summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2000-01-18numeric_in accepts exponents; numeric to int4 rounds; float4/8 to numericTom Lane
is considerably more robust and accurate than it used to be. Also, get rid of numeric's private allocation freelist, which is no longer a win since Jan rewrote palloc.
2000-01-18Fix multibyte supportTatsuo Ishii
2000-01-18Adopt for new psqlTatsuo Ishii
2000-01-18Updated user interfaces on initdb, initlocation, pg_dump, ipcclean to a ↵Peter Eisentraut
GNU-compliant'ish state. Made ipcclean work on Linux.
2000-01-17setheapoverride() is history. Uses replaced with CommandCounterIncrement()Tom Lane
where necessary --- several of them didn't really need it, though. tqual-checking macros simplified accordingly.
2000-01-17Correct minor typos.Tom Lane
2000-01-17Update subquery error message.Bruce Momjian
2000-01-17Add .cvsignore so cvs update doesn't complain about derivedTom Lane
files being left around.
2000-01-17Modify libpq's pqexpbuffer to eliminate length restriction on how muchTom Lane
data can be formatted per call. This requires relying on vsnprintf(). On machines that haven't got vsnprintf, link in the version from backend/port/.
2000-01-17Pass atttypmod to CoerceTargetExpr, so that it can pass it on toTom Lane
coerce_type, so that the right things happen when coercing a previously- unknown constant to a destination data type.
2000-01-17Hmm, numeric array type was missing too. Added.Tom Lane
Of the standard types, only 'timestamp' seems not to have an array type; should it be added, or are we going to remove that type for 7.0 anyway?
2000-01-17Fix for TODO item * spinlock stuck problem when elog(FATAL)Hiroshi Inoue
and elog(ERROR) inside bufmgr.
2000-01-17Update unused_oids script so it works with non-GNU awk --- /* ... */Tom Lane
comment style apparently isn't portable to other awks.
2000-01-17Apparently, no one's ever used float4abs(), because it's got incorrectTom Lane
data in its pg_proc entry. abs() doesn't require two arguments, last I heard.
2000-01-17Update strings test to reflect the fact that casting to char() willTom Lane
now truncate or pad to the specified length.
2000-01-17Create a new parsetree node type, TypeCast, so that transformation ofTom Lane
SQL cast constructs can be performed during expression transformation instead of during parsing. This allows constructs like x::numeric(9,2) and x::int2::float8 to behave as one would expect.
2000-01-16Rearrange coding in COPY so that expansible string buffer for data beingTom Lane
read is reused for successive attributes, instead of being deleted and recreated from scratch for each value read in. This reduces palloc/pfree overhead a lot. COPY IN still seems to be noticeably slower than it was in 6.5 --- we need to figure out why. This change takes care of the only major performance loss I can see in copy.c itself, so the performance problem is at a lower level somewhere.
2000-01-16Sigh, I'm an idiot ... I broke the async startup logic a couple days ago,Tom Lane
by creating a race condition. It wasn't waiting for select() to say write-ready immediately after connect, which meant that you might get an unhelpful 'broken pipe' error message if connect failed, rather than the intended error message.
2000-01-16Put back change to 'connection failed' message formatting that someoneTom Lane
overwrote.
2000-01-16Removed lextest, because lex'ed files are now in the distribution.Peter Eisentraut
2000-01-16Included all yacc and lex files into the distribution.Peter Eisentraut
2000-01-16Add check that inherited constraints and defaults work.Tom Lane
2000-01-16Repair breakage of inherited constraint expressions --- needed aTom Lane
CommandCounterIncrement to make new relation visible before trying to parse/deparse the expressions. Also, eliminate unnecessary setheapoverride calls in AddNewAttributeTuples.
2000-01-16Fix broken FOR UPDATE error message.Tom Lane
2000-01-16Fix passing of atttypmod that Tom found.Bruce Momjian
2000-01-16Clean up pg_dump coredumps caused by change of output formatting forTom Lane
oidvector/int2vector. pg_dump code was assuming that it would see exactly FUNC_MAX_ARGS integers in the string returned by the backend. That's no longer true. (Perhaps that change wasn't such a good idea after all --- will it break any other applications??)
2000-01-15Add some examples to numeric regress test to verify that recently-fixedTom Lane
problems are indeed fixed.
2000-01-15Clean up problems with rounding/overflow code in NUMERIC, particularlyTom Lane
the case wherein zero was rejected for a field like NUMERIC(4,4). Miscellaneous other code beautification efforts.
2000-01-15Fix a passel of problems with incorrect calls to typinput and typoutputTom Lane
functions, which would lead to trouble with datatypes that paid attention to the typelem or typmod parameters to these functions. In particular, incorrect code in pg_aggregate.c explains the platform-specific failures that have been reported in NUMERIC avg().
2000-01-15Now that new psql is fflush()'ing properly, it emerges that severalTom Lane
regress test expected outputs were committed with NOTICEs appearing out of order. Update to correct results.
2000-01-15Update arrays regress test to reflect fact that several thingsTom Lane
work now that did not work in 6.5.
2000-01-15- Allow array on int8Peter Eisentraut
- Prevent permissions on indexes - Instituted --enable-multibyte option and tweaked the MB build process where necessary - initdb prompts for superuser password
2000-01-15Adapt to the changes of libpq(eliminateing using putenv()).Tatsuo Ishii
2000-01-15Eliminate using putenv().Tatsuo Ishii
2000-01-15Prepare for new psqlTatsuo Ishii
2000-01-15Fixed all elog related warnings, as well as a few others.Peter Eisentraut
2000-01-14Fixed psql variables vs array syntax, as well as minor psql enhancementsPeter Eisentraut
2000-01-14* User management commands no longer user pg_exec_query_dest -> more robustPeter Eisentraut
* Let unprivileged users change their own passwords. * The password is now an Sconst in the parser, which better reflects its text datatype and also forces users to quote them. * If your password is NULL you won't be written to the password file, meaning you can't connect until you have a password set up (if you use password authentication). * When you drop a user that owns a database you get an error. The database is not gone.
2000-01-14Clean up some problems in new asynchronous-connection logicTom Lane
in libpq --- mostly, poor response to error conditions. You now actually get to see the postmaster's 'The Data Base System is starting up' message, which you didn't before. I suspect the SSL code is still broken though.
2000-01-14Make PSQLexec's behavior on loss of connection more reasonable;Tom Lane
report original error before attempting reset, not after.
2000-01-14Make connection-failed messages a little friendlier onTom Lane
80-column displays...
2000-01-14Remove redundant and now-incorrect declaration of pstrdup.Tom Lane
2000-01-14Bump catversion to ensure initdb.Tom Lane
2000-01-14Revise quoting conventions in outfuncs/readfuncs so that nodeRead doesn'tTom Lane
choke on relation or attribute names containing spaces, quotes, or other special characters. This fixes a TODO item. It also forces initdb, since stored rule strings change.
2000-01-13Fixed everything in and surrounding createdb and dropdb to make it morePeter Eisentraut
error-proof. Rearranged some old code and removed dead sections.
2000-01-13initdb didn't load pg_descriptionPeter Eisentraut
2000-01-13Add UDC (User Defined Characters) support to SJIS/EUC_JP conversionTatsuo Ishii
Update README so that it reflects all source file names Add an entry to make sjistest (testing between SJIS/EUC_JP conversion)
2000-01-12Fixed a few "fixes" and bugs. Adjusted messages and options to GNU suggestions.Peter Eisentraut
2000-01-12Multi-byte case fix by Oliver Elphick (olly@lfix.co.uk)Tatsuo Ishii
2000-01-12Use fmgr_array_args() to avoid dependency on FUNC_MAX_ARGS.Tom Lane