PostgreSQL v7.0beta5 now available ...

After a failed attempt to get beta4 out the door, we've skipped it and put
out beta5, with a change log from beta3->beta5 included below ...

The *biggest* modifications since beta3 are the Buffer Manager changes
that Tom Lane recently put into place.

The key things remaining before release are docs related, but we still
need to get more ports tested and confirmed as being ready to go, so
please download and report your findings on this release.

ON Friday, April 14th, we hope to put out our first Release Candidate,
based on any reports received over the next 4 days.  If all goes well, and
pending documentation submit'd, maybe RC1 will be the only one ...

All bug reports should be submitted to pgsql-bugs@postgresql.org, and all
regression test results should be sent to pgsql-ports@postgresql.org ...

Thanks ...

Marc G. Fournier                   ICQ#7615664               IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy@hub.org           secondary: scrappy@{freebsd|postgresql}.org

===============

WinNT Changes:

   - I was unable to compile ecpg due to the ":=" instead of "=" in
     defining LIBPQDIR and some other variables in Makefile.global.in
   - pg_id (and also pg_encoding) executable was not removed during "make
     clean" - there was no $(X) appended to the executable name for rm
   - I have added result for int2, int4, float8 and geometry regression
     tests
        - int2, int2 - yet another message for too large numbers ;-)
        - float8 - it is problably a bug in the newlib C library - it has
          no error message for numbers with exponent -400
        - geometry - differences in precision of float numbers
   - I have added appropriate lines into resultmap file
   - I have modified the script regress.sh to use "case" statement when
     testing the hostname. For cygwin the script is called with
     "i686-pc-cygwin" (on my machine) as a parameter and this was not
     catched with the "if" statement.  The check was done for PORTNAME
     (win) and not HOSTNAME (i.86-pc-cygwin*).

- Add configure checks to see if 'using namespace std' and '#include
  <string>' work in the local C++ compiler.

- Added latest config.guess and config.sub

- Unixware fix for new config.guess output,

- Make configure --help produce slightly more useful and consistently-
  formatted descriptions of --with options.

- Adjusted for new (again!) config.guess output under QNX

- Fix configure.in. Remove SJIS, BIG5, WIN1250 from the selection list of
  multibyte encodings. These cannot be used as a backend/database encoding
  (OK to use as a client encoding).

- Add new pg_options.sample file.

- Get rid of SetBufferWriteMode(), which was an accident waiting to
  happen.

- Buffer manager modifications to keep a local buffer-dirtied bit as well
  as a shared dirtybit for each shared buffer.

- nodeAgg has always been willing to accept an aggregate with a finalFunc
  and only one transition state, but the CREATE AGGREGATE code rejected
  this combination.

- Fix bug noted by Bruce: FETCH in an already-aborted transaction block
  would crash, due to premature invocation of SetQuerySnapshot().

- Clean up problems with handling of multiple queries by splitting
  pg_parse_and_plan into two routines.

- absolute pathnames for the copy

- Fixed bug in createdb/alternative location

- Add transcendental math functions (sine, cosine, etc)

- Add a random number generator and seed setter (random(), SET SEED)

- Fix up the interval*float8 math to carry partial months
  into the time field.

- Add float8*interval so we have symmetry in the available math.

- Fix the parser and define.c to accept SQL92 types as field arguments.

- Fix the parser to accept SQL92 types for CREATE TYPE, etc. This is
  necessary to allow...

- Bit/varbit support in contrib/bit cleaned up to compile and load
  cleanly. Still needs some work before final release.

- Implement the "SOME" keyword as a synonym for "ANY" per SQL92.

- Implement ascii(text), ichar(int4), repeat(text,int4) to help
  support the ODBC driver.

- Enable the TRUNCATE() function mapping in the ODBC driver.

- Allow vacuum of temporary tables

- Repair assert failure in tuple-chain-moving logic

- Commebts & one check in EndEvalPlanQual().

- Fix (I hope) resource leakage in EvalPlanQual

- Partial fix for EvalPlanQual bugs reported by Magnus Hagander, 3-Apr.

- ExecSubPlan needs to be able to cope with RelabelType nodes atop the
  Const placeholder nodes for subplan result values.

- Add copyObject logic for TruncateStmt and a few other utility-statement
  parse node types.

- Fix extremely nasty little bug observed when a sub-SELECT appears in
  WHERE in a place where it can be part of a nestloop inner indexqual.

- outfuncs.c was missing a print routine for Material plan nodes, leading
  to trouble when trying to EXPLAIN VERBOSE a plan containing one.

- A little further tweaking of the range-query selectivity logic

- Further tweaking of indexscan cost estimates.

- Tweak indexscan cost estimation: round estimated # of tuples visited up
  to next integer.

- Save a few cycles in simple cases: no need to call cost_sort() when
  there is no presorted path to compare with.

- transformCreateStmt should put Ident nodes, not ColumnDef nodes, into
  keys lists of Constraint nodes.

- Also, change parser's uses of strcasecmp() to strcmp().

- New coding for SET provoked a 'var might be used uninitialized' warning
  from gcc.  Which wasn't actually a code bug, but I don't like warnings.

- Enable more flexible syntax for the SET command. Now allows single
  floats, single integers, and lists of names, without surrounding them
  with quotes.

- Remove all tokens which are defined as operators from ColID and ColLabel
  to avoid precedence confusion. Thanks to Tom Lane for catching this.

- Allow full type names in CREATE FUNCTION arguments and return type.

- Move CREATE FUNCTION/WITH clause to end of statement to get around
  shift/reduce conflicts with type names containing "WITH".

- Add lots of tokens as allowed ColId's and/or ColLabel's, so this should
  be a complete set for the v7.0 release.

- Remove no-longer-necessary restriction against uplevel correlation vars
  outside WHERE clause.

- Fix a couple of places that didn't handle uplevel refs cleanly.

- Tweak TypeCategory to treat new BIT types as of STRING category, rather
  than not knowing what they are at all.

- Hack parse_coerce so it won't try to constant-fold the dummy Const
  nodes introduced by make_subplan().

- When rewriting an aggregate introduced into WHERE, allow agg argument to
  be an expression not just a simple Var, so long as only one table is
  referenced (so that code isn't really any more difficult than before).

- Fix relcache refcount leakage when inv_drop is applied to a non-LO
  relation.

- Fix an Assert bug(was my fault) pointed out by Tom Lane.

- Add zpbit and varbit data types from Adrian Joubert

- Change rules for interpreting date/time input to disallow 1 and 3
  character years. Rejects dates like '0.085', which were accepted
  previously.

- Float-to-int conversion functions should return NULL when given NULL
  input, not throw a gratuitous elog().

- Update create_rule manual page.

- Rename bytea functions to not have upper-case letters in their names.

- Fix low-probability bug in relcache startup: write_irels wrote the
  pg_internal.init file in-place, which meant that if another backend
  started at about the same time, it might read the incomplete file.

- Add colo to display at end of initdb.

- Update initdb display of postmaster startup

- Fix syntax error reported by old shells ("if ! command..." is a
  neologism, apparently).

- Enhance pg_ctl so that it prints error messages from postmaster if it
  fails to start up it (this is only vaild if -w is given).

- Add a check to pg_dump to see whether backend is same version as
  pg_dump.  If not, abort by default.  Abort can be prevented by using -i
  or --ignore-version switch.

- Update pgaccess to 0.98.5

- Fix pgaccess __wish__ thing.

- Add new pgaccess languages.

- Fixed bug with repeated \e in psql (failed to clear buffers correctly)

- Old patch from Mark Hollomon to add plperl to createlang's repertoire.

- Add POLLUTE=1 to perl Makefile.PL creations.

- Makefile should not pre-empt user's decision about whether to use -g

- Fold PQsetenv working state into PGconn, rather than trying to maintain
  it in a separate object.

- remove PQsetenv routines from public API, until and unless they are
  redesigned to have a safer interface.

- Use the CXXFLAGS exported by configure (amazingly, we weren't before!)

- Make use of configure symbols for unportable constructs.  Make
  inclusions of system headers more consistent.

- Correct off-by-one error in strncat() usage.

- Change form of query which used "DISTINCT ON" to help support primary
  keys.

- Allow compile to finish even if plperl fails, which it does now.

- Remove bogus complexity from build/install of plperl.

- For no good reason at all, libplpgsql was getting linked with libpq.
  Since plpgsql is a backend extension, it doesn't need libpq...

- Remove -O2 for aix.

- Back out // compiler flag.

- Update alpha templates as suggested by SL Baur and Adriaan Joubert.

- Update make_mkid for mkid version 4.0.

- resultmap additions for solaris x86 ...

- Rename geometry.*bsdi to geo.*bsd. for Freebsd.

- Update regression for bsdi.

- Add expected output for netbsd, per report from Patrick Welche.

- Add resultmap entries for datetime-related tests on alpha-dec-osf.

- Tweak parallel test script so that command line options for postmaster
  can be set via environment variable PMOPTIONS.  Default is -o -F.

- Use initdb --noclean so that test database is available for examination
  if initdb fails.

- Update alternate float8 output files to match current float8.sql.

- New platform specific geometry regression expected file for
  powerpc-linux-gnulibc1.

- Update horology-solaris-1947.out per results from SL Baur.

- Update for to_char change.

- Update numeric_big regress test for 7.0.

- subselect regress test was kind of silly; it claimed to test correlation
  cases but actually did no such thing.  Make it test some more cases than
  before (including things that didn't work in 6.5).