Tom Lane [Mon, 26 Oct 1998 01:03:24 +0000 (01:03 +0000)]
Fix some actual bugs exposed by compiler warnings.
(Someone forgot whether their subroutine signaled errors by a NULL pointer
return value, or a negative integer... I'm surprised gcc -Wall doesn't
catch this...)
Tom Lane [Mon, 26 Oct 1998 01:00:13 +0000 (01:00 +0000)]
Get rid of some minor compiler warnings.
(HP's cc doesn't like if you forward-declare a routine static,
and then don't make it static in the actual definition...)
Tom Lane [Mon, 26 Oct 1998 00:56:16 +0000 (00:56 +0000)]
Fix cpp-from-stdin detection so that it works even if the
compiler silently accepts an empty command line.
Don't bother searching for bison and yacc unless we are going to use them.
Add massive new docs from Peter Mount.
Includes markup for the introductory portions suitable for the User's Guide
but just put the class reference material inside a huge <programlisting>
block. Figure out what to do later...
Tom Lane [Sat, 24 Oct 1998 22:21:12 +0000 (22:21 +0000)]
Update README to reflect the fact that you don't have to
run the regression tests with a special postmaster TZ setting anymore.
Correct a few typos too.
Tom Lane [Sat, 24 Oct 1998 22:05:01 +0000 (22:05 +0000)]
Update instructions for running configure (the old ones
seem to date from some pre-autoconf interactive configure script).
Revise sequence of instructions to allow you to start the regular
postmaster run before running regression tests; there's no need to
start and kill a special postmaster for regression tests.
Update the description of interpreting regression test results.
Arbitrary addition of a comment line to get ownership of file in the
cvs tree. Am trying to fix the permissions to allow file execution,
since I had damaged the permissions when transferring in with ftp.
Tom Lane [Mon, 19 Oct 1998 00:00:51 +0000 (00:00 +0000)]
Centralized shared-library build knowledge in a new file,
src/Makefile.shlib. Updated all the makefiles that try to build shlibs
to include that file instead of having duplicate (and mostly incomplete)
copies of shared-library options. It works on HPUX, a lot better than it
did before in fact, but there's a chance I broke some other platforms.
At least now you only have to fix one place not six...
Tom Lane [Sun, 18 Oct 1998 19:41:00 +0000 (19:41 +0000)]
Clean up shared makefile install procedures.
Get the permissions right, don't overwrite real files with symlinks, etc.
plpgsql and odbc still aren't fully up to speed, but at least they don't crash and burn...
Tom Lane [Sun, 18 Oct 1998 19:16:09 +0000 (19:16 +0000)]
Use AC_CONFIG_HEADER as documented by Autoconf manual: call it
only once at the top of the file. I'm surprised include/config.h wasn't
being overwritten from interfaces/odbc/config.h.in ...
Bruce Momjian [Sun, 18 Oct 1998 04:16:08 +0000 (04:16 +0000)]
The TCL/TK configuration cleanup patches I submitted have the following
problem:
'tclsh' still had to be found even if --with-libs (or
--with-libraries) was
specified to configure.
--with-libs is really an overloaded option. It really should only be used
to specify additions directories to search in order to file needed
libraries. It was also being used to locate the *Config.sh files.
Tom Lane [Sun, 18 Oct 1998 02:36:48 +0000 (02:36 +0000)]
Still another round of Perl-module installation tweaks.
Now, src/interfaces/perl5/Makefile.PL is pretty simple, and instead we
work a little harder in src/interfaces/Makefile.
Tom Lane [Sat, 17 Oct 1998 23:33:23 +0000 (23:33 +0000)]
Add code to pltcl's Makefile to cope with the situation where
libtcl has been installed as a non-shared library. pltcl cannot be
built in that situation; we want to do nothing and let the overall Postgres
build complete, rather than failing.
Bruce Momjian [Fri, 16 Oct 1998 06:05:18 +0000 (06:05 +0000)]
The problem is that read_pg_options needs DataDir to read its file but
DataDir is set after read_pg_options if postgres is called
interactively. If postgres is forked by postgres DataDir is read from
the PGDATA enviromnent variable set by the postmaster and this explains
while the bug disappears. I have written this patch but I don't like
it. Any better idea?
Bruce Momjian [Wed, 14 Oct 1998 17:59:22 +0000 (17:59 +0000)]
I have included a couple of patches that will make postgres
compile out of the tar file on Solaris with the SUN 5.0 compilers.
These compilers will be needed if you wan to compile the libpg++
interface without using the gcc/g++. The SC4.2 compilers do not
understand the string class.
The first patch changes the ecpg intermediate shared library
name from *.sho to *.sho.o so that the SUN compiler will
allow it to be used in conjunction with the -o option.
Add mention of the SERIAL data type.
Change claim of "no limit" on maximum date for datetime type to
more realistic 1465001 AD. Thanks Tatsuo for pointing this out :)
Clean up markup.
Chapter on indices intended for the User's Guide.
Currently not included in the UG, and this now only has a discussion of
partial indices by Paul Aoki culled from the mailing lists.
But, didn't want to loose it...
Clean up library names and installation paths.
Include soft links in the installation directory for major version number.
Use the existing $(DLSUFFIX) parameter to name shared library.
Add Taral's test for flags required by the detected cpp program to read
from stdin. Necessary at least for systems missing cpp but having gcc -E.
Include new configured file backend/catalog/genbki.sh.
Allow TIMESTAMP as a column name or general identifier.
timestamp had become a token a while ago, but had been omitted from the
keywords.c until recently. This uncovered the omission in the ColId decl.
projects / postgresql / log