (00:26 +0000)]
Added a fflush(stdout) to the \p command required for the
multisession test tool.
Jan
Bruce Momjian [Thu, 9 Dec 1999 19:15:45 +0000 (19:15 +0000)]
Make LD -r as macros that can be changed for QNX.
Bruce Momjian [Thu, 9 Dec 1999 15:56:16 +0000 (15:56 +0000)]
Update for QNX from Kardos, Dr. Andrea
Tom Lane [Thu, 9 Dec 1999 05:58:56 +0000 (05:58 +0000)]
Replace generic 'Illegal use of aggregates' error message with one that
shows the specific ungrouped variable being complained of. Perhaps this
will reduce user confusion...
Bruce Momjian [Thu, 9 Dec 1999 05:02:24 +0000 (05:02 +0000)]
Hi,
I was able to crash postgres 6.5.3 when I did an 'alter user' command.
After I started a debugger I found the problem in the timezone handling
of
datetime (my Linux box lost its timezone information, that's how the
problem occurred).
Only 7 bytes are reserved for the timezone, without checking for
boundaries.
Attached is a patch that fixes this problem and emits a NOTICE if a
timezone is encountered that is longer than MAXTZLEN bytes, like this:
Jeroen van Vianen
Bruce Momjian [Thu, 9 Dec 1999 04:36:57 +0000 (04:36 +0000)]
Remove extra vacuum, from Peter.
Jan Wieck [Wed, 8 Dec 1999 20:41:22 +0000 (20:41 +0000)]
Fixed concurrent visibility bug.
Jan
Bruce Momjian [Wed, 8 Dec 1999 11:37:38 +0000 (11:37 +0000)]
Fix indexing of cidr.
Bruce Momjian [Wed, 8 Dec 1999 11:17:20 +0000 (11:17 +0000)]
Cleanup
Bruce Momjian [Wed, 8 Dec 1999 10:29:55 +0000 (10:29 +0000)]
Change $< to $+ in scripts Makefile, and clean up pg_encoding if logic.
Michael Meskes [Wed, 8 Dec 1999 09:52:29 +0000 (09:52 +0000)]
*** empty log message ***
Bruce Momjian [Tue, 7 Dec 1999 22:43:03 +0000 (22:43 +0000)]
Peter's file removals.
Bruce Momjian [Tue, 7 Dec 1999 22:41:44 +0000 (22:41 +0000)]
Okay, that should put us back in sync. These two patches (src & doc) are
against the sources from one hour ago and contain all the portable and
up
to date stuff.
A few other CVS "householding" things you might want to take care of:
* Remove the src/bin/cleardbdir directory
* Remove the file src/bin/psql/sql_help.h from the repository, as it is
a derived file and is build by the release_prep.
Peter Eisentraut
Michael Meskes [Tue, 7 Dec 1999 10:29:18 +0000 (10:29 +0000)]
*** empty log message ***
Tom Lane [Tue, 7 Dec 1999 04:09:39 +0000 (04:09 +0000)]
Clean up memory leakage in find_inheritors() by using pg_list lists
(which are palloc'd) instead of DLLists (which are malloc'd). Not very
significant, since this routine seldom has anything useful to do, but
a leak is a leak...
Jan Wieck [Tue, 7 Dec 1999 00:13:41 +0000 (00:13 +0000)]
Again updated INSIDE TODO
Jan
Jan Wieck [Tue, 7 Dec 1999 00:11:35 +0000 (00:11 +0000)]
Added SET DEFAULT action for ON DELETE/UPDATE
Jan
Jan Wieck [Mon, 6 Dec 1999 19:52:36 +0000 (19:52 +0000)]
Corrected TODO in source
Jan
Jan Wieck [Mon, 6 Dec 1999 19:50:49 +0000 (19:50 +0000)]
Added ON DELETE/UPDATE SET NULL
Jan
Jan Wieck [Mon, 6 Dec 1999 18:02:47 +0000 (18:02 +0000)]
Completed FOREIGN KEY syntax.
Added functionality for automatic trigger creation during CREATE TABLE.
Added ON DELETE RESTRICT and some others.
Jan
Thomas G. Lockhart [Mon, 6 Dec 1999 16:40:45 +0000 (16:40 +0000)]
Update entries with recent results for Alpha and ARM processors.
Remove references to Ultrix; haven't heard from it for a *long* time :(
Thomas G. Lockhart [Mon, 6 Dec 1999 16:38:25 +0000 (16:38 +0000)]
Fix markup (a couple of missing tags) and renormalize.
Thomas G. Lockhart [Mon, 6 Dec 1999 16:37:11 +0000 (16:37 +0000)]
Remove references to PostgreSQL as "public-domain" since that has a
specific meaning wrt copyright (or lack thereof).
Tatsuo Ishii [Mon, 6 Dec 1999 08:49:00 +0000 (08:49 +0000)]
Fix the case of -o given
Tatsuo Ishii [Mon, 6 Dec 1999 08:35:34 +0000 (08:35 +0000)]
Fix minor bugs
Tatsuo Ishii [Mon, 6 Dec 1999 07:23:41 +0000 (07:23 +0000)]
pg_ctl: a script to start/stop/restart and report status of post
Tatsuo Ishii [Mon, 6 Dec 1999 07:21:12 +0000 (07:21 +0000)]
Fix bugs regarding pid file.
Tom Lane [Mon, 6 Dec 1999 02:38:56 +0000 (02:38 +0000)]
Update rules test for extra quoting in rule deparser.
Tom Lane [Mon, 6 Dec 1999 02:37:17 +0000 (02:37 +0000)]
Rule deparser needs to quote identifiers that are spelled the same as
SQL keywords.
Bruce Momjian [Sun, 5 Dec 1999 21:23:00 +0000 (21:23 +0000)]
Clean up doc/Makefile to install manual pages.
Bruce Momjian [Sun, 5 Dec 1999 20:52:54 +0000 (20:52 +0000)]
Fix echo -n and read -r in scripts.
Bruce Momjian [Sun, 5 Dec 1999 20:29:18 +0000 (20:29 +0000)]
Remove destroylang.sgml.
Bruce Momjian [Sun, 5 Dec 1999 20:22:00 +0000 (20:22 +0000)]
Update.
Bruce Momjian [Sun, 5 Dec 1999 20:04:42 +0000 (20:04 +0000)]
Update
Bruce Momjian [Sun, 5 Dec 1999 20:02:49 +0000 (20:02 +0000)]
I cleaned those out as well (the echo -n "bug" was in there ;) and moved
them into the scripts dir. I also added a --list option to show already
installed languages.
This whole moving and renaming totally confused CVS and my checked out
copy got completely fried last night. When you apply the source patch,
please make sure that all the directories src/bin/{create|destroy}* as
well as vacuumdb, cleardbdir are gone and that all the scripts (7) are
in
scripts/.
Meanwhile I am still puzzled about what happened with the docs patch.
Because I don't know what you got now, the second attachment contains
the
files
ref/allfiles.sgml
ref/commands.sgml
ref/createlang.sgml
ref/droplang.sgml
doc/src/sgml/Makefile
Peter Eisentraut Sernanders väg 10:115
Bruce Momjian [Sun, 5 Dec 1999 19:29:37 +0000 (19:29 +0000)]
This *should* fix the large object segfault problem. Note linewrap may
be interfering with last line of patch.
Vince.
Bruce Momjian [Sat, 4 Dec 1999 12:50:54 +0000 (12:50 +0000)]
Re-add destroylang.
Tatsuo Ishii [Sat, 4 Dec 1999 08:23:43 +0000 (08:23 +0000)]
Unlink the pid file if it's bogus (no associated process exists)
Bruce Momjian [Sat, 4 Dec 1999 05:03:49 +0000 (05:03 +0000)]
Update scripts from Peter.
Bruce Momjian [Sat, 4 Dec 1999 04:53:22 +0000 (04:53 +0000)]
New scripts for create/drop user/db from Peter Eisentraut
Bruce Momjian [Sat, 4 Dec 1999 04:34:56 +0000 (04:34 +0000)]
No distclean for pgaccess/Makefile.
Bruce Momjian [Fri, 3 Dec 1999 18:28:32 +0000 (18:28 +0000)]
Replace examples .H to .h.
Bruce Momjian [Fri, 3 Dec 1999 17:35:05 +0000 (17:35 +0000)]
Rename libpq++.H to libpq++.h.
Jan Wieck [Fri, 3 Dec 1999 12:47:43 +0000 (12:47 +0000)]
Arranged that plpgsql.so is used from new build, not installed
one in production PGLIB.
Jan
Tatsuo Ishii [Fri, 3 Dec 1999 06:26:34 +0000 (06:26 +0000)]
Create postpid and postmaster.opts under $PGDATA
Tom Lane [Thu, 2 Dec 1999 00:26:15 +0000 (00:26 +0000)]
Type 'socklen_t' might be the right way to declare getsockopt()'s last
parameter in some flavor of Unix, but Linux, HPUX, and SunOS all say
it's int. For now I'm just going to make it int so that I can compile.
If the other way is actually necessary on some Unix somewhere, I guess
we will need a configure test...
Bruce Momjian [Wed, 1 Dec 1999 18:03:26 +0000 (18:03 +0000)]
Improve descriptions of date/time functions.
Bruce Momjian [Wed, 1 Dec 1999 00:29:54 +0000 (00:29 +0000)]
Mention index name when reporting corruption.
Bruce Momjian [Tue, 30 Nov 1999 04:29:57 +0000 (04:29 +0000)]
Fix compile error on older patch.
Bruce Momjian [Tue, 30 Nov 1999 03:57:29 +0000 (03:57 +0000)]
create/alter user extension
This one should work much better than the one I sent in previously. The
functionality is the same, but the patch was missing one file resulting
in
the compilation failing. The docs also received a minor fix.
Peter Eisentraut Sernanders väg 10:115
Bruce Momjian [Tue, 30 Nov 1999 03:08:19 +0000 (03:08 +0000)]
This patch (against the current CVS sources) adds to libpq the functions
PQconnectStart
PQconnectPoll
PQresetStart
PQresetPoll
PQsetenvStart
PQsetenvPoll
PQsetenvAbort
and brings into the published interface
PQsetenv.
The first four are asynchronous analogues of PQconnectdb and PQreset -
they allow an application to connect to the DB without blocking on
remote I/O.
The PQsetenv functions perform an environment negotiation with the
server.
Internal to libpq, pqReadReady and pqWriteReady have been made available
across the library (they were previously static functions inside
fe-misc.c). A lot of internal rearrangement has been necessary to
support these changes.
The API documentation has been updated also.
Caveats:
o The Windows code does not default to using non-blocking sockets,
since I have no documentation: Define WIN32_NON_BLOCKING_CONNECTIONS to
do that.
o The SSL code still blocks.
Ewan Mellor.
Bruce Momjian [Tue, 30 Nov 1999 02:44:09 +0000 (02:44 +0000)]
Rename readonly to onlyread in odbc. Use varargs properly.
Bruce Momjian [Tue, 30 Nov 1999 02:14:13 +0000 (02:14 +0000)]
Update TODO list.
Bruce Momjian [Mon, 29 Nov 1999 23:42:03 +0000 (23:42 +0000)]
Small patch which fixes the ODBC driver so it doesn't segfault if:
You have CommLog and Debug enabled
You encounter in error in any operation (SQLConnect/SQLExec).
Previously, the extra logging didn't check for NULL pointers
when trying to print some of the strings- the socket error
message could frequently be NULL by design (if there was no socket
error)
and Solaris does not handle NULLS passed to things like printf
("%s\n",string);
gracefully.
This basically duplicates the functionality found in Linux where passing
a null pointer
to printf prints "(NULL)". No very elegant, but the logging is for debug
only anyway.
Dirk Niggemann
Bruce Momjian [Mon, 29 Nov 1999 23:26:18 +0000 (23:26 +0000)]
I finish devel. of Oracle compatible DateTime routines TO_CHAR(),
TO_DATE()
and PgSQL extension FROM_CHAR().
TO_CHAR() routine allow formating text output with a datetime values:
SELECT TO_CHAR('now'::datetime, '"Now is: "HH24:MI:SS');
to_char
----------------
Now is: 21:04:10
FROM_CHAR() routine allow convert text to a datetime:
SELECT FROM_CHAR('September 1999 10:20:30', 'FMMonth YYYY
HH:MI:SS');
from_char
-----------------------------
Wed Sep 01 10:20:30 1999 CEST
TO_DATE() is equal with FROM_CHAR(), but output a Date only:
SELECT TO_DATE('September 1999 10:20:30', 'FMMonth YYYY
HH:MI:SS');
to_date
----------
09-01-1999
In attache is compressed dir for the contrib. All is prepared, but I'am
not
sure if Makefile is good (probably yes).
Comments & suggestions ?
Thomas, thank you for your good advices.
Karel
------------------------------------------------------------------------------
Karel Zak <zakkr@zf.jcu.cz>
http://home.zf.jcu.cz/~zakkr/
Bruce Momjian [Mon, 29 Nov 1999 22:34:36 +0000 (22:34 +0000)]
Attached are the C-routines that implement a BIT and BIT VARYING type.
Adriaan Joubert
Bruce Momjian [Mon, 29 Nov 1999 05:28:11 +0000 (05:28 +0000)]
I pulled this one out via CVS, I hope you can make use of it:
Peter.
Tom Lane [Mon, 29 Nov 1999 04:43:15 +0000 (04:43 +0000)]
Add permissions check: now one must be the Postgres superuser or the
table owner in order to vacuum a table. This is mainly to prevent
denial-of-service attacks via repeated vacuums. Allow VACUUM to gather
statistics about system relations, except for pg_statistic itself ---
not clear that it's worth the trouble to make that case work cleanly.
Cope with possible tuple size overflow in pg_statistic tuples; I'm
surprised we never realized that could happen. Hold a couple of locks
a little longer to try to prevent deadlocks between concurrent VACUUMs.
There still seem to be some problems in that last area though :-(
Tom Lane [Mon, 29 Nov 1999 04:34:55 +0000 (04:34 +0000)]
Try to detect oversize tuple before corrupting relation, instead of
after...
Tom Lane [Sun, 28 Nov 1999 22:02:17 +0000 (22:02 +0000)]
Fix "Unable to identify an operator =$" problem that occurred when pgsql
expressions were written without spaces between operators and operands.
Problem was that something like "if new.f1=new.f2 then" would be translated
to "if $1=$2 then", and the Postgres lexer would tokenize that the wrong
way. Fix is to emit spaces around $paramno constructs to ensure they are
seen as separate tokens.