OSDN Git Service
PostgreSQL Daemon [Mon, 14 Jan 2002 13:15:28 +0000 (13:15 +0000)]
tag as beta 5 for *hopefully* a very very short beta cycle on this one?
Bruce Momjian [Sun, 13 Jan 2002 17:52:00 +0000 (17:52 +0000)]
Fix recreation of sequence files for 7.2 also because of XID change from
beta4 to beta5.
Michael Meskes [Sun, 13 Jan 2002 08:52:09 +0000 (08:52 +0000)]
- Fixed variable handling for struct members.
- Removed check for array input. An attribut might store the
complete array.
Bruce Momjian [Sun, 13 Jan 2002 05:36:42 +0000 (05:36 +0000)]
Now works for 7.1 databases.
Bruce Momjian [Sun, 13 Jan 2002 04:55:44 +0000 (04:55 +0000)]
Now tested on 7.2->7.2 migration; still disabled. Need 7.1 testing now.
Bruce Momjian [Sun, 13 Jan 2002 01:22:27 +0000 (01:22 +0000)]
Keep it disabled.
Bruce Momjian [Sun, 13 Jan 2002 01:13:59 +0000 (01:13 +0000)]
Wow, it actually worked for the regression database. More testing needed.
Tom Lane [Sat, 12 Jan 2002 18:09:04 +0000 (18:09 +0000)]
Bump catversion to ensure people apply recent timestamp and sequence
fixes.
Tom Lane [Sat, 12 Jan 2002 18:07:34 +0000 (18:07 +0000)]
Update horology expected results (this variant extrapolated by yours
truly; should be verified by someone with an appropriate system).
Tom Lane [Sat, 12 Jan 2002 18:06:53 +0000 (18:06 +0000)]
Update horology expected results (this variant verified on HPUX).
Thomas G. Lockhart [Sat, 12 Jan 2002 04:38:38 +0000 (04:38 +0000)]
Repair bugs in declarations of routines to add timestamptz and interval.
Thanks to Bruce for spotting it and Tom Lane for diagnosing it.
Since horology test output is changing anyway, add some date/time input
tests to horology.sql. Some of these should move to the tests for the
individual data types, and we perhaps should add an entire new test
for "timezone" to allow manipulating the current time zone without
risking damage to the results of other tests.
Tom Lane [Fri, 11 Jan 2002 23:21:55 +0000 (23:21 +0000)]
Fix pg_dump to read-lock all tables to be dumped as soon as it's read
their names from pg_class. This considerably reduces the window wherein
someone could DROP or ALTER a table that pg_dump is intending to dump.
Not a perfect solution, but definitely an improvement. Per complaints
from Marc Fournier; patch by Brent Verner with some kibitzing by Tom Lane.
Tom Lane [Fri, 11 Jan 2002 21:27:13 +0000 (21:27 +0000)]
Code review for latest changes.
Bruce Momjian [Fri, 11 Jan 2002 20:48:47 +0000 (20:48 +0000)]
More man pages improvements; also needed to test sgml build.
Bruce Momjian [Fri, 11 Jan 2002 20:34:14 +0000 (20:34 +0000)]
Sync up.
Tom Lane [Fri, 11 Jan 2002 20:07:03 +0000 (20:07 +0000)]
VACUUM must make sure that a HEAP_MARKED_FOR_UPDATE tuple gets marked
as either HEAP_XMAX_COMMITTED or HEAP_XMAX_INVALID once the updating
transaction is gone. Otherwise some other transaction may come along
and try to test the commit status of t_xmax later --- which could be
after VACUUM has recycled the CLOG status for that xact. Bug introduced
in post-beta4 bug fix.
Tom Lane [Fri, 11 Jan 2002 18:16:04 +0000 (18:16 +0000)]
Fix sequence creation to set the t_xmin of a sequence's tuple to
FrozenTransactionId, not the XID of the creating transaction. Without
this it's possible for a reference to a long-gone CLOG record to occur,
per Christian Meunier's bug report of 10-Jan-02. Worse, the sequence
tuple would become invisible to SELECTs after 2 billion transactions.
Since the fix is applied during sequence creation it does not help
existing databases, unless you drop and recreate every sequence.
However, we intend to force initdb for 7.2RC1 anyway, to fix a pg_proc
error, so I see no need to do more for this problem.
Michael Meskes [Fri, 11 Jan 2002 14:43:11 +0000 (14:43 +0000)]
Added Christof's fixes.
Bruce Momjian [Fri, 11 Jan 2002 06:48:41 +0000 (06:48 +0000)]
More testing, more cleanups.
Bruce Momjian [Fri, 11 Jan 2002 06:33:01 +0000 (06:33 +0000)]
pg_resetxlog specifies log location in hex; more pg_upgrade testing
improvments.
Bruce Momjian [Fri, 11 Jan 2002 06:08:02 +0000 (06:08 +0000)]
More cleanups.
Hiroshi Inoue [Fri, 11 Jan 2002 06:01:47 +0000 (06:01 +0000)]
*** empty log message ***
Bruce Momjian [Fri, 11 Jan 2002 05:54:59 +0000 (05:54 +0000)]
More updates for int8 sequence restoration. Old dump file not required
anymore.
Bruce Momjian [Fri, 11 Jan 2002 04:39:19 +0000 (04:39 +0000)]
Handle int4-int8 sequence migration without full data dump.
Hiroshi Inoue [Fri, 11 Jan 2002 02:50:01 +0000 (02:50 +0000)]
Add a directory to save the changes until 7.3-tree is branched.
Bruce Momjian [Fri, 11 Jan 2002 00:27:42 +0000 (00:27 +0000)]
More improvements using pg_resetxlog. still disabled.
Bruce Momjian [Thu, 10 Jan 2002 23:46:13 +0000 (23:46 +0000)]
Have pg_resetxlog -n return zero exit status.
Bruce Momjian [Thu, 10 Jan 2002 23:41:49 +0000 (23:41 +0000)]
Update FAQ.
Bruce Momjian [Thu, 10 Jan 2002 23:40:58 +0000 (23:40 +0000)]
Remove 7.1beta WAL handling from /contrib/pg_resetxlog.
Bruce Momjian [Thu, 10 Jan 2002 23:35:20 +0000 (23:35 +0000)]
Update FAQ.
Bruce Momjian [Thu, 10 Jan 2002 23:35:19 +0000 (23:35 +0000)]
Update FAQ.
Bruce Momjian [Thu, 10 Jan 2002 23:34:14 +0000 (23:34 +0000)]
Update FAQ.
Bruce Momjian [Thu, 10 Jan 2002 23:33:35 +0000 (23:33 +0000)]
Update FAQ.
Bruce Momjian [Thu, 10 Jan 2002 23:32:11 +0000 (23:32 +0000)]
Update FAQ.
Bruce Momjian [Thu, 10 Jan 2002 23:07:07 +0000 (23:07 +0000)]
Update FAQ.
Bruce Momjian [Thu, 10 Jan 2002 20:09:06 +0000 (20:09 +0000)]
Add new -l flag to set checkpoint location for /contrib/pg_resetxlog.
Peter Eisentraut [Thu, 10 Jan 2002 18:22:44 +0000 (18:22 +0000)]
Temporary workaround for links within bibliography references fouling up
footnote count.
Bruce Momjian [Thu, 10 Jan 2002 18:08:29 +0000 (18:08 +0000)]
Per Tom, do full reset on -x flag.
Bruce Momjian [Thu, 10 Jan 2002 17:51:52 +0000 (17:51 +0000)]
Add new -x XID option to /contrib/pg_resetxlog for future pg_upgrade use.
Michael Meskes [Thu, 10 Jan 2002 10:42:54 +0000 (10:42 +0000)]
Include sqlca.h automatically.
Bruce Momjian [Thu, 10 Jan 2002 04:58:19 +0000 (04:58 +0000)]
More pg_upgrade improvements. Almost done, except for max transaction
setting.
Bruce Momjian [Thu, 10 Jan 2002 03:05:48 +0000 (03:05 +0000)]
More improvements; still disabled. (Don't panic.)
Tom Lane [Thu, 10 Jan 2002 01:11:45 +0000 (01:11 +0000)]
The result of getopt() should be compared to -1, not EOF, per
pgsql-hackers discussion of this date.
Tom Lane [Wed, 9 Jan 2002 23:38:06 +0000 (23:38 +0000)]
Describe type casts under the heading of Value Expressions, and explain
the difference between a run-time type cast and casting a literal string
to a specific type. Minor editorial work in same area.
Bruce Momjian [Wed, 9 Jan 2002 21:50:52 +0000 (21:50 +0000)]
Update with new features. Still disabled.
Tom Lane [Wed, 9 Jan 2002 19:13:41 +0000 (19:13 +0000)]
Fix use of 'char' to hold result of getc, per bug report forwarded by
Oliver Elphick. A few other minor cleanups while at it.
Tom Lane [Wed, 9 Jan 2002 18:21:46 +0000 (18:21 +0000)]
Fix portability problem (result of getopt is int, not char).
Bruce Momjian [Wed, 9 Jan 2002 16:08:54 +0000 (16:08 +0000)]
Remove clog copy.
Bruce Momjian [Wed, 9 Jan 2002 16:00:46 +0000 (16:00 +0000)]
Make improvements to pg_upgrade; is still disabled.
Bruce Momjian [Wed, 9 Jan 2002 04:56:44 +0000 (04:56 +0000)]
Add quotes around $bindir.
Peter Eisentraut [Wed, 9 Jan 2002 00:52:38 +0000 (00:52 +0000)]
Make spelling of operating system names match official version closely.
And listing QNX 6 as both supported and unsupported is probably not helpful.
Tom Lane [Wed, 9 Jan 2002 00:06:42 +0000 (00:06 +0000)]
Fix include paths for case of VPATH build.
Peter Eisentraut [Tue, 8 Jan 2002 23:50:37 +0000 (23:50 +0000)]
Adjust spelling of article titles to match the originals, add some links.
Tom Lane [Tue, 8 Jan 2002 23:40:02 +0000 (23:40 +0000)]
make install failed in VPATH build.
Tom Lane [Tue, 8 Jan 2002 23:34:47 +0000 (23:34 +0000)]
Fix copy-and-paste mistake exposed by gcc warning.
Tom Lane [Tue, 8 Jan 2002 21:00:30 +0000 (21:00 +0000)]
In a VPATH build, resultmap must be one of the symlinked files, else
pg_regress doesn't see it and you don't get any port-specific
comparisons.
Peter Eisentraut [Tue, 8 Jan 2002 20:41:28 +0000 (20:41 +0000)]
Revert last change (CFLAGS+=-g). Probably was a mistake...
Bruce Momjian [Tue, 8 Jan 2002 20:04:31 +0000 (20:04 +0000)]
Update Lockhart's email address.
< * Thomas is Thomas Lockhart <lockhart@alumni.caltech.edu>
---
> * Thomas is Thomas Lockhart <lockhart@fourpalms.org>
Bruce Momjian [Tue, 8 Jan 2002 20:03:58 +0000 (20:03 +0000)]
Update Thomas Lockhart's email address to lockhart@fourpalms.org.
Tom Lane [Tue, 8 Jan 2002 19:02:51 +0000 (19:02 +0000)]
Remove shift/reduce conflicts introduced by last change.
Peter Eisentraut [Tue, 8 Jan 2002 18:03:26 +0000 (18:03 +0000)]
Repair the GEQO bibliography. It was marked up all wrong. Move some of
the info to the main bibliography.
Tom Lane [Tue, 8 Jan 2002 17:03:41 +0000 (17:03 +0000)]
Fix lpad() and rpad() to produce correct results in variable-length
multibyte encodings.
Peter Eisentraut [Tue, 8 Jan 2002 16:13:41 +0000 (16:13 +0000)]
Point to main CPAN site, and reword a little.
Tom Lane [Tue, 8 Jan 2002 15:38:42 +0000 (15:38 +0000)]
Minor wordsmithing.
Peter Eisentraut [Tue, 8 Jan 2002 15:16:16 +0000 (15:16 +0000)]
Make FreeBSD URL point to the right chapter again.
Michael Meskes [Tue, 8 Jan 2002 14:25:06 +0000 (14:25 +0000)]
Fixed array pointers, no longer using void * in arithmetics.
Bruce Momjian [Tue, 8 Jan 2002 05:45:19 +0000 (05:45 +0000)]
Email address no longer valid.
Bruce Momjian [Tue, 8 Jan 2002 05:39:26 +0000 (05:39 +0000)]
Update bad URL's with good ones.
Bruce Momjian [Mon, 7 Jan 2002 22:36:51 +0000 (22:36 +0000)]
Update my2pg to version 1.22.
Tom Lane [Mon, 7 Jan 2002 18:56:09 +0000 (18:56 +0000)]
Add variants of digest() and hmac() that accept text inputs.
Marko Kreen says:
This is so obvious that I would like to make it 'official'.
Seems like the theology around bytea<>text casting kept me from
seeing the simple :)
Tom Lane [Mon, 7 Jan 2002 16:33:00 +0000 (16:33 +0000)]
Tweak LWLock algorithms so that an awakened waiter for a lock is not
granted the lock when awakened; the signal now only means that the lock
is potentially available. The waiting process must retry its attempt
to get the lock when it gets to run. This allows the lock releasing
process to re-acquire the lock later in its timeslice. Since LWLocks
are usually held for short periods, it is possible for a process to
acquire and release the same lock many times in a timeslice. The old
spinlock-based implementation of these locks allowed for that; but the
original coding of LWLock would force a process swap for each acquisition
if there was any contention. Although this approach reopens the door to
process starvation (a waiter might repeatedly fail to get the lock),
the odds of that being a big problem seem low, and the performance cost
of the previous approach is considerable.
Michael Meskes [Mon, 7 Jan 2002 16:25:45 +0000 (16:25 +0000)]
Fixed parser to accept initializing expressions starting with "(".
Peter Eisentraut [Mon, 7 Jan 2002 02:29:15 +0000 (02:29 +0000)]
Editorial review
Tom Lane [Sun, 6 Jan 2002 21:40:02 +0000 (21:40 +0000)]
If we fail to fork a new backend process, (try to) report the failure
to the client before closing the connection. Before 7.2 this was done
correctly, but new code would simply close the connection with no report
to the client.
Bruce Momjian [Sun, 6 Jan 2002 18:12:00 +0000 (18:12 +0000)]
Fix ecpg wording to be consistent.
Bruce Momjian [Sun, 6 Jan 2002 17:54:14 +0000 (17:54 +0000)]
Remove ecpg TODO section:
> Looking at this I also found an ecpg TODO list in the docs:
>
>
http://candle.pha.pa.us/main/writings/pgsql/sgml/ecpg-develop.html
>
> Seems that TODO section should be removed. Some items are done,
others
> are on the main TODO list.
That's correct. I did not fix the docs for quite some time.
Michael
--
Michael Meskes
Tom Lane [Sun, 6 Jan 2002 03:33:26 +0000 (03:33 +0000)]
Fix copy-and-paste brain damage. (These commands do not create a
database.)
Tom Lane [Sun, 6 Jan 2002 00:37:44 +0000 (00:37 +0000)]
Add CHECK_FOR_INTERRUPTS() in various strategic spots, per comments
from Hiroshi.
Barry Lind [Sat, 5 Jan 2002 22:26:23 +0000 (22:26 +0000)]
Bugfix for bug reported by Marcus Better (marcus@dactylis.com). When preforming
a get on a bytea value the code was running the raw value from the server
through character set conversion, which if the character set was SQL_ASCII
would cause all 8bit characters to become ?'s.
Bruce Momjian [Sat, 5 Jan 2002 04:03:47 +0000 (04:03 +0000)]
Spelling fix:
< * Consider use of open/fctl(O_DIRECT) to minimize OS caching
---
> * Consider use of open/fcntl(O_DIRECT) to minimize OS caching
Bruce Momjian [Fri, 4 Jan 2002 21:32:28 +0000 (21:32 +0000)]
Re-order items:
> * Consider use of open/fctl(O_DIRECT) to minimize OS caching
> * Make blind writes go through the file descriptor cache
391d392
< * Make blind writes go through the file descriptor cache
409d409
< * Consider use of open/fctl(O_DIRECT) to minimize OS caching
Bruce Momjian [Fri, 4 Jan 2002 21:08:42 +0000 (21:08 +0000)]
Added:
> * Consider use of open/fctl(O_DIRECT) to minimize OS caching
Bruce Momjian [Fri, 4 Jan 2002 18:58:46 +0000 (18:58 +0000)]
ecpg.sgml manual page updated to ON.
---------------------------------------------------------------------------
When you run 'ecpg --help' you get the following:
-t turn on autocommit of transactions
amongst the other options... Shouldn't this be OFF as per the
documentation?
Best regards, Lee.
--
Lee Kindness, Senior Software Engineer, lkindness@csl.co.uk
Bruce Momjian [Fri, 4 Jan 2002 18:42:40 +0000 (18:42 +0000)]
Change:
< perhaps CREATE DATABASE dbname WITH USER = "user" (Gavin)
---
> perhaps CREATE DATABASE dbname WITH OWNER = "user" (Gavin)
Bruce Momjian [Fri, 4 Jan 2002 18:23:56 +0000 (18:23 +0000)]
Add from Vadim:
> * Allow command blocks that can ignore certain types of errors
Tom Lane [Fri, 4 Jan 2002 17:06:51 +0000 (17:06 +0000)]
Fix some incorrect and obsolete commentary.
Thomas G. Lockhart [Fri, 4 Jan 2002 17:02:25 +0000 (17:02 +0000)]
Add more complete details on date/time keywords and parsing rules.
Move some tabular information on these from the chapter on data types to
the appendix on dates and times.
Thomas G. Lockhart [Fri, 4 Jan 2002 15:49:42 +0000 (15:49 +0000)]
Have to_date() call timestamptz_date() per Karel's email instructions.
Fixes time zone problems introduced by Thomas' implementation of
TIMESTAMP WITHOUT TIME ZONE which caused the behavior of the previously
appropriate routine, timestamp_date(), to change for the worse in this
context.
Bruce Momjian [Fri, 4 Jan 2002 05:50:25 +0000 (05:50 +0000)]
Replace #ifdef, #endif with # requires comment.
Bruce Momjian [Fri, 4 Jan 2002 05:44:45 +0000 (05:44 +0000)]
Update FAQ.
Bruce Momjian [Fri, 4 Jan 2002 05:03:15 +0000 (05:03 +0000)]
Give attribution:
< perhaps CREATE DATABASE dbname WITH USER = "user"
---
> perhaps CREATE DATABASE dbname WITH USER = "user" (Gavin)
Bruce Momjian [Fri, 4 Jan 2002 04:53:26 +0000 (04:53 +0000)]
Remove:
< * Allow only owner to create indexes
Tom Lane [Thu, 3 Jan 2002 23:21:32 +0000 (23:21 +0000)]
Require ownership permission for CREATE INDEX, per bug report.
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.
Tom Lane [Thu, 3 Jan 2002 21:52:05 +0000 (21:52 +0000)]
Instead of waiting a fixed amount of time for the test postmaster to
start up, wait until a psql connection attempt succeeds. Per gripe
from Jason Tishler.
Bruce Momjian [Thu, 3 Jan 2002 21:15:04 +0000 (21:15 +0000)]
Prefer savepoints term:
< 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]
Bruce Momjian [Thu, 3 Jan 2002 20:42:23 +0000 (20:42 +0000)]
Update:
< allow error codes to be specified
---
> allow error codes to be specified; requires nested transactions
Tom Lane [Thu, 3 Jan 2002 20:30:47 +0000 (20:30 +0000)]
SPI_cursor_open must copy by-reference parameter values into the
portal's memory context, so that they will live as long as the portal does.
Bruce Momjian [Thu, 3 Jan 2002 18:25:43 +0000 (18:25 +0000)]
Add:
> o Allow COPY to report error lines and continue; optionally
> allow error codes to be specified
Bruce Momjian [Thu, 3 Jan 2002 18:22:43 +0000 (18:22 +0000)]
Add:
> * Make equals sign optional in CREATE DATABASE WITH param = 'val'
Bruce Momjian [Thu, 3 Jan 2002 18:13:14 +0000 (18:13 +0000)]
Added:
> * Prevent create/drop scripts from allowing extra args (Bruce)