summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2001-12-11Applied patch from Thomas O'Dowd that fixes timestamp parsing. The jdbc codeBarry Lind
wasn't updated to handle more than two decimal digits for fractional seconds that now are possible in 7.2. This patch fixes the timestamp parsing logic. I have built and tested on both jdbc1 and jdbc2.
2001-12-11Patch from Ned Wolpert that fixes a bug that caused the cache of types notBarry Lind
to be used, causing extra sql statements to be executed. This was a significant performance problem with the database meta data classes. The fix is a simple one liner.
2001-12-11Make sure that inlined S_UNLOCK is marked as an update of a 'volatile'Tom Lane
object. This should prevent the compiler from reordering loads and stores into or out of a critical section.
2001-12-11Repair roundoff-error problem for stddev/variance results near zero,Tom Lane
per complaint from Kemin Zhou. Fix lack of precision in numeric stddev/variance.
2001-12-10Suppress subquery pullup and pushdown when the subquery has anyTom Lane
set-returning functions in its target list. This ensures that we won't rewrite the query in a way that places set-returning functions into quals (WHERE clauses). Cf. bug reports from Joe Conway.
2001-12-10Declare LWLock pointers as volatile to prevent AIX compiler fromTom Lane
reordering operations at its whim. Releasing TAS lock before we've finished updating proc structure is uncool.
2001-12-10Update from Weiping HePeter Eisentraut
2001-12-10Update from Serguei MokhovPeter Eisentraut
2001-12-10AIX 5 needs same resultmap tweaks as AIX 4.Tom Lane
2001-12-10Repair case-conversion error in to_date's handling of roman numerals.Tom Lane
From Manuel Sugawara, approved by Karel Zak.
2001-12-10committed the missing filesMichael Meskes
2001-12-10Add French translationPeter Eisentraut
2001-12-10updatePeter Eisentraut
2001-12-10Remove duplicate lines from previous patch attempt. Trouble with myThomas G. Lockhart
DSL line at home broke things right in the middle of an update. :(
2001-12-10Support ODBC-style CURRENT_TIME, CURRENT_USER, etc with trailing emptyThomas G. Lockhart
parens. This is not SQL spec syntax, so later we will remove this extension from gram.y.
2001-12-09Fixed several bugs concerning indicators and added error messages instead of ↵Michael Meskes
segfaults.
2001-12-09Allow variable (unrestricted) precision for TIME and TIMESTAMP types in parser.Thomas G. Lockhart
2001-12-09Honor the typmod field for time zone intervals.Thomas G. Lockhart
2001-12-08Fix ecpg to allow pointer to structs.Michael Meskes
2001-12-06 - Removed debug message from preproc.y.Michael Meskes
- Fixed some bugs in exec sql var and exec sql type command.
2001-12-05 - Fixed variable handling in AT statement.Michael Meskes
- Fixed bug that caused segfault when given incorrect DB name. - Fixed bug in ecpglib causing indicator to list the size of the variable instead of the size of the data.
2001-12-05Fix for usage of spirntf in more portable way.Tatsuo Ishii
2001-12-05Re-enable SunOS4 port. ReplaceTatsuo Ishii
RTLD_LAZY | RTLD_GLOBAL to 1. It seems sunos4.h was accidentally modified between 7.1 and 7.2.
2001-12-04Enforce restriction that COPY DELIMITERS string must be exactly oneTom Lane
character; replace strchr() search with simple comparison to speed up COPY IN. Per discussion in pghackers.
2001-12-04Change appendStringInfoChar to appendStringInfoCharMacro in a couple ofTom Lane
hot spots --- buys about 10% in Verner's INSERT example.
2001-12-04Replace pq_getbytes(&ch, 1) calls with pq_getbyte(), which is easierTom Lane
to use and significantly faster. This tweak saves 25% (!) of the runtime of COPY IN in a test with 8000-character lines. I wouldn't normally commit a performance improvement this late in the cycle, but 25% got my attention...
2001-12-04Remove now-dead code for processing CONSTR_UNIQUE nodes inTom Lane
AlterTableAddConstraint. Earlier reorganization of the parser's processing of ALTER TABLE means that these node types no longer get here.
2001-12-04Hi Bruce, here is a first version of the hu.po file for the backend. IBruce Momjian
tested it with 7.2b3 and worked fine. Regards, Zoltan
2001-12-04Suppress data directory access-permissions check when __CYGWIN__,Tom Lane
per recent pghackers discussions. Improving this should be a TODO for 7.3.
2001-12-04 Fixed dumping of structs without indicators.Michael Meskes
2001-12-03Update to get around backslash problemsPeter Eisentraut
2001-12-03Fix latent bug in checking for failure of hashtable creation.Tom Lane
2001-12-03attachement is the NLS patch for backend/po & pg_dump,Bruce Momjian
they are all against the current CVS tree. the patch is about 70K. regards laser
2001-12-03Update IRIX geometry results to match newer MIPS FPUs, rather thanTom Lane
older ones. We really gotta get out of the business of matching every last low-order bit in this test ...
2001-12-03Add configure-time check whether sig_atomic_t is defined; if not,Tom Lane
provide a default definition equating it to 'int'. Should trigger only on machines with pre-ANSI-C header files, eg SunOS 4.1.x.
2001-12-03Pg_lo_open must OR together multiple modes, not AND them. Per reportTom Lane
from Andreas Kretzer.
2001-12-03Bump version to 3.3. Mostly this is because there is some confusion aboutD'Arcy J.M. Cain
the latest version and I wanted to make sure that there was a clean release. I also change the build files as I discussed in my letter of Nov 6, 2001. At the time I was asked to hold off until after the release.
2001-12-03Remove code to lookup WinSock error strings in netmsg.dll; according toTom Lane
Magnus Hagander that DLL only contains error strings for the Net*** functions, *not* WinSock. We need to look for a workable solution for older Windows flavors ... but it won't happen for PG 7.2.
2001-12-02HPUX 11 should use geometry-solaris-precision, not geometry-positive-zerosTom Lane
as is needed for HPUX 9 and 10.
2001-12-02More correct way to check for existence of types, which allows to specifyPeter Eisentraut
which include files to consider. Should fix BeOS problems with int8 types.
2001-12-02Update since left and right are no longer key words.Peter Eisentraut
2001-11-30Try to make genbki.sh a little more bulletproof, per trouble reportTom Lane
from Paul Stavrides.
2001-11-30Repair failure to mark an inserted Materialize node with the appropriateTom Lane
extParam/locParam lists. Per bug #526.
2001-11-30updatesPeter Eisentraut
2001-11-29Tweak plpgsql's expression reader to be smarter about parentheses andTom Lane
to give more useful error messages. Stephen Szabo's example of this morning ('loop' used as a variable name inside a subselect) works correctly now, and a FOR that is misinterpreted as an integer FOR will draw 'missing .. at end of SQL expression', which is at least marginally helpful.
2001-11-29Fix array_out's failure to backslash backslashes, per bug# 524. Also,Tom Lane
remove brain-dead rule that double quotes are needed if and only if the datatype is pass-by-reference; neither direction of the implication holds water. Instead, examine the actual data string to see if it contains any characters that force us to quote it. Add some documentation about quoting of array values, which was previously explained nowhere AFAICT.
2001-11-29NLS updates, most notably fixed zh_TW translationsPeter Eisentraut
2001-11-29Fix results to work with normal systems again.Peter Eisentraut
2001-11-28Load netmsg.dll locally in winsock_strerror, to avoid actual andTom Lane
potential problems discussed in pgsql-interfaces.
2001-11-28No longer need define test in md5.h because of Hiroshi's win_md5.c.Bruce Momjian