Commit Graph

5121 Commits

Author SHA1 Message Date
Bruce Momjian
578e71fee5 This should fix a bug where a row that was updated or
deleted that had another row inserted/updated to its old
value during the same statement or other statements before the
integrity check for noaction would incorrectly error.  This
could happen in deferred constraints or due to triggers or
functions. It's effectively a reworking of the previous patch that
did a not exists to instead do a separate check.

Stephan Szabo
2002-07-30 16:33:21 +00:00
Tom Lane
9f1fc1080e Since we're depending on %option noyywrap in the main scanner now,
we may as well use it in all our flex files.  Make all the flex files
have a consistent set of options.
2002-07-30 16:33:08 +00:00
Bruce Momjian
23a8b77d42 Here are two patches. The guc_and_tablefunc patch addresses the two
changes mentioned above, and also adds a new function to the tablefunc
API. The tablefunc API change adds the following function:

* Oid foidGetTypeId(Oid foid) - Get a function's typeid given the
* function Oid. Use this together with TypeGetTupleDesc() to get a
* TupleDesc which is derived from the function's declared return type.

In the next post I'll send the contrib/tablefunc patch, which
illustrates the usage of this new function. Also attached is a doc patch
for this change. The doc patch also adds a function that I failed to
document previously.

Joe Conway
2002-07-30 16:20:03 +00:00
Bruce Momjian
ceb438ed8c This patch fixes one serious bug (runaway INSERT) and a few rare (and
hard to reproduce) error conditions.

Manfred Koizar
2002-07-30 16:08:33 +00:00
Bruce Momjian
6bdbe4ca2c Move alarm timers to proper location. 2002-07-30 05:35:53 +00:00
Bruce Momjian
b6440a7ece Make statement_timeout apply to entire query string, not per statement. 2002-07-30 05:13:06 +00:00
Bruce Momjian
a89fc4e6de --enable-syslog no longer required to use syslog. 2002-07-30 04:24:54 +00:00
Tom Lane
b5eebc1fd4 Centralize code for interpreting schema references, which had gotten
copied more places than I first thought it would.  This fixes a bug:
a couple of these places were neglecting to enforce USAGE access on
explicitly-referenced schemas.
2002-07-29 23:46:35 +00:00
Tom Lane
7b970bc1bc Use aclcheck_error() in place of ad-hoc permissions complaints. 2002-07-29 23:44:44 +00:00
Tom Lane
ea4686e3e1 Implement CREATE/DROP OPERATOR CLASS. Work still remains: need more
documentation (xindex.sgml should be rewritten), need to teach pg_dump
about it, need to update contrib modules that currently build pg_opclass
entries by hand.  Original patch by Bill Studenmund, grammar adjustments
and general update for 7.3 by Tom Lane.
2002-07-29 22:14:11 +00:00
Tom Lane
a28375140c Fix minor coding error: don't ReleaseSysCache until after last use of
cache entry.
2002-07-29 20:45:44 +00:00
Peter Eisentraut
b0c3c48eb3 Assemble portability modules into libpgport library.
Some makefile simplifications.
2002-07-27 20:10:05 +00:00
Tatsuo Ishii
0345f58496 Implement DROP CONVERSION
Add regression test
2002-07-25 10:07:13 +00:00
Peter Eisentraut
43515ba3f8 Remove _deadcode. 2002-07-24 19:16:43 +00:00
Peter Eisentraut
739adf32ee Remove unused system table columns:
pg_language.lancompiler
pg_operator.oprprec
pg_operator.oprisleft
pg_proc.proimplicit
pg_proc.probyte_pct
pg_proc.properbyte_cpu
pg_proc.propercall_cpu
pg_proc.prooutin_ratio
pg_shadow.usetrace
pg_type.typprtlen
pg_type.typreceive
pg_type.typsend

Attempts to use the obsoleted attributes of pg_operator or pg_proc
in the CREATE commands will be greeted by a warning.  For pg_type,
there is no warning (yet) because pg_dump scripts still contain these
attributes.

Also remove new but already obsolete spellings
isVolatile, isStable, isImmutable in WITH clause.  (Use new syntax
instead.)
2002-07-24 19:11:14 +00:00
Tatsuo Ishii
19a20e04bd Add Japanese README explaining how to add new conversion.
English README will come soon...
2002-07-24 07:05:41 +00:00
Peter Eisentraut
e9c013f4bd Add unique index on pg_cast.oid, and document pg_cast table. 2002-07-22 20:23:19 +00:00
Tom Lane
b70b78656b Tweak CreateTrigger() so that the OID used in the name of an
RI_ConstraintTrigger is the same OID assigned to the pg_trigger row.
This reduces consumption of OIDs and may ease debugging.
2002-07-20 19:55:38 +00:00
Tom Lane
c33a6343cd Code review for SHOW output changes; fix horology expected files for
new SHOW output format.
2002-07-20 15:12:56 +00:00
Bruce Momjian
63cfc6a8e8 Fix problems caused by code drift of API for without-oids. 2002-07-20 06:17:43 +00:00
Bruce Momjian
b34cbe006f The attached patch fixes 2 trivial warnings generated by bison 1.35,
as a result of Peter's recent CREATE CAST changes.

Neil Conway
2002-07-20 05:58:34 +00:00
Bruce Momjian
1ce03603cc > 2. This patch includes the same Table Function API fixes that I
>    submitted on July 9:
>
>    http://archives.postgresql.org/pgsql-patches/2002-07/msg00056.php
>
>    Please disregard that one *if* this one is applied. If this one is
>    rejected please go ahead with the July 9th patch.

The July 9th Table Function API patch mentioned above is now in CVS, so
here is an updated version of the guc patch which should apply cleanly
against CVS tip.

Joe Conway
2002-07-20 05:49:28 +00:00
Bruce Momjian
dcbacde2e3 Patch problems caused by code drift since OID patch creation. 2002-07-20 05:37:45 +00:00
Bruce Momjian
e36f9cd440 Manually apply part of oid patch that didn't apply cleanly. 2002-07-20 05:29:01 +00:00
Bruce Momjian
b0f5086e41 oid is needed, it is added at the end of the struct (after the null
bitmap, if present).

Per Tom Lane's suggestion the information whether a tuple has an oid
or not is carried in the tuple descriptor.  For debugging reasons
tdhasoid is of type char, not bool.  There are predefined values for
WITHOID, WITHOUTOID and UNDEFOID.

This patch has been generated against a cvs snapshot from last week
and I don't expect it to apply cleanly to current sources.  While I
post it here for public review, I'm working on a new version against a
current snapshot.  (There's been heavy activity recently; hope to
catch up some day ...)

This is a long patch;  if it is too hard to swallow, I can provide it
in smaller pieces:

Part 1:  Accessor macros
Part 2:  tdhasoid in TupDesc
Part 3:  Regression test
Part 4:  Parameter withoid to heap_addheader
Part 5:  Eliminate t_oid from HeapTupleHeader

Part 2 is the most hairy part because of changes in the executor and
even in the parser;  the other parts are straightforward.

Up to part 4 the patched postmaster stays binary compatible to
databases created with an unpatched version.  Part 5 is small (100
lines) and finally breaks compatibility.

Manfred Koizar
2002-07-20 05:16:59 +00:00
Bruce Momjian
38dd3ae7d0 The attached patch fixes a build problem with GEQO when using the
PX recombination operator, changes some elog() messages from LOG
to DEBUG1, puts some debugging functions inside the appropriate
#ifdef (not enabled by default), and makes a few other minor
cleanups.

BTW, the elog() change is motivated by at least one user who
has sent a concerned email to -general asking exactly what the
"ERX recombination operator" is, and what it is doing to their
DBMS.

Neil Conway
2002-07-20 04:59:10 +00:00
Bruce Momjian
e77054e029 This patch fixes a regression caused by my recent changes to heap
tuple header.  The fix is based on the thought that HEAP_MOVED_IN is
not needed any more as soon as HEAP_XMIN_COMMITTED has been set.  So
in tqual.c and vacuum.c the HEAP_MOVED bits are cleared when
HEAP_XMIN_COMMITTED is set.

Vacuum robustness is enhanced by rearranging ifs, so that we have a
chance to elog(ERROR, ...) before an assertion fails.

A new regression test is included.

Manfred Koizar
2002-07-20 04:57:13 +00:00
Tom Lane
009d593c48 Small performance improvement --- use indexscan not heapscan to find
pg_statistic rows to delete while dropping a relation.
2002-07-19 22:21:17 +00:00
Bruce Momjian
33b3c95200 Complete TODO item:
* -Add BSD-licensed qsort() for Solaris
2002-07-19 17:35:11 +00:00
Tatsuo Ishii
86270024ff Oops. Too much ifdef out. 2002-07-19 11:09:25 +00:00
Tatsuo Ishii
248cbb5796 Temporary ifdef out migrating functions to avoid compiler warnings. 2002-07-19 00:22:24 +00:00
Bruce Momjian
b75fcf9326 Complete TODO item:
* -HOLDER/HOLDERTAB rename to PROCLOCK/PROCLOCKTAG
2002-07-19 00:17:40 +00:00
Peter Eisentraut
97377048b4 pg_cast table, and standards-compliant CREATE/DROP CAST commands, plus
extension to create binary compatible casts.  Includes dependency tracking
as well.

pg_proc.proimplicit is now defunct, but will be removed in a separate
commit.

pg_dump provides a migration path from the previous scheme to declare
casts.  Dumping binary compatible casts is currently impossible, though.
2002-07-18 23:11:32 +00:00
Bruce Momjian
981d045e88 Complete TODO item:
* Merge LockMethodCtl and LockMethodTable into one shared structure (Bruce)
2002-07-18 23:06:20 +00:00
Peter Eisentraut
85d2a629c6 Create directory before installing files. 2002-07-18 22:58:08 +00:00
Bruce Momjian
7d78bac108 Back out BETWEEN node patch, was causing initdb failure. 2002-07-18 17:14:20 +00:00
Tom Lane
11333426f1 Implement DROP SCHEMA. It lacks support for dropping conversions and
operator classes, both of which are schema-local and so should really
be droppable.
2002-07-18 16:47:26 +00:00
Bruce Momjian
e6f02c8231 REVOKE ALL ON FUNCTION nonexistant() FROM PUBLIC;
Used to report that GRANT could not find function nonexistant.

Rod Taylor
2002-07-18 04:50:10 +00:00
Bruce Momjian
a90db34b54 The attached patch (against HEAD) implements
COPY x (a,d,c,b) from stdin;
  COPY x (a,c) to stdout;

as well as the corresponding changes to pg_dump to use the new
functionality.  This functionality is not available when using
the BINARY option.  If a column is not specified in the COPY FROM
statement, its default values will be used.

In addition to this functionality, I tweaked a couple of the
error messages emitted by the new COPY <options> checks.

Brent Verner
2002-07-18 04:43:51 +00:00
Bruce Momjian
3e22406ec6 Finished the Between patch Christopher started.
Implements between (symmetric / asymmetric) as a node.

Executes the left or right expression once, makes a Const out of the
resulting Datum and executes the >=, <= portions out of the Const sets.

Of course, the parser does a fair amount of preparatory work for this to
happen.

Rod Taylor
2002-07-18 04:41:46 +00:00
Bruce Momjian
7ea5f1d7f1 Here is a patch for the Table Function API. It fixes a bug found by Neil
Conway (BuildTupleFromCStrings sets NULL for pass-by-value types when
intended value is 0). It also implements some other improvements
suggested by Neil.

Joe Conway
2002-07-18 04:40:30 +00:00
Bruce Momjian
404e9a12a5 Move libc replacement files from src/backend/port to src/port. 2002-07-18 04:13:59 +00:00
Tatsuo Ishii
eb335a034b I have committed many support files for CREATE CONVERSION. Default
conversion procs and conversions are added in initdb. Currently
supported conversions are:

UTF-8(UNICODE) <--> SQL_ASCII, ISO-8859-1 to 16, EUC_JP, EUC_KR,
		    EUC_CN, EUC_TW, SJIS, BIG5, GBK, GB18030, UHC,
		    JOHAB, TCVN

EUC_JP <--> SJIS
EUC_TW <--> BIG5
MULE_INTERNAL <--> EUC_JP, SJIS, EUC_TW, BIG5

Note that initial contents of pg_conversion system catalog are created
in the initdb process. So doing initdb required is ideal, it's
possible to add them to your databases by hand, however. To accomplish
this:

psql -f your_postgresql_install_path/share/conversion_create.sql your_database

So I did not bump up the version in cataversion.h.

TODO:
Add more conversion procs
Add [CASCADE|RESTRICT] to DROP CONVERSION
Add tuples to pg_depend
Add regression tests
Write docs
Add SQL99 CONVERT command?
--
Tatsuo Ishii
2002-07-18 02:02:30 +00:00
Tom Lane
5af19e4227 Add more dependency insertions --- this completes the basic pg_depend
functionality.  Of note: dropping a table that has a SERIAL column
defined now drops the associated sequence automatically.
2002-07-16 22:12:20 +00:00
Bruce Momjian
846ea08d34 Oops, sometimes strtol isn't called in pg_atoi, so we do need that badp
check.
2002-07-16 18:34:16 +00:00
Bruce Momjian
7aee5ed3b7 Fix tid to in/out as unsigned. 2002-07-16 17:55:25 +00:00
Tatsuo Ishii
3c7798f068 Add conversion procs for CREATE CONVERSION 2002-07-16 09:25:06 +00:00
Tatsuo Ishii
743b747264 Fix typo 2002-07-16 06:58:44 +00:00
Tatsuo Ishii
d49108fd73 Add FindDefaultConversionProc 2002-07-16 06:58:14 +00:00
Tom Lane
30ec31604d Add code to extract dependencies from an expression tree, and use it
to build dependencies for rules, constraint expressions, and default
expressions.  Repair some problems in the original design of
recursiveDeletion() exposed by more complex dependency sets.  Fix
regression tests that were deleting things in illegal sequences.
2002-07-16 05:53:34 +00:00
Bruce Momjian
1e07ab78cc Add DLLINIT mention in port/Makefile. 2002-07-16 05:49:38 +00:00
Bruce Momjian
3cbb9eb265 Move few remaining src/utils files to backend/port so everything is in
one place.  Everything may be moved to src/utils eventually.

Add DLLINIT variable to simplify makfiles.
2002-07-16 05:46:36 +00:00
Bruce Momjian
b3341ddbf2 Mark 'line' as 'not implemented' in SGML and psql \dT, per Thomas
Lockhart.  initdb not forced.
2002-07-16 03:30:27 +00:00
Bruce Momjian
b9104e3a97 Remove certain Makefile dependencies by using full pathnames in
configure.in.
2002-07-15 21:34:05 +00:00
Tom Lane
7bd631bfa4 Use the dependency mechanism to manage column defaults. We need this
so that dependencies in default expressions (on operators, functions,
etc) can be expressed properly.
2002-07-15 16:33:32 +00:00
Bruce Momjian
e552e9e512 Improve relcache.c error reporting for the next guy who has to debug
this thing.
2002-07-15 01:57:51 +00:00
Tom Lane
d5fa19c6ee Add COMMENT ON CONSTRAINT facility (from Rod Taylor's pg_constraint patch).
Fix comment.c to not depend on parser token values, per discussion awhile
back.
2002-07-14 23:38:13 +00:00
Tom Lane
df3f5dfd19 In DeleteAttributeTuples, use a single indexscan instead of the multiple
scans that will most likely be caused by SearchSysCache probes.  Also,
share some code between index deletion and table deletion.
2002-07-14 21:08:08 +00:00
Tom Lane
942a2e94fa Fix testing of partial-index predicates to work correctly in cases where
varno of index's relation is not 1.  This embarrassing oversight pointed
out by Dmitry Tkach 12-Jul-02.
2002-07-13 19:20:34 +00:00
Bruce Momjian
4db8718e84 Add SET statement_timeout capability. Timeout is in ms. A value of
zero turns off the timer.
2002-07-13 01:02:14 +00:00
Tom Lane
7c6df91dda Second phase of committing Rod Taylor's pg_depend/pg_constraint patch.
pg_relcheck is gone; CHECK, UNIQUE, PRIMARY KEY, and FOREIGN KEY
constraints all have real live entries in pg_constraint.  pg_depend
exists, and RESTRICT/CASCADE options work on most kinds of DROP;
however, pg_depend is not yet very well populated with dependencies.
(Most of the ones that are present at this point just replace formerly
hardwired associations, such as the implicit drop of a relation's pg_type
entry when the relation is dropped.)  Need to add more logic to create
dependency entries, improve pg_dump to dump constraints in place of
indexes and triggers, and add some regression tests.
2002-07-12 18:43:19 +00:00
Bruce Momjian
29dfd5fa26 Change error messages ExecAppend->ExecInsert and ExecReplace->ExecUpdate
as discussed on hackers.
2002-07-11 21:36:20 +00:00
Tatsuo Ishii
fcc962566a Add new CREATE CONVERSION/DROP CONVERSION command.
This is the first cut toward CREATE CONVERSION/DROP CONVERSION implementaion.
The commands can now add/remove tuples to the new pg_conversion system
catalog, but that's all. Still need work to make them actually working.
Documentations, regression tests also need work.
2002-07-11 07:39:28 +00:00
Bruce Momjian
92a77cb870 Oops, proper initialization for domainTypMod was none at all, not 0. 2002-07-09 13:52:14 +00:00
Bruce Momjian
81e7e71941 Fix compiler warning:
int32           domainTypMod = NULL;

should be:

	int32           domainTypMod = 0;
2002-07-09 05:21:35 +00:00
Bruce Momjian
1666970275 I've fixed up the way domain constraints (not null and type length)
are managed as per request.

Moved from merging with table attributes to applying themselves during
coerce_type() and coerce_type_typmod.

Regression tests altered to test the cast() scenarios.

Rod Taylor
2002-07-06 20:16:36 +00:00
Bruce Momjian
c8c2b7fff9 Fix compile error in assert coded added by new DISTINCT ON patch. 2002-07-04 16:44:08 +00:00
Thomas G. Lockhart
68d9fbeb55 Implement the IS DISTINCT FROM operator per SQL99.
Reused the Expr node to hold DISTINCT which strongly resembles
 the existing OP info. Define DISTINCT_EXPR which strongly resembles
 the existing OPER_EXPR opType, but with handling for NULLs required
 by SQL99.
We have explicit support for single-element DISTINCT comparisons
 all the way through to the executor. But, multi-element DISTINCTs
 are handled by expanding into a comparison tree in gram.y as is done for
 other row comparisons. Per discussions, it might be desirable to move
 this into one or more purpose-built nodes to be handled in the backend.
Define the optional ROW keyword and token per SQL99.
 This allows single-element row constructs, which were formerly disallowed
 due to shift/reduce conflicts with parenthesized a_expr clauses.
Define the SQL99 TREAT() function. Currently, use as a synonym for CAST().
2002-07-04 15:24:11 +00:00
Bruce Momjian
c9a7345217 >the extra level of struct naming for pd_opaque has no obvious
>usefulness.
>
>> [...] should I post a patch that puts pagesize directly into
>> PageHeaderData?
>
>If you're so inclined.  Given that pd_opaque is hidden in those macros,
>there wouldn't be much of any gain in readability either, so I haven't
>worried about changing the declaration.

Thanks for the clarification.  Here is the patch.  Not much gain, but at
least it saves the next junior hacker from scratching his head ...


Manfred Koizar
2002-07-02 06:18:57 +00:00
Bruce Momjian
33f1687879 There already was a macro PageGetItemId; this is now used in (almost)
all places, where pd_linp is accessed.  Also introduce new macros
SizeOfPageHeaderData and BTMaxItemSize. This is just source code
cosmetic, no behaviour changed.

Manfred Koizar
2002-07-02 05:48:44 +00:00
Bruce Momjian
8864603f3c Minor code cleanup in bufmgr.c and bufmgr.h, mainly by moving repeated
lines of code into internal routines (drop_relfilenode_buffers,
release_buffer) and by hiding unused routines (PrintBufferDescs,
PrintPinnedBufs) behind #ifdef NOT_USED. Remove AbortBufferIO()
declaration from bufmgr.c (already declared in bufmgr.h)

Manfred Koizar
2002-07-02 05:47:37 +00:00
Bruce Momjian
97bfffe50e This patch, which is built upon the "HeapTupleHeader accessor macros"
patch from 2002-06-10, is supposed to reduce the heap tuple header size
by four bytes on most architectures.  Of course it changes the on-disk
tuple format and therefore requires initdb.

This overlays cmin/cmax/xmax fields into only two fields.

Manfred Koizar
2002-07-02 05:46:14 +00:00
Tom Lane
131f801d37 First phase of applying Rod Taylor's pg_depend patch. This just adds
RESTRICT/CASCADE syntax to the DROP commands that need it, and propagates
the behavioral option through the parser to the routines that execute
drops.  Doesn't do anything useful yet, but I figured I'd commit these
changes so I could get out of the parser area while working on the rest.
2002-07-01 15:27:56 +00:00
Bruce Momjian
4f0de24486 Restore error message I accidentally changed. 2002-06-26 22:16:54 +00:00
Bruce Momjian
73ad6ca96c The attached patch fixes some spelling mistakes, makes the
comments on one of the optimizer functions a lot more
clear, adds a summary of the recent KSQO discussion to the
comments in the code, adds regression tests for the bug with
sequence state Tom fixed recently and another reg. test, and
removes some PostQuel legacy stuff: ExecAppend -> ExecInsert,
ExecRetrieve -> ExecSelect, etc.

Error messages remain unchanged until a vote.

Neil Conway
2002-06-26 21:58:56 +00:00
Tom Lane
893fe4919d Treat \r as white space when parsing pg_hba and related files.
Should make life easier for DBAs who insist on editing files with
Windoze tools.
2002-06-26 14:52:08 +00:00
Bruce Momjian
e2c007046f Back out cleanup patch. Got old version and needs work.
Neil Conway
2002-06-25 17:58:10 +00:00
Bruce Momjian
ed275aea42 The attached patch fixes some spelling mistakes, makes the
comments on one of the optimizer functions a lot more
clear, adds a summary of the recent KSQO discussion to the
comments in the code, adds regression tests for the bug with
sequence state Tom fixed recently and another reg. test, and
removes some PostQuel legacy stuff: ExecAppend -> ExecInsert,
ExecRetrieve -> ExecSelect, etc. This was changed because the
elog() messages from this routine are user-visible, so we
should be using the SQL terms.

Neil Conway
2002-06-25 17:27:20 +00:00
Bruce Momjian
8a9462867a Here is a patch for a memory leak in rtree.c, version 7.2.1 (in code
that I submitted last year, alas).

Kenneth Been
2002-06-25 17:26:11 +00:00
Bruce Momjian
5a15149736 It seems that ExecInit/EndIndexScan is leaking some memory...
For example, if I run a query, that uses an index scan, and call
MemoryContextSt ats (CurrentMemoryContext) before ExecutorStart() and
after ExecutorEnd() in ProcessQuery(), I am consistently see ing that
the 'after' call shows 256 bytes more used, then 'before'...

The problem seems to be in ExecEndIndexScan - it does not release
scanstate, ind exstate, indexstate->iss_RelationDescs and indexstate ->
iss_ScanDescs...

Dmitry Tkach
2002-06-23 21:29:32 +00:00
Thomas G. Lockhart
90edb265e3 Implement SQL99 CREATE CAST and DROP CAST statements.
Also implement alternative forms to expose the PostgreSQL CREATE FUNCTION
 features.
Implement syntax for READ ONLY and READ WRITE clauses in SET TRANSACTION.
 READ WRITE is already implemented (of course).
Implement syntax for "LIKE table" clause in CREATE TABLE. Should be fairly
 easy to complete since it resembles SELECT INTO.
Implement MATCH SIMPLE clause for foreign key definitions. This is explicit
 SQL99 syntax for the default behavior, so we now support it :)
Start implementation of shorthand for national character literals in
 scanner. For now, just swallow the leading "N", but sometime soon let's
 figure out how to pass leading type info from the scanner to the parser.
 We should use the same technique for binary and hex bit string literals,
 though it might be unusual to have two apparently independent literal
 types fold into the same storage type.
2002-06-22 02:04:55 +00:00
Bruce Momjian
db650b451e Remove Jan's copyright, with his approval. 2002-06-21 02:59:38 +00:00
Bruce Momjian
42ef2c9cb7 Clean up some copyrights. 2002-06-20 20:51:45 +00:00
Bruce Momjian
27bce57766 Add missing SRF file. 2002-06-20 20:37:00 +00:00
Bruce Momjian
d84fe82230 Update copyright to 2002. 2002-06-20 20:29:54 +00:00
Bruce Momjian
ba790a5608 Here is a patch for Composite and Set returning function support. I made
two small changes to the API since last patch, which hopefully completes
the decoupling of composite function support from SRF specific support.

Joe Conway
2002-06-20 17:19:08 +00:00
Bruce Momjian
c2c2fd57ee Improve COPY syntax to use WITH clause, keep backward compatibility. 2002-06-20 16:00:44 +00:00
Bruce Momjian
e136986afb Properly mark rules that return no value. 2002-06-19 15:40:58 +00:00
Bruce Momjian
8080ac74d5 Wrap long gram.y lines. 2002-06-18 17:56:41 +00:00
Bruce Momjian
71fd49e28d Change CREATE DATABASE to use DefElem instead of constructing structure
members in gram.y.  This is the prefered method for WITH and arbitrary
param/value pairs.
2002-06-18 17:27:58 +00:00
Bruce Momjian
3d564953cd Improve rule action ordering in gram.y, more cleanups. 2002-06-18 00:28:11 +00:00
Bruce Momjian
e25f9e4f5d Mark noise keyword actions with {} rather than returning TRUE, like
opt_with and from_in.
2002-06-17 20:38:04 +00:00
Bruce Momjian
42423c7195 Manually indent gram.y to be consistent. 2002-06-17 20:27:43 +00:00
Bruce Momjian
9e1c5fba03 Fix missing 'buf' variable in SSL sources and add missing includes. 2002-06-17 15:23:36 +00:00
Bruce Momjian
7005e4c076 Fix SSL DEBUG levels to match CVS elog flags. 2002-06-17 15:19:28 +00:00
Tom Lane
507ed0332a Repair AlterTableOwner --- was failing for relations with indexes. 2002-06-17 14:31:32 +00:00
Bruce Momjian
e4cd7c315f Rename DEBUG to DEBUG1 in SSL code. 2002-06-17 07:33:25 +00:00
Bruce Momjian
10f05d8127 Merge ALTER GROUP ADD/DROP rules by creating add_drop action. 2002-06-17 07:00:27 +00:00
Bruce Momjian
f91ee129a7 Simplify optional WITH handling in CREATE USER, ALTER USER, CREATE
GROUP.  Make WITH optional in CREATE DATABASE for consistency.
2002-06-17 05:40:32 +00:00
Bruce Momjian
0dbfea39f3 Remove KSQO from GUC and move file to _deadcode. 2002-06-16 00:09:12 +00:00
Tom Lane
b50cbbd66b Fix up memory leakage created by recent changes. 2002-06-15 22:25:40 +00:00
Tom Lane
c63bc32c2c Fix up gcc warnings, improve comments. 2002-06-15 21:52:31 +00:00
Bruce Momjian
eb1ad5b4b5 Patch for current_schemas to optionally include implicit ...
Second cut attached. This one just adds a boolean option to the existing
function to indicate that implicit schemas are to be included (or not).
I remembered the docs as well this time :-)

Dave Page
2002-06-15 20:03:51 +00:00
Bruce Momjian
6e8a1a6717 WriteBuffer return value:
>I'd vote for changing WriteBuffer to
>return void, and have it elog() on bad argument.

Manfred Koizar
2002-06-15 19:59:59 +00:00
Bruce Momjian
918e864f14 Remove some pre-WAL relics:
SharedBufferChanged
  BufferRelidLastDirtied
  BufferTagLastDirtied
  BufferDirtiedByMe

Manfred Koizar
2002-06-15 19:55:38 +00:00
Bruce Momjian
3c35face41 This patch wraps all accesses to t_xmin, t_cmin, t_xmax, and t_cmax in
HeapTupleHeaderData in setter and getter macros called
HeapTupleHeaderGetXmin, HeapTupleHeaderSetXmin etc.

It also introduces a "virtual" field xvac by defining
HeapTupleHeaderGetXvac and HeapTupleHeaderSetXvac.  Xvac is used by
VACUUM, in fact it is stored in t_cmin.

Manfred Koizar
2002-06-15 19:54:24 +00:00
Bruce Momjian
6b9e742458 The macaddr datatype understands most formats of MAC address, except 12
hex digits with no separators, eg 00AABBCCDDEE. This is easily remedied
with the following patch (against 7.2.1):

Mike Wyer
2002-06-15 19:39:33 +00:00
Thomas G. Lockhart
133df7ce70 Add LOCALTIME and LOCALTIMESTAMP functions per SQL99 standard.
Remove ODBC-compatible empty parentheses from calls to SQL99 functions
 for which these parentheses do not match the standard.
Update the ODBC driver to ensure compatibility with the ODBC standard
 for these functions (e.g. CURRENT_TIMESTAMP, CURRENT_USER, etc).
Include a new appendix in the User's Guide which lists the labeled features
 for SQL99 (the labeled features replaced the "basic", "intermediate",
 and "advanced" categories from SQL92). features.sgml does not yet split
 this list into "supported" and "unsupported" lists.
2002-06-15 03:00:09 +00:00
Thomas G. Lockhart
bad5fe9797 Search the existing regular expression cache as a ring buffer.
Will optimize the case for repeated calls for the same expression,
 which seems to be the most common case. Formerly, always searched
 from the first entry.
May want to look at the least-recently-used algorithm to make sure it
 is identifying the right slots to reclaim. Seems silly to do math when
 it seems that we could simply use an incrementing counter...
2002-06-15 02:49:47 +00:00
Bruce Momjian
63a354c4bb Make encryption of stored passwords the default, as discussed months ago. 2002-06-15 01:29:50 +00:00
Bruce Momjian
579eab9880 Remove pfrees of cached pg_pwd file, per Tom Lane. 2002-06-15 00:52:23 +00:00
Bruce Momjian
1ed4a922b8 Yet another SSL patch. :-) This one adds some informational messages
on the server, if DebugLvl >= 2.

The patch also includes a late addition to the last patch
(X509_check_private_key()).  I'm not sure why it the currect
revision wasn't tagged.


Bear Giles
2002-06-14 04:38:04 +00:00
Bruce Momjian
eb7afc1407 SSL patch that adds support for optional client certificates.
If the user has certificates in $HOME/.postgresql/postgresql.crt
and $HOME/.postgresql/postgresql.key exist, they are provided
to the server.  The certificate used to sign this cert must be
known to the server, in $DataDir/root.crt.  If successful, the
cert's "common name" is logged.

Client certs are not used for authentication, but they could be
via the port->peer (X509 *), port->peer_dn (char *) or
port->peer_cn (char *) fields.  Or any other function could be
used, e.g., many sites like the issuer + serial number hash.

Bear Giles
2002-06-14 04:36:58 +00:00
Bruce Momjian
b8b6691e52 Patch that checks ownership and permissions on server static
private key.  (You want it to be a regular file owned by the
database process, with 0400 or 0600 permissions.)

Bear Giles
2002-06-14 04:35:02 +00:00
Bruce Momjian
8f44024639 SSL patch to periodically renegotiate session key.
In order to reduce the risk of cryptanalysis during extended
sessions (or brief ones involving a substantial amount of data),
this patch renegotiates the session key after 64kib has been
transferred.

Bear Giles
2002-06-14 04:33:53 +00:00
Bruce Momjian
55d053233c SSL support for ephemeral DH keys.
As the comment headers in be-secure.c discusses, EPH preserves
confidentiality even if the static private key (which is usually
kept unencrypted) is compromised.

Because of the value of this, common default values are hard-coded
to protect the confidentiality of the data even if an attacker
successfully deletes or modifies the external file.

Bear Giles
2002-06-14 04:31:49 +00:00
Bruce Momjian
19570420f5 UPDATED PATCH:
Attached are a revised set of SSL patches.  Many of these patches
are motivated by security concerns, it's not just bug fixes.  The key
differences (from stock 7.2.1) are:

*) almost all code that directly uses the OpenSSL library is in two
   new files,

     src/interfaces/libpq/fe-ssl.c
     src/backend/postmaster/be-ssl.c

   in the long run, it would be nice to merge these two files.

*) the legacy code to read and write network data have been
   encapsulated into read_SSL() and write_SSL().  These functions
   should probably be renamed - they handle both SSL and non-SSL
   cases.

   the remaining code should eliminate the problems identified
   earlier, albeit not very cleanly.

*) both front- and back-ends will send a SSL shutdown via the
   new close_SSL() function.  This is necessary for sessions to
   work properly.

   (Sessions are not yet fully supported, but by cleanly closing
   the SSL connection instead of just sending a TCP FIN packet
   other SSL tools will be much happier.)

*) The client certificate and key are now expected in a subdirectory
   of the user's home directory.  Specifically,

	- the directory .postgresql must be owned by the user, and
	  allow no access by 'group' or 'other.'

	- the file .postgresql/postgresql.crt must be a regular file
	  owned by the user.

	- the file .postgresql/postgresql.key must be a regular file
	  owned by the user, and allow no access by 'group' or 'other'.

   At the current time encrypted private keys are not supported.
   There should also be a way to support multiple client certs/keys.

*) the front-end performs minimal validation of the back-end cert.
   Self-signed certs are permitted, but the common name *must*
   match the hostname used by the front-end.  (The cert itself
   should always use a fully qualified domain name (FDQN) in its
   common name field.)

   This means that

	  psql -h eris db

   will fail, but

	  psql -h eris.example.com db

   will succeed.  At the current time this must be an exact match;
   future patches may support any FQDN that resolves to the address
   returned by getpeername(2).

   Another common "problem" is expiring certs.  For now, it may be
   a good idea to use a very-long-lived self-signed cert.

   As a compile-time option, the front-end can specify a file
   containing valid root certificates, but it is not yet required.

*) the back-end performs minimal validation of the client cert.
   It allows self-signed certs.  It checks for expiration.  It
   supports a compile-time option specifying a file containing
   valid root certificates.

*) both front- and back-ends default to TLSv1, not SSLv3/SSLv2.

*) both front- and back-ends support DSA keys.  DSA keys are
   moderately more expensive on startup, but many people consider
   them preferable than RSA keys.  (E.g., SSH2 prefers DSA keys.)

*) if /dev/urandom exists, both client and server will read 16k
   of randomization data from it.

*) the server can read empheral DH parameters from the files

     $DataDir/dh512.pem
     $DataDir/dh1024.pem
     $DataDir/dh2048.pem
     $DataDir/dh4096.pem

   if none are provided, the server will default to hardcoded
   parameter files provided by the OpenSSL project.

Remaining tasks:

*) the select() clauses need to be revisited - the SSL abstraction
   layer may need to absorb more of the current code to avoid rare
   deadlock conditions.  This also touches on a true solution to
   the pg_eof() problem.

*) the SIGPIPE signal handler may need to be revisited.

*) support encrypted private keys.

*) sessions are not yet fully supported.  (SSL sessions can span
   multiple "connections," and allow the client and server to avoid
   costly renegotiations.)

*) makecert - a script that creates back-end certs.

*) pgkeygen - a tool that creates front-end certs.

*) the whole protocol issue, SASL, etc.

 *) certs are fully validated - valid root certs must be available.
    This is a hassle, but it means that you *can* trust the identity
    of the server.

 *) the client library can handle hardcoded root certificates, to
    avoid the need to copy these files.

 *) host name of server cert must resolve to IP address, or be a
    recognized alias.  This is more liberal than the previous
    iteration.

 *) the number of bytes transferred is tracked, and the session
    key is periodically renegotiated.

 *) basic cert generation scripts (mkcert.sh, pgkeygen.sh).  The
    configuration files have reasonable defaults for each type
    of use.

Bear Giles
2002-06-14 04:23:17 +00:00
Bruce Momjian
eb43af3210 Back out SSL changes. Newer patch available. 2002-06-14 04:09:37 +00:00
Bruce Momjian
a9bd17616e Attached are a revised set of SSL patches. Many of these patches
are motivated by security concerns, it's not just bug fixes.  The key
differences (from stock 7.2.1) are:

*) almost all code that directly uses the OpenSSL library is in two
   new files,

     src/interfaces/libpq/fe-ssl.c
     src/backend/postmaster/be-ssl.c

   in the long run, it would be nice to merge these two files.

*) the legacy code to read and write network data have been
   encapsulated into read_SSL() and write_SSL().  These functions
   should probably be renamed - they handle both SSL and non-SSL
   cases.

   the remaining code should eliminate the problems identified
   earlier, albeit not very cleanly.

*) both front- and back-ends will send a SSL shutdown via the
   new close_SSL() function.  This is necessary for sessions to
   work properly.

   (Sessions are not yet fully supported, but by cleanly closing
   the SSL connection instead of just sending a TCP FIN packet
   other SSL tools will be much happier.)

*) The client certificate and key are now expected in a subdirectory
   of the user's home directory.  Specifically,

	- the directory .postgresql must be owned by the user, and
	  allow no access by 'group' or 'other.'

	- the file .postgresql/postgresql.crt must be a regular file
	  owned by the user.

	- the file .postgresql/postgresql.key must be a regular file
	  owned by the user, and allow no access by 'group' or 'other'.

   At the current time encrypted private keys are not supported.
   There should also be a way to support multiple client certs/keys.

*) the front-end performs minimal validation of the back-end cert.
   Self-signed certs are permitted, but the common name *must*
   match the hostname used by the front-end.  (The cert itself
   should always use a fully qualified domain name (FDQN) in its
   common name field.)

   This means that

	  psql -h eris db

   will fail, but

	  psql -h eris.example.com db

   will succeed.  At the current time this must be an exact match;
   future patches may support any FQDN that resolves to the address
   returned by getpeername(2).

   Another common "problem" is expiring certs.  For now, it may be
   a good idea to use a very-long-lived self-signed cert.

   As a compile-time option, the front-end can specify a file
   containing valid root certificates, but it is not yet required.

*) the back-end performs minimal validation of the client cert.
   It allows self-signed certs.  It checks for expiration.  It
   supports a compile-time option specifying a file containing
   valid root certificates.

*) both front- and back-ends default to TLSv1, not SSLv3/SSLv2.

*) both front- and back-ends support DSA keys.  DSA keys are
   moderately more expensive on startup, but many people consider
   them preferable than RSA keys.  (E.g., SSH2 prefers DSA keys.)

*) if /dev/urandom exists, both client and server will read 16k
   of randomization data from it.

*) the server can read empheral DH parameters from the files

     $DataDir/dh512.pem
     $DataDir/dh1024.pem
     $DataDir/dh2048.pem
     $DataDir/dh4096.pem

   if none are provided, the server will default to hardcoded
   parameter files provided by the OpenSSL project.

Remaining tasks:

*) the select() clauses need to be revisited - the SSL abstraction
   layer may need to absorb more of the current code to avoid rare
   deadlock conditions.  This also touches on a true solution to
   the pg_eof() problem.

*) the SIGPIPE signal handler may need to be revisited.

*) support encrypted private keys.

*) sessions are not yet fully supported.  (SSL sessions can span
   multiple "connections," and allow the client and server to avoid
   costly renegotiations.)

*) makecert - a script that creates back-end certs.

*) pgkeygen - a tool that creates front-end certs.

*) the whole protocol issue, SASL, etc.

 *) certs are fully validated - valid root certs must be available.
    This is a hassle, but it means that you *can* trust the identity
    of the server.

 *) the client library can handle hardcoded root certificates, to
    avoid the need to copy these files.

 *) host name of server cert must resolve to IP address, or be a
    recognized alias.  This is more liberal than the previous
    iteration.

 *) the number of bytes transferred is tracked, and the session
    key is periodically renegotiated.

 *) basic cert generation scripts (mkcert.sh, pgkeygen.sh).  The
    configuration files have reasonable defaults for each type
    of use.

Bear Giles
2002-06-14 03:56:47 +00:00
Tatsuo Ishii
15378a53f8 Add support for GB18030 2002-06-14 03:30:56 +00:00
Bruce Momjian
c66eb00adc Allow ANALYZE to run in a transaction. 2002-06-13 19:52:02 +00:00
Tom Lane
f67a931aa4 Make WHERE conditions pulled up from subqueries be executed before outer
WHERE conditions, if there is no reason to do it differently.
2002-06-13 15:10:25 +00:00
Thomas G. Lockhart
bbc9b759d6 Add PLACING to the list of reserved keywords. Added as a token in the last
update to support the new OVERLAY() function.
2002-06-13 14:16:43 +00:00
Tatsuo Ishii
14f72b9a4d Add GB18030 support. Contributed by Bill Huang <bill_huanghb@ybb.ne.jp>
(ODBC support has not been committed yet. left for Hiroshi...)
2002-06-13 08:30:22 +00:00
Tatsuo Ishii
620dbc98cd make namein multibyte aware 2002-06-13 06:19:45 +00:00
Tom Lane
ecb5269404 Further tweaks to support display of sort keys in EXPLAIN --- initial
implementation didn't work for Sort nodes associated with Append plans.
2002-06-13 03:40:49 +00:00
Tom Lane
59c325bb73 Repair for bug #691 --- CREATE TABLE AS column aliases fail to be
applied when the select is a UNION (or other set-operation).

An alternative route to a fix would be to leave analyze.c alone and
change plan_set_operations in prepunion.c to take column names from
the topmost targetlist.  But I am not sure that would work in all
cases.  This patch seems the minimum-risk fix.
2002-06-13 02:04:46 +00:00
Thomas G. Lockhart
ea01a451cc Implement SQL99 OVERLAY(). Allows substitution of a substring in a string.
Implement SQL99 SIMILAR TO as a synonym for our existing operator "~".
Implement SQL99 regular expression SUBSTRING(string FROM pat FOR escape).
 Extend the definition to make the FOR clause optional.
 Define textregexsubstr() to actually implement this feature.
Update the regression test to include these new string features.
 All tests pass.
Rename the regular expression support routines from "pg95_xxx" to "pg_xxx".
Define CREATE CHARACTER SET in the parser per SQL99. No implementation yet.
2002-06-11 15:44:38 +00:00
Jan Wieck
469cb65aca Katherine Ward wrote:
> Changes to avoid collisions with WIN32 & MFC names...
> 1.  Renamed:
>       a.  PROC => PGPROC
>       b.  GetUserName() => GetUserNameFromId()
>       c.  GetCurrentTime() => GetCurrentDateTime()
>       d.  IGNORE => IGNORE_DTF in include/utils/datetime.h & utils/adt/datetim
>
> 2.  Added _P to some lex/yacc tokens:
>       CONST, CHAR, DELETE, FLOAT, GROUP, IN, OUT

Jan
2002-06-11 13:40:53 +00:00
Bruce Momjian
4cb53d71b9 Remove DEBUG_LEVEL from postgresql.conf. Now uses CLIENT/SERVER_MIN_MESSAGES. 2002-06-08 04:08:19 +00:00
Bruce Momjian
585fef1671 Improve readability of factorial, from Florian Weimer. 2002-06-07 23:43:17 +00:00
Bruce Momjian
8a3f34b558 Fix for factorial(0::int2) returning 1, from sugita@sra.co.jp. 2002-06-07 23:41:53 +00:00
Bruce Momjian
2f297a2fcf The attached patch fixes a problem with InstallXLogFileSegment()'s use
of link() under Cygwin:

    http://archives.postgresql.org/pgsql-cygwin/2002-04/msg00072.php

Note that it appears that BeOS and Netware also have the above or
similar problem.

I have only verified that PostgreSQL builds under Cygwin with this
patch.
Since I cannot reproduce the problem, I cannot verify that the proposed
patch solves it.  Nevertheless, both Barry Pederson and David P.
Caldwell
attest that this patch solves the problem.  See the following for
details:

    http://archives.postgresql.org/pgsql-cygwin/2002-05/msg00043.php
    http://archives.postgresql.org/pgsql-cygwin/2002-05/msg00040.php

Jason Tishler
2002-06-07 21:47:45 +00:00
Thomas G. Lockhart
17adf80b47 Fix timestamp to date conversion for the case where timestamp uses a double
precision storage format. Previously applied the same math as used for the
 64-bit integer storage format case, which was wrong.
 Problem introduced recently when the 64-bit storage format was
 implemented.
2002-06-01 15:52:15 +00:00
Tom Lane
75c93c0a68 Rearrange LOG_CONNECTIONS code so that two log messages are made:
one immediately upon forking to handle a new connection, and one after
the authentication cycle is finished.  Per today's pggeneral discussion.
2002-05-28 23:56:51 +00:00
Tom Lane
5a8ab29adf Queries used by ruleutils were not schema-proof. 2002-05-28 22:16:15 +00:00
Tom Lane
e80d6a1bac In default nextval('foo') expression for a SERIAL column, use double
quotes only when necessary.
2002-05-28 22:15:42 +00:00
Tom Lane
a71a53079c Repair error with not adjusting active scans properly after gistSplit.
Patch from Teodor Sigaev.
2002-05-28 15:22:33 +00:00
Tom Lane
3212cf9417 Distinguish between MaxHeapAttributeNumber and MaxTupleAttributeNumber,
where the latter is made slightly larger to allow for in-memory tuples
containing resjunk attributes.  Responds to today's complaint that one
cannot UPDATE a table containing the allegedly-legal maximum number of
columns.

Also, apply Manfred Koizar's recent patch to avoid extra alignment padding
when there is a null bitmap.  This saves bytes in some cases while not
creating any backward-compatibility problem AFAICS.
2002-05-27 19:53:33 +00:00
Tom Lane
4d567013cf Remove AMI_OVERRIDE tests from tqual.c routines; they aren't necessary
and just slow down normal operations (only fractionally, but a cycle saved
is a cycle earned).  Improve documentation of AMI_OVERRIDE behavior.
2002-05-25 20:00:12 +00:00
Tom Lane
29737d83d9 AlterDatabaseSet() forgot to update the indexes on pg_database. 2002-05-25 16:30:59 +00:00
Tom Lane
de09da547a Wups, managed to break ANALYZE with one aspect of that heap_fetch change. 2002-05-24 19:52:43 +00:00
Tom Lane
3f4d488022 Mark index entries "killed" when they are no longer visible to any
transaction, so as to avoid returning them out of the index AM.  Saves
repeated heap_fetch operations on frequently-updated rows.  Also detect
queries on unique keys (equality to all columns of a unique index), and
don't bother continuing scan once we have found first match.

Killing is implemented in the btree and hash AMs, but not yet in rtree
or gist, because there isn't an equally convenient place to do it in
those AMs (the outer amgetnext routine can't do it without re-pinning
the index page).

Did some small cleanup on APIs of HeapTupleSatisfies, heap_fetch, and
index_insert to make this a little easier.
2002-05-24 18:57:57 +00:00
Tom Lane
c0fdec2b6a Add $(LDFLAGS) to Windows make rule for postgres executable. Needed
to do profiling on Cygwin, per report from Dave Page.
2002-05-22 21:46:40 +00:00
Tom Lane
a2597ef179 Modify sequence state storage to eliminate dangling-pointer problem
exemplified by bug #671.  Moving the storage to relcache turned out to
be a bad idea because relcache might decide to discard the info.  Instead,
open and close the relcache entry on each sequence operation, and use
a record of the current XID to discover whether we already hold
AccessShareLock on the sequence.
2002-05-22 21:40:55 +00:00
Tom Lane
b8ffc99602 Minor kibitzing. 2002-05-22 18:33:15 +00:00
Peter Eisentraut
d60f10b0e7 Add optional "validator" function to languages that can validate the
function body (and other properties) as a function in the language
is created.  This generalizes ad hoc code that already existed for
the built-in languages.

The validation now happens after the pg_proc tuple of the new function
is created, so it is possible to define recursive SQL functions.

Add some regression test cases that cover bogus function definition
attempts.
2002-05-22 17:21:02 +00:00
Tom Lane
df9c8e1a39 Make RelationForgetRelation error out if the relcache entry has nonzero
reference count.  This avoids leaving dangling pointers around, as in
recent bug report against sequences (bug# 671).
2002-05-22 15:57:40 +00:00
Tom Lane
0352e3a783 Un-break table creation. 2002-05-22 15:35:43 +00:00
Hiroshi Inoue
a3519a2fcc Allow
CREATE VIEW as SELECT CTID, ....
  SELECT currtid( a view, ..).
2002-05-22 07:46:58 +00:00