Tom Lane [Wed, 24 Aug 2005 18:16:58 +0000 (18:16 +0000)]
Fix up plperl 'use_strict' so that it can be enabled or disabled on the
fly. Fix problem with incompletely duplicated setup code. Andrew Dunstan,
from an idea of Michael Fuhr's.
Tom Lane [Wed, 24 Aug 2005 17:50:00 +0000 (17:50 +0000)]
Add small hack to support use of Unicode-based locales on WIN32. This
is not adequately tested yet, but let's get it into beta1 so it can be
tested. Magnus Hagander and Tom Lane.
Tom Lane [Wed, 24 Aug 2005 17:38:35 +0000 (17:38 +0000)]
Fix ExecBRDeleteTriggers so that deletion is not suppressed when delete
triggers exist but are all disabled. Problem noted by Gavin Sherry in
original discussion of enable/disable trigger patch, but was never
addressed.
Michael Meskes [Wed, 24 Aug 2005 10:34:19 +0000 (10:34 +0000)]
- Check for NULL before checking whether argument is an array.
- Remove stary character from string quoting.
- Fixed check to report missing varchar pointer implementation.
Bruce Momjian [Tue, 23 Aug 2005 23:51:18 +0000 (23:51 +0000)]
Mark item as complete, add another:
< * Allow triggers to be disabled [trigger]
> * -Allow triggers to be disabled [trigger]
> * Allow triggers to be disabled in only the current session.
< Currently the only way to disable triggers is to modify the system
< tables.
> This is currently possible by starting a multi-statement transaction,
> modifying the system tables, performing the desired SQL, restoring the
> system tables, and committing the transaction. ALTER TABLE ...
> TRIGGER requires a table lock so it is not idea for this usage.
Tom Lane [Tue, 23 Aug 2005 22:40:47 +0000 (22:40 +0000)]
Add ALTER TABLE ENABLE/DISABLE TRIGGER commands. Change pg_dump to
use these instead of its previous hack of changing pg_class.reltriggers.
Documentation is lacking, will add that later.
Patch by Satoshi Nagayasu, review and some extra work by Tom Lane.
Bruce Momjian [Tue, 23 Aug 2005 21:02:05 +0000 (21:02 +0000)]
The attached patch updates the thread test program to run stand-alone on
Windows. The test itself is bypassed in configure as discussed, and
libpq has been updated appropriately to allow it to build in thread-safe
mode.
Tom Lane [Tue, 23 Aug 2005 20:49:47 +0000 (20:49 +0000)]
Clean up some very old and crufty code for TID scan planning. Not much
functional difference really, but make use of stuff added to the planner
since this code was touched last.
Tom Lane [Mon, 22 Aug 2005 23:59:04 +0000 (23:59 +0000)]
Rewrite gather-write patch into something less obviously bolted on
after the fact. Fix bug with incorrect test for whether we are at end
of logfile segment. Arrange for writes triggered by XLogInsert's
is-cache-more-than-half-full test to synchronize with the cache boundaries,
so that in long transactions we tend to write alternating halves of the
cache rather than randomly chosen portions of it; this saves one more
write syscall per cache load.
Tom Lane [Mon, 22 Aug 2005 19:40:37 +0000 (19:40 +0000)]
Cause ALTER INDEX OWNER to generate a warning and do nothing, rather than
erroring out as it has done for the last couple weeks. Document that this
form is now ignored because indexes can't usefully have different owners
from their parent tables. Fix pg_dump to not generate ALTER OWNER commands
for indexes.
Tom Lane [Mon, 22 Aug 2005 17:38:20 +0000 (17:38 +0000)]
Fix unwanted denial of ALTER OWNER rights to superusers. There was some
discussion of getting around this by relaxing the checks made for regular
users, but I'm disinclined to toy with the security model right now,
so just special-case it for superusers where needed.
Bruce Momjian [Mon, 22 Aug 2005 16:59:47 +0000 (16:59 +0000)]
Improve xid wraparound message (the server isn't really shut down, just
not accepting queries).
errmsg("database is not accepting queries to avoid
wraparound data loss in database \"%s\"",
errhint("Stop the postmaster and use a standalone
backend to VACUUM database \"%s\".",
Tom Lane [Mon, 22 Aug 2005 16:27:36 +0000 (16:27 +0000)]
Use an initdb-time test to see if the local version of getaddrinfo()
chokes on IPv6 addresses, and comment out the IPv6 entry in the default
pg_hba.conf if so. Per Andrew Dunstan.
Tom Lane [Mon, 22 Aug 2005 00:41:28 +0000 (00:41 +0000)]
Fix some inconsistent choices of datatypes in xlog.c. Make buffer
indexes all be int, rather than variously int, uint16 and uint32;
add some casts where necessary to support large buffer arrays.
Tom Lane [Sun, 21 Aug 2005 03:39:37 +0000 (03:39 +0000)]
Minor GUC cleanups: document krb_server_hostname and custom_variable_classes
in postgresql.conf.sample, mark custom_variable_classes as SIGHUP not
POSTMASTER to agree with the documentation (I can't see a reason it has
to be POSTMASTER so I think the docs are right).
Tom Lane [Sat, 20 Aug 2005 23:26:37 +0000 (23:26 +0000)]
Convert the arithmetic for shared memory size calculation from 'int'
to 'Size' (that is, size_t), and install overflow detection checks in it.
This allows us to remove the former arbitrary restrictions on NBuffers
etc. It won't make any difference in a 32-bit machine, but in a 64-bit
machine you could theoretically have terabytes of shared buffers.
(How efficiently we could manage 'em remains to be seen.) Similarly,
num_temp_buffers, work_mem, and maintenance_work_mem can be set above
2Gb on a 64-bit machine. Original patch from Koichi Suzuki, additional
work by moi.
Tom Lane [Sat, 20 Aug 2005 19:19:21 +0000 (19:19 +0000)]
Invoke mksafefunc and mkunsafefunc with :: decoration. This seems a good
idea on consistency grounds, whether or not it really fixes bug #1831.
Michael Fuhr
Tom Lane [Sat, 20 Aug 2005 00:40:32 +0000 (00:40 +0000)]
Repair problems with VACUUM destroying t_ctid chains too soon, and with
insufficient paranoia in code that follows t_ctid links. (We must do both
because even with VACUUM doing it properly, the intermediate state with
a dangling t_ctid link is visible concurrently during lazy VACUUM, and
could be seen afterwards if either type of VACUUM crashes partway through.)
Also try to improve documentation about what's going on. Patch is a bit
bulky because passing the XMAX information around required changing the
APIs of some low-level heapam.c routines, but it's not conceptually very
complicated. Per trouble report from Teodor and subsequent analysis.
This needs to be back-patched, but I'll do that after 8.1 beta is out.
Tom Lane [Thu, 18 Aug 2005 17:51:12 +0000 (17:51 +0000)]
Fix up LIMIT/OFFSET planning so that we cope with non-constant LIMIT
or OFFSET clauses by using estimate_expression_value(). The main advantage
of this is that if the expression is a Param and we have a value for the
Param, we'll use that value rather than defaulting. Also, fix some
thinkos in the logic for combining LIMIT/OFFSET with an externally
supplied tuple fraction (this covers cases like EXISTS(...LIMIT...)).
And make sure the results of all this are shown by EXPLAIN. Per a
gripe from Merlin Moncure.
Bruce Momjian [Thu, 18 Aug 2005 04:37:08 +0000 (04:37 +0000)]
Fix to_char(interval) to return proper year and century values.
Fix to_char(interval) to return large year/month/day/hour values that
are larger than possible timestamp values.
Prevent to_char(interval) format specifications that make no sense, like
Month.
Clean up formatting.c code to more logically handle return lengths.
Tom Lane [Wed, 17 Aug 2005 22:14:34 +0000 (22:14 +0000)]
Avoid an Assert failure if OuterUserId hasn't been set yet during
AbortTransaction. This can happen if a backend's InitPostgres transaction
fails (eg, because the given username is invalid). Per Alvaro.
Tom Lane [Tue, 16 Aug 2005 00:48:12 +0000 (00:48 +0000)]
Reject operator names >= NAMEDATALEN characters. These will not work
anyway, and in assert-enabled builds you are likely to get an assertion
failure. Backpatch as far as 7.3; 7.2 seems not to have the problem.
Tom Lane [Mon, 15 Aug 2005 21:50:15 +0000 (21:50 +0000)]
Clean up recent patch for PL handler functions in pg_catalog: the patch
caused PL languages and handlers to be dumped ALWAYS, even in the face
of contrary --schema or --table switches. Adopt a slightly saner
definition.
Tom Lane [Mon, 15 Aug 2005 19:40:20 +0000 (19:40 +0000)]
array_in() and array_recv() need to be more paranoid about validating
their OID parameter. It was possible to crash the backend with
select array_in('{123}',0,0); because that would bypass the needed step
of initializing the workspace. These seem to be the only two places
with a problem, though (record_in and record_recv don't have the issue,
and the other array functions aren't depending on user-supplied input).
Back-patch as far as 7.4; 7.3 does not have the bug.
Tom Lane [Mon, 15 Aug 2005 19:05:16 +0000 (19:05 +0000)]
int_array_enum function should be using fcinfo->flinfo->fn_extra for
working state, not fcinfo->context. Silly oversight on my part in last
go-round of fixes.
Tom Lane [Mon, 15 Aug 2005 16:25:19 +0000 (16:25 +0000)]
Allow the pgstat views to show toast tables as well as regular tables
(the stats system has always collected this info, but the views were
filtering it out). Modify autovacuum so that over-threshold activity
in a toast table can trigger a VACUUM of the parent table, even if the
parent didn't appear to need vacuuming itself. Per discussion a month
or so back about "short, wide tables".
Tom Lane [Sun, 14 Aug 2005 18:49:30 +0000 (18:49 +0000)]
Fix up tab completion for ROLEs and add some more completion logic for
other stuff; change \du and \dg to be role-aware (Stefan Kaltenbrunner).
Also make tab completion fetch the list of GUC variables from pg_settings
instead of having a hard-wired copy of the list (Tom Lane).
Tom Lane [Sat, 13 Aug 2005 22:18:07 +0000 (22:18 +0000)]
Tweak catalog cache management algorithms to reduce cost of
SearchCatCacheList and ReleaseCatCacheList. Previously, we incremented
and decremented the refcounts of list member tuples along with the list
itself, but that's unnecessary, and very expensive when the list is big.
It's cheaper to change only the list refcount. When we are considering
deleting a cache entry, we have to check not only its own refcount but
its parent list's ... but it's easy to arrange the code so that this
check is not made in any commonly-used paths, so the cost is really nil.
The bigger gain though is to refrain from DLMoveToFront'ing each individual
member tuple each time the list is referenced. To keep some semblance
of fair space management, lists are just marked as used or not since the
last cache cleanout search, and we do a MoveToFront pass only when about
to run a cleanout. In combination, these changes reduce the costs of
SearchCatCacheList and ReleaseCatCacheList from about 4.5% of pgbench
runtime to under 1%, according to my gprof results.
Tom Lane [Sat, 13 Aug 2005 19:02:34 +0000 (19:02 +0000)]
Make pg_stat_file() use OUT parameters so that the user doesn't have to
remember the output parameter set for himself. It's a bit of a kluge
but fixing array_in to work in bootstrap mode looks worse.
I removed the separate pg_file_length() function, as it no longer has any
real notational advantage --- you can write (pg_stat_file(...)).length.
Bruce Momjian [Sat, 13 Aug 2005 02:06:21 +0000 (02:06 +0000)]
The large one adds support for RSA keys and reorganizes
the pubkey functions a bit. The actual RSA-specific code
there is tiny, most of the patch consists of reorg of the
pubkey code, as lots of it was written as elgamal-only.
I was bit hasty making the random exponent 'k' a prime. Further researh
shows that Elgamal encryption has no specific needs in respect to k,
any random number is fine.
It is bit different for signing, there it needs to be 'relatively prime'
to p - 1, that means GCD(k, p-1) == 1, which is also a lot lighter than
full primality. As we don't do signing, this can be ignored.
o pgp_mpi_free: Accept NULLs
o pgp_mpi_cksum: result should be 16bit
o Remove function name from error messages - to be similar to other
SQL functions, and it does not match anyway the called function
o remove couple junk lines
o Tom stuck a CVS id into file. I doubt the usefulness of it,
but if it needs to be in the file then rather at the end.
Also tag it as comment for asciidoc.
o Mention bytea vs. text difference
o Couple clarifications
There is a choice whether to update it with pgp functions or
remove it. I decided to remove it, updating is pointless.
I've tried to keep the core of pgcrypto relatively independent
from main PostgreSQL, to make it easy to use externally if needed,
and that is good. Eg. that made development of PGP functions much
nicer.
But I have no plans to release it as generic library, so keeping such
doc
up-to-date is waste of time. If anyone is interested in using it in
other products, he can probably bother to read the source too.
Commented source is another thing - I'll try to make another pass
over code to see if there is anything non-obvious that would need
more comments.
Bruce Momjian [Sat, 13 Aug 2005 01:55:41 +0000 (01:55 +0000)]
Add documentation:
>>>I think a more accurate description would be "permissions not
>>>inherited by children," and that isn't necessarily a bug.
>>
>>I agree it may not be a bug - but it's more than the permissions not
>>being inherited: the parent is affected.
>
>
> Not really, once you understand what's happening. Unless you use
> FROM ONLY, selecting from the parent selects from the parent *and*
> its children. The parent itself isn't affected, as queries with
> FROM ONLY should demonstrate. I understand what you're saying --
> that there's an apparent effect on the parent -- but there really
> isn't.