user/sven/postgresql.git

Age	Commit message (Collapse)	Author
2004-10-02	Clean up handling of inherited-table update queries, per bug report	Tom Lane
	from Sebastian Böck. The fix involves being more consistent about when rangetable entries are copied or modified. Someday we really need to fix this stuff to not scribble on its input data structures in the first place...
2004-10-02	Update length from 75 to 79.	Bruce Momjian

2004-10-01	Convert pg_stat_get_backend_idset to use the existing SRF support.	Tom Lane
	This seems the cleanest way of fixing its lack of a shutdown callback, which was preventing it from working correctly in a query that didn't run it to completion. Per bug report from Szima GÄbor.
2004-10-01	fmgr_security_definer had better do a PG_TRY to ensure the outer userid	Tom Lane
	is restored on error exit.
2004-10-01	Adjust postmaster to recognize that a lockfile containing its parent's PID	Tom Lane
	must be stale. Tweak example startup scripts to not use pg_ctl but launch the postmaster directly, thereby ensuring that only the postmaster's direct parent shell will be a postgres-owned process. In combination these should fix the longstanding problem of the postmaster sometimes refusing to start during reboot because it thinks the old lockfile is not stale.
2004-10-01	Don't assume PQdb() will return a valid result from a failed connection.	Tom Lane

2004-10-01	Don't assume PQdb() will return a valid result from a failed connection.	Tom Lane

2004-10-01	Fallout from changing index locking rules: we can reduce the strength	Tom Lane
	of locking used by REINDEX. REINDEX needs only ShareLock on the parent table, same as CREATE INDEX, plus an exclusive lock on the specific index being processed.
2004-10-01	Code review for NOWAIT patch: downgrade NOWAIT from fully reserved keyword	Tom Lane
	to unreserved keyword, use ereport not elog, assign a separate error code for 'could not obtain lock' so that applications will be able to detect that case cleanly.
2004-10-01	Use 'override' when appending text to LDFLAGS, so the build doesn't break	Neil Conway
	if the user has defined LDFLAGS themselves.
2004-09-30	Adjust index locking rules as per my proposal of earlier today. You	Tom Lane
	now are supposed to take some kind of lock on an index whenever you are going to access the index contents, rather than relying only on a lock on the parent table.
2004-09-30	Remove unnecessary use of index_open just to get the index name.	Tom Lane

2004-09-30	isinf.c needs <float.h> on some platforms, per Darcy Buskermolen.	Tom Lane

2004-09-30	Come to think of it, functions in FROM have the same syntactic restriction	Tom Lane
	as CREATE INDEX did, and can be fixed the same way, for another small improvement in usability and reduction in grammar size.
2004-09-29	Split out everything that looks like a function call from c_expr into	Tom Lane
	a separate production func_expr. This allows us to accept all these variants in the backwards-compatible syntax for creating a functional index; which beats documenting exactly which things work and which don't. Interestingly, it also seems to make the generated state machine a little bit smaller.
2004-09-29	Minor adjustments to improve the accuracy of our computation of required	Tom Lane
	shared memory size.
2004-09-28	Adjust obsolete comment about memory needs for lock table.	Tom Lane

2004-09-28	Arrange to preallocate all required space for the buffer and FSM hash	Tom Lane
	tables in shared memory. This ensures that overflow of the lock table creates no long-lasting problems. Per discussion with Merlin Moncure.
2004-09-28	An include path got missed out of the patch - see the attached fix.	Bruce Momjian
	This also adds debug build support to src/bin/psql/win32.mak. Dave Page
2004-09-28	Revert patch that removed BUFSIZ usage. The memory has to hold the	Bruce Momjian
	structures plus pointers used by the structure.
2004-09-27	Small WIN32_CLIENT_ONLY cleanup for new build patch.	Bruce Momjian

2004-09-27	Remove use of large BUFSIZ for buffers and use the proper struct sizes.	Bruce Momjian
	This greatly helps threaded libpq programs.
2004-09-27	Add new macro as shorthand for MS VC and Borland C++:	Bruce Momjian
	+ #if defined(_MSC_VER) \|\| defined(__BORLANDC__) + #define WIN32_CLIENT_ONLY + #endif
2004-09-27	More MS VC and BCC psql compile cleanups.	Bruce Momjian

2004-09-27	More Win32 cleanups for Dave Page.	Bruce Momjian

2004-09-27	Make adjustment for MS VC and BCC compile of psql, from Dave Page.	Bruce Momjian

2004-09-27	Improve MS VC builds for psql by handlling flex properly and prevent	Bruce Momjian
	rename prototype conflict.
2004-09-27	Translation update	Peter Eisentraut

2004-09-27	Translation updates	Peter Eisentraut

2004-09-27	Translation update	Peter Eisentraut

2004-09-27	New translations	Peter Eisentraut

2004-09-27	Remove inclusion of windows.h now that it is included in c.h, per idea	Bruce Momjian
	from Peter.
2004-09-27	Synced parser.	Michael Meskes
	Removed Oracle transaction syntax to fix shift/reduce error.
2004-09-27	Remove references to the ODBC driver from the main source tree. From Kris	Neil Conway
	Jurka.
2004-09-27	A few minor list-related cleanups:	Neil Conway
	(1) Replace while loop with the new forboth() construct in parser/analyze.c (2) Replace lcons() with lappend() in SearchCatCacheList(). Since these now have the same performance, there is no reason to prefer lcons() in this case, and using lappend() leads to cleaner code. (3) Improve the name of the second parameter to for_each_cell()
2004-09-27	Code cleanup: don't bother casting the argument to pfree() to void *	Neil Conway
	from another pointer type. Per C89, this is unnecessary, and it is common practice throughout the rest of the tree anyway.
2004-09-27	Rewrite ReadArrayStr() to avoid O(N^2) behavior on large strings,	Tom Lane
	and hopefully improve code clarity while at it. One intentional semantics change: a backslashed space will not be treated as removable trailing whitespace, as the prior coding would do. ISTM that if it wouldn't be considered removable leading whitespace, it shouldn't be stripped at the end either.
2004-09-26	Fix multiple breakages in our support for SSL certificates.	Tom Lane

2004-09-26	Use ULL not LL in UINT64CONST.	Tom Lane

2004-09-26	Here is a patch that adds the version info from libpq.rc to the DLL	Bruce Momjian
	build in mingw. The MSVC build already did this, but it was not linked into the mingw one. This is not the same as the versioninfo patch that's in the queue. Please apply this one before beta-3 if at all possible. Magnus Hagander
2004-09-26	Use <> not "" for include of pthread.h.	Bruce Momjian

2004-09-26	Repair bug that would allow libpq to think a command had succeeded when	Tom Lane
	it really hadn't, due to double output of previous command's response. Fix prevents recursive entry to libpq routines. Found by Jan Wieck.
2004-09-24	Persuade plpython to build on OS X.	Tom Lane

2004-09-24	GUC assign hooks that look at external state in deciding whether a	Tom Lane
	setting is valid must ignore that state and permit the assignment anyway when source is PGC_S_OVERRIDE. Otherwise they may disallow a rollback at transaction abort, which is The Wrong Thing. Per example from Michael Fuhr 12-Sep-04.
2004-09-24	Cleanup some ancient Ultrix / Alpha code in main() that is intended to	Neil Conway
	modify how unaligned memory accesses are dealt with. Document that this is really what is going on, and merge the NOFIXADE and NOPRINTADE code paths.
2004-09-24	shmdt takes a void* parameter, per Kris Jurka.	Tom Lane

2004-09-24	Fix gratuitous weirdness in function prototype, per Kris Jurka.	Tom Lane

2004-09-24	Okay, I'm tired of reading gripes from OS X users ... add a regression	Tom Lane
	variant file to cover OS X 10.3's bizarre minus-zero behavior.
2004-09-24	Get rid of /*-inside-comment warning. My fault.	Tom Lane

2004-09-24	ExecProcAppend() wasn't called ExecAppend() because the latter name was	Neil Conway
	formerly used in execMain. Since that is no longer the case, this patch renames ExecProcAppend() to ExecAppend() for the sake of consistency.