Commit Graph

15162 Commits

Author SHA1 Message Date
Tom Lane
0249c24235 More binary I/O routines. 2003-05-13 18:03:08 +00:00
Michael Meskes
72f311b86a Fixed order of include file search path. 2003-05-13 11:29:14 +00:00
Tom Lane
1c9ac7dfd0 Change pg_amop's index on (amopclaid,amopopr) to index (amopopr,amopclaid).
This makes no difference for existing uses, but allows SelectSortFunction()
and pred_test_simple_clause() to use indexscans instead of seqscans to
locate entries for a particular operator in pg_amop.  Better yet, they can
use the SearchSysCacheList() API to cache the search results.
2003-05-13 04:38:58 +00:00
Tom Lane
30f609484d Add binary I/O routines for a bunch more datatypes. Still a few to go,
but that was enough tedium for one day.  Along the way, move the few
support routines for types xid and cid into a more logical place.
2003-05-12 23:08:52 +00:00
Tom Lane
b02832719c Avoid unnecessary copying of parameter values in BIND. This allows
efficient insertion of large bytea values through the BIND interface.
2003-05-12 16:48:17 +00:00
Tom Lane
d97c9b3662 Apply fixes for problems with dropped columns whose types have also been
dropped.  The simplest fix for INSERT/UPDATE cases turns out to be for
preptlist.c to insert NULLs of a known-good type (I used INT4) rather
than making them match the deleted column's type.  Since the representation
of NULL is actually datatype-independent, this should work fine.
I also re-reverted the patch to disable the use_physical_tlist optimization
in the presence of dropped columns.  It still doesn't look worth the
trouble to be smarter, if there are no other bugs to fix.
Added a regression test to catch future problems in this area.
2003-05-12 00:17:03 +00:00
Tom Lane
a7c50cd92b Finish removing HPUXMATHLIB. 2003-05-11 20:50:53 +00:00
Tom Lane
6f261461c1 Revert previous change --- turns out the underlying problem affects
more cases than I thought, so ExecTypeFromTL() will have to be fixed
anyway.
2003-05-11 20:25:50 +00:00
Tom Lane
8f2e53bc10 Disable the recently-added use_physical_tlist optimization in cases
where the table contains dropped columns.  If the columns are dropped,
then their types may be gone as well, which causes ExecTypeFromTL() to
fail if the dropped columns appear in a plan node's tlist.  This could
be worked around but I don't think the optimization is valuable enough
to be worth the trouble.
2003-05-11 15:03:52 +00:00
Tom Lane
a4e775a263 Make use of new error context stack mechanism to allow random errors
detected during buffer dump to be labeled with the buffer location.
For example, if a page LSN is clobbered, we now produce something like
ERROR:  XLogFlush: request 2C000000/8468EC8 is not satisfied --- flushed only
to 0/8468EF0
CONTEXT:  writing block 0 of relation 428946/566240
whereas before there was no convenient way to find out which page had
been trashed.
2003-05-10 19:04:30 +00:00
Tom Lane
ecc8c3652e In child-crashed log messages, identify a checkpoint process as such
instead of calling it a 'server process'.
2003-05-10 18:15:42 +00:00
Tom Lane
8d86a96068 Adjust CreateCheckpoint so that buffer dumping activities and cleanup of
dead xlog segments are not considered part of a critical section.  It is
not necessary to force a database-wide panic if we get a failure in these
operations.  Per recent trouble reports.
2003-05-10 18:01:31 +00:00
Bruce Momjian
199f24e19d More crypt() cleanup. 2003-05-10 04:06:55 +00:00
Bruce Momjian
37627eb571 Remove HPUX 9 rint() test.
Cleanup new crypt() source code.
2003-05-10 04:03:37 +00:00
Bruce Momjian
cb8b1299a3 Add Makefile rules for new crypt/rint() /port functions. Add include
for rint prototype.
2003-05-10 02:05:50 +00:00
Tom Lane
ba1e066e46 Implement array_send/array_recv (binary I/O for arrays). This exposed
the folly of not passing element type to typsend/typreceive, so fix that.
2003-05-09 23:01:45 +00:00
Tom Lane
b1ee615a7f COPY BINARY uses the new binary I/O routines. Update a few more datatypes
so that COPY BINARY regression test passes.
2003-05-09 21:19:50 +00:00
Tom Lane
38d9919d1a Now that fastpath protocol allows null arguments to be passed,
fastpath.c had better check for strict functions.
2003-05-09 18:18:54 +00:00
Tom Lane
0ac6298bb8 Implement new-protocol binary I/O support in DataRow, Bind, and FunctionCall
messages.  Binary I/O is now up and working, but only for a small set
of datatypes (integers, text, bytea).
2003-05-09 18:08:48 +00:00
Bruce Momjian
d85a0a6bef More cleanups now that we have crypt(). 2003-05-09 16:59:43 +00:00
Bruce Momjian
384bbdd0e1 autoconf run for crypt(). 2003-05-09 16:52:48 +00:00
Bruce Momjian
66e0a8cb39 Add crypt() to /port for Win32. 2003-05-09 16:52:11 +00:00
Bruce Momjian
89eae96b2f Remove rint() prototype from QNX. 2003-05-09 16:35:57 +00:00
Bruce Momjian
3afb84cd0f Remove another old rint() replacement. 2003-05-09 16:31:24 +00:00
Bruce Momjian
0afe5417d7 Add rint() to /port, remove from qnx/. 2003-05-09 16:26:29 +00:00
Bruce Momjian
995773be1c Pass canAcceptConnections to exec'ed backend, for Jan. 2003-05-09 15:57:24 +00:00
Tom Lane
5ecd4e3f30 Binary send/receive routines for a few basic datatypes --- enough for
testing purposes.
2003-05-09 15:44:42 +00:00
Bruce Momjian
4207d6bf30 Add opendir/readdir/closedir() for Win32.
Keep SRA copyright on file because it contains BSD license clause.
2003-05-09 01:16:29 +00:00
Tom Lane
45d04099df Reinstate pg_type's typsend and typreceive columns. They don't do much
yet, but they're there.  Also some editorial work on CREATE TYPE reference
page.
2003-05-08 22:19:58 +00:00
Bruce Momjian
f2c7a27635 Write non-defaults file before signalling the children to reread it. 2003-05-08 20:43:07 +00:00
Bruce Momjian
789a5b340c Move attached to shared memory out into a separate function for clarity. 2003-05-08 19:17:07 +00:00
Tom Lane
075a0078df I think we're done with protocol instability, so mark server and libpq
as speaking the one true 3.0 protocol.
2003-05-08 18:33:39 +00:00
Tom Lane
c0a8c3ac13 Update 3.0 protocol support to match recent agreements about how to
handle multiple 'formats' for data I/O.  Restructure CommandDest and
DestReceiver stuff one more time (it's finally starting to look a bit
clean though).  Code now matches latest 3.0 protocol document as far
as message formats go --- but there is no support for binary I/O yet.
2003-05-08 18:16:37 +00:00
Bruce Momjian
5e7a5c9511 Pass shared memory address on command line to exec'ed backend.
Allow backends to attached to specified shared memory address.
2003-05-08 14:49:04 +00:00
Tom Lane
4e08d35e7d Fix documentation problem pointed out by Barry Lind: code expected a
terminating zero byte in StartupMessage, docs didn't mention it.
2003-05-08 14:35:24 +00:00
Tom Lane
1718f4c66c Update COPY BINARY file format spec to reflect recent decisions about
external representation of binary data.
2003-05-07 22:23:27 +00:00
Tom Lane
2de6da832f Revise protocol doc for recently-agreed modifications. This is the
first version of the 3.0 protocol that can be considered non-draft.
2003-05-07 21:46:15 +00:00
Bruce Momjian
07d35e4277 Change alternate database location patch to test for symlink() rather
than WIN32.
2003-05-07 03:47:08 +00:00
Tom Lane
6165a278e5 Fix erroneous example, per Tim Knowles. 2003-05-07 03:41:31 +00:00
Barry Lind
137b123fc7 Fixed bug with Warnings. Warnings are not added to the Statement instead of
the connection when appropriate.
This checkin also adds the type map for jdbc3, however currently it is
identical to the jdbc2 mapping.

 Modified Files:
 	jdbc/org/postgresql/core/BaseStatement.java
 	jdbc/org/postgresql/core/QueryExecutor.java
 	jdbc/org/postgresql/jdbc3/AbstractJdbc3Connection.java
2003-05-07 03:03:30 +00:00
Bruce Momjian
d9fd7d12f6 Pass shared memory id and socket descriptor number on command line for
fork/exec.
2003-05-06 23:34:56 +00:00
Bruce Momjian
e8f4f2f92d Properly test for buggy flex 2.5.3. 2003-05-06 23:33:52 +00:00
Tom Lane
efa58e105d Minor editorializing. 2003-05-06 23:10:04 +00:00
Tom Lane
755d191700 Add display of eventual result RowDescription (if any) to the output
of Describe on a prepared statement.  This was in the original 3.0
protocol proposal, but I took it out for reasons that seemed good at
the time.  Put it back per yesterday's pghackers discussion.
2003-05-06 21:51:42 +00:00
Tom Lane
8f6a6b7e9a Ensure that an Execute operation can't send tuples in cases where
Describe would claim that no tuples will be returned.  Only affects
SELECTs added to non-SELECT base queries by rewrite rules.  If you
want to see the output of such a select, you gotta use 'simple Query'
protocol.
2003-05-06 21:01:04 +00:00
Tom Lane
79913910d4 Restructure command destination handling so that we pass around
DestReceiver pointers instead of just CommandDest values.  The DestReceiver
is made at the point where the destination is selected, rather than
deep inside the executor.  This cleans up the original kluge implementation
of tstoreReceiver.c, and makes it easy to support retrieving results
from utility statements inside portals.  Thus, you can now do fun things
like Bind and Execute a FETCH or EXPLAIN command, and it'll all work
as expected (e.g., you can Describe the portal, or use Execute's count
parameter to suspend the output partway through).  Implementation involves
stuffing the utility command's output into a Tuplestore, which would be
kind of annoying for huge output sets, but should be quite acceptable
for typical uses of utility commands.
2003-05-06 20:26:28 +00:00
Michael Meskes
299fbb4b37 Added rfmtlong compatibility function. 2003-05-06 09:48:25 +00:00
Bruce Momjian
3ad406bbc9 Rename internal variables DBName|dbName to dbname, for consistency. 2003-05-06 05:15:45 +00:00
Bruce Momjian
77a0219ab1 Comment out some of the conditional tests until we have exec().
Reorder non-default variable loading until PGDATA is defined.
2003-05-06 04:16:35 +00:00
Tom Lane
2cf57c8f8d Implement feature of new FE/BE protocol whereby RowDescription identifies
the column by table OID and column number, if it's a simple column
reference.  Along the way, get rid of reskey/reskeyop fields in Resdoms.
Turns out that representation was not convenient for either the planner
or the executor; we can make the planner deliver exactly what the
executor wants with no more effort.
initdb forced due to change in stored rule representation.
2003-05-06 00:20:33 +00:00