summaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
...
* Backpatch SearchSysCacheTupleCopy fix for init_fcache.Tom Lane2000-06-061-4/+7
|
* Back-patch typeTypeName() fix into REL7_0.Tom Lane2000-06-061-3/+5
|
* Fix error message wordingBruce Momjian2000-06-051-2/+2
|
* Update for 7.0.2.Bruce Momjian2000-06-052-6/+6
|
* Lexer defended us against overlength plain identifiers, but not againstTom Lane2000-06-011-1/+8
| | | | overlength quoted identifiers. Death and destruction ensue...
* Fixups for 7.0.1Bruce Momjian2000-06-012-6/+6
|
* Back-patch fix to recognize clause pairs involving ParamsTom Lane2000-05-311-36/+45
| | | | as being range queries.
* Update pg_hba.conf with more examplesBruce Momjian2000-05-301-1/+1
|
* Update pg_hba.conf.sample with better examples and descriptionsBruce Momjian2000-05-301-92/+106
|
* Copy pg_ident.conf.sample into /lib directory. Rename USERAUTH toBruce Momjian2000-05-302-7/+10
| | | | AUTHTYPE in config file. Patch both branches.
* Back-patch tz name length fixes into REL7_0.Tom Lane2000-05-294-17/+14
|
* array_map() needs to guard against a NULL input array. This problemTom Lane2000-05-281-1/+5
| | | | should go away in 7.1 with the new fmgr, but for 7.0.1 ...
* fastpath neglected to pass a valid isNull flag to the function beingTom Lane2000-05-281-1/+2
| | | | | | called. We hadn't noticed this because C-coded functions mostly don't pay attention to isNull anyway ... but plpgsql functions sure do ... This is fixed in another way for 7.1, but 7.0.1 needs this patch.
* Make EUC_CN support more robust.Tatsuo Ishii2000-05-281-2/+6
|
* Add README.kerbrosBruce Momjian2000-05-276-0/+0
|
* Improve pg_hba.conf examplesBruce Momjian2000-05-271-6/+6
|
* Reduce eqsel()'s fudge-factor for estimating the frequency of valuesTom Lane2000-05-261-3/+6
| | | | | | | other than the most common value in a column. We had had 0.5, make it 0.1 to make it more likely that an indexscan will be chosen. Really need better statistics instead, but this should stem the bleeding meanwhile ...
* Fix \h SELECT to not show SELECT INTO.Bruce Momjian2000-05-261-1/+4
|
* Fix too long syslog message problemTatsuo Ishii2000-05-261-1/+81
|
* Generate a reasonable error message when an aggregate function is appliedTom Lane2000-05-261-21/+34
| | | | to an undecorated relation name (cf. example from Ed Loehr, 5/25/00).
* Spell fixBruce Momjian2000-05-261-2/+2
|
* More paranoia about global variables containing references to long-Tom Lane2000-05-261-4/+12
| | | | since-closed file descriptors...
* After closing frontend socket, set MyProcPort->sock = -1 to ensure thatTom Lane2000-05-261-1/+5
| | | | | | | | | subsequent I/O attempts fail cleanly. I'm speculating about failure scenarios in which we do pq_close, then something in a proc_exit routine opens a file (re-using that kernel FD number), then something else fails and tries to write an elog message to the frontend ... message ends up in opened file, oops. No known examples of this but it seems like a potential hole.
* Add some elog(DEBUG)'s to help diagnose mdblindwrt failures.Tom Lane2000-05-251-1/+17
|
* Clean up sloppy coding of _outAExpr().Tom Lane2000-05-251-3/+7
|
* Modify raw parsetree representation returned by gram.y for SubLinks:Tom Lane2000-05-253-17/+19
| | | | | | | | the oper field should be a valid Node structure so it can be dumped by outfuncs.c without risk of coredump. (We had been using a raw pointer to character string, which surely is NOT a valid Node.) This doesn't cause any backwards compatibility problems for stored rules, since raw unanalyzed parsetrees are never stored.
* Do table renaming in a sane order: physical file rename must happenTom Lane2000-05-251-28/+36
| | | | | | | *last*, after all updating of system catalogs. In old code, an error detected during TypeRename left the relation hosed. Also, add a call to flush the relation's relcache entry, rather than trusting to shared cache invalidation to flush it for us.
* heap_drop() should flush the relcache entry for theTom Lane2000-05-251-1/+4
| | | | relation being dropped.
* Quote database name so that not-all-lowercase names are handled safely.Tom Lane2000-05-251-7/+7
|
* ! * Users doing non-blocking connections need to handle the resetBruce Momjian2000-05-251-2/+10
| | | | | | | ! * themselves, they'll need to check the connection status if we ! * return an error. Alfred Perlstein
* On solaris, createdb/dropdb fails because of strange behavior of system().Tatsuo Ishii2000-05-251-1/+13
| | | | | (it returns error with errno ECHILD upon successful completion of commands). This fix ignores an error from system() if errno == ECHILD.
* Make the world safe for QNX's busted shell ...Tom Lane2000-05-242-444/+456
|
* Make the world safe for QNX's busted shell ...Tom Lane2000-05-242-18/+37
|
* add a --with-setproctitle switch to configure that leaves the use of ↵Marc G. Fournier2000-05-242-432/+532
| | | | | | setproctitle() disabled by default ...
* Update so init displays status with setproctitle too, like the rest of them.Bruce Momjian2000-05-241-0/+1
|
* Make setproctitle update for every query.Bruce Momjian2000-05-242-5/+6
|
* Comment out no-op ps updates for setproctitle(), until we figure outBruce Momjian2000-05-231-2/+2
| | | | what to do.
* comment cleanupBruce Momjian2000-05-231-3/+3
|
* Fix problem in which sloppily-coded test in ExecInitIndexScan wouldTom Lane2000-05-232-60/+36
| | | | | | | | | | | | think that both sides of indexqual look like index keys. An example is create table inside (f1 float8 primary key); create table outside (g1 float8, g2 float8); select * from inside,outside where f1 = atan2(g1+1, g2); ERROR: ExecInitIndexScan: both left and right ops are rel-vars (note that failure is potentially platform-dependent). Solution is a cleanup I had had in mind to make anyway: functional index keys should be represented as Var nodes in the fixed indexqual, just like regular index keys.
* Twiddle C++ configuration checks so that we #include <string> (ifTom Lane2000-05-232-443/+463
| | | | | | | it exists) before testing 'using namespace std'. This is necessary on some C++ setups where the compiler won't take a 'using' until you've included a header that mentions namespace std. (Pretty braindead if you ask me, but...)
* Transmit -I switches selected by configure (particularly those fromTom Lane2000-05-235-405/+409
| | | | | | --with-includes) to makefiles for pltcl and plperl, so that these switches will be used even though we do not want other top-level CFLAGS. Ain't it fun trying to support multiple-compiler platforms?
* Cast parameters to ckfree to (char *) to suppress compiler warnings.Tom Lane2000-05-231-9/+9
|
* Irix fixes from Murad NayalBruce Momjian2000-05-222-3/+3
|
* CleanupBruce Momjian2000-05-221-2/+2
|
* I am attempting to integrate postgres (v 7.0) with an open sourceBruce Momjian2000-05-224-11/+12
| | | | | | | | | | | | | | | | | | | | project I am working on (Recall - a distributed, fault-tolerant, replicated, storage framework @ http://www.fault-tolerant.org). Recall is written in C++. I need to include the postgres headers and there are some problems when including the headers w/C++. Attached is a patch generated from postgres/src that fixes my problems. I was hoping to get this into the main source. It's very small (2k) and 3 files are changed: backend/utils/fmgr/fmgr.c, backend/utils/Gen_fmgrtab.sh.in, and include/access/tupdesc.h. In C++, you get a multiply defined symbol because the variable (FmgrInfo *fmgr_pl_finfo) is defined in the header (the patch moves it to the .c file). The other problem in tupdesc.h is the use of typeid is a problem in c++ (I renamed it to oidtypeid). Thanks, Neal Norwitz
* Remove calls to getprotobyname(), which we now know leaks memory onTom Lane2000-05-212-22/+6
| | | | | | some platforms --- and I also see that it is documented as not thread- safe on HPUX and possibly other platforms. No good reason not to just use IPPROTO_TCP constant from <netinet/in.h> instead.
* Repair memory leaks that caused CacheCxt to grow without bound. WeTom Lane2000-05-212-23/+27
| | | | | | | | | really ought to fix relcache entry construction so that it does not do so much with CurrentMemoryContext = CacheCxt. As is, relatively harmless leaks in either sequential or index scanning translate to permanent leaks if they occur when called from relcache build. For the moment, however, the path of least resistance is to repair all such leaks...
* Add debug code to aid in memory-leak tracking: if SHOW_MEMORY_STATS isTom Lane2000-05-215-9/+71
| | | | | defined then statistics about memory usage of all the global memory contexts are printed after each commit.
* Clean up grotty references to CacheCxt (externs inside functions,Tom Lane2000-05-203-10/+8
| | | | duplicate global declarations, no points for style at all!)
* Enhance multibyte support.Tatsuo Ishii2000-05-201-1181/+425
| | | | SJIS UDC (NEC selection IBM kanji) support contributed by Eiji Tokuya
View Lorry Controller Status