c4d6bda2c5
There is a bug in aclinsert3 in the code which update the acl arrays.
...
When an acl item is added or updated the new entry is deleted if it has no
permissions and the acl array is shrinked. This is is done by decrementing
the number of items without updating the corresponding array size.
The array with the incorrect size is later read by pg_aclcheck and the entry
count is used to allocate a new array while the array size is used to copy
the old one. This causes a memory corruption and a backend crash.
This happens only to normal user as the administrator bypasses acl checks.
Massimo Dal Zotto
1996-11-20 22:53:10 +00:00
2abbf392bd
Removed old Assert's used during development.
1996-11-19 05:06:39 +00:00
13312a007d
The routines in magic.c have moved to the more accessible version.c.
1996-11-19 02:55:37 +00:00
e2a8163c40
Remove linux define of __USE_POSIX, which doesn't appear to do anything.
1996-11-18 02:26:57 +00:00
b55d49f5fe
Add comments about the diversity of signal functionality. Came from
...
Makefile.global.
1996-11-18 02:25:09 +00:00
87f5fda6b9
Make port-specific link libraries defined for linking backend more global
...
so you can also link pgtclsh.
1996-11-18 02:23:41 +00:00
bd57c3afe5
Quiet compiler warnings about missing prototypes in Linux's bitops.h.
1996-11-18 01:45:44 +00:00
003b76f9ac
Fix for error when viewing group access privs.
1996-11-17 04:26:59 +00:00
4b130b2209
Fix for Alter TABLE add column varchar(). Was causing zero length.
1996-11-17 04:23:10 +00:00
aef6163388
Remove old GetPGData reference that causes warning.
1996-11-16 09:03:44 +00:00
7bf3bae1a4
Add #include <sys/types.h> so it works with Ultrix's in.h
1996-11-16 08:09:16 +00:00
7085aba53c
Remove quote removal. Moved to parser.
1996-11-15 18:38:55 +00:00
36aaf2505d
Correct permissions on lo_export(). Open up permissions.
1996-11-15 18:38:20 +00:00
3f53311088
Fix quote/doublequote handling.
1996-11-15 18:37:37 +00:00
b847d29785
Remove PERFECT_MEM
...
Unallocate opaque.
1996-11-15 18:37:10 +00:00
f8e02c192a
Unallocate opaque.
1996-11-15 18:36:31 +00:00
54c9905d6e
Mulitple inclusion of sys/socket.h
...
Pointed out by: Erik Bertelsen <erik@sockdev.uni-c.dk>
1996-11-15 09:54:28 +00:00
8abc490181
Features added:
...
* Wrote max(date) and min(date) aggregates
* Wrote operator "-" for date; date - date yields number of days
difference
* Wrote operator+(date,int) and operator-(date,int); the int is the
number of days. Each operator returns a new date.
By: Tom Tromey <tromey@creche.cygnus.com>
1996-11-14 21:39:14 +00:00
eedc75bfe1
Bring in Randy's NOROOT patch
1996-11-14 20:49:09 +00:00
2081153e69
Fix a comment...was going to fix the <sys/types.h> problem, but
...
Randy is submitting a more comprehensive patch :)
1996-11-14 20:27:06 +00:00
d84057db71
Couple of Ultrix4 related patches submited by:
...
Erik Bertelsen <erik@sockdev.uni-c.dk>
1996-11-14 20:08:58 +00:00
f64b840387
Remove most compile-time options, add a few runtime options to make up for it.
...
In particular, no more compiled-in default for PGDATA or LIBDIR. Commands
that need them need either invocation options or environment variables.
PGPORT default is hardcoded as 5432, but overrideable with options or
environment variables.
1996-11-14 10:25:54 +00:00
6b135c9391
Use port-protos.h instead of trying to declare random() and srandom()
...
explicitly.
1996-11-14 08:23:31 +00:00
902d2b8bac
Use -Wno-error option only if compiler is gcc.
1996-11-14 07:33:34 +00:00
07a65b2255
Commit of a *MAJOR* patch from Dan McGuirk <djm@indirect.com>
...
Changes:
* Unique index capability works using the syntax 'create unique
index'.
* Duplicate OID's in the system tables are removed. I put
little scripts called 'duplicate_oids' and 'find_oid' in
include/catalog that help to find and remove duplicate OID's.
I also moved 'unused_oids' from backend/catalog to
include/catalog, since it has to be in the same directory
as the include files in order to work.
* The backend tries converting the name of a function or aggregate
to all lowercase if the original name given doesn't work (mostly
for compatibility with ODBC).
* You can 'SELECT NULL' to your heart's content.
* I put my _bt_updateitem fix in instead, which uses
_bt_insertonpg so that even if the new key is so big that
the page has to be split, everything still works.
* All literal references to system catalog OID's have been
replaced with references to define'd constants from the catalog
header files.
* I added a couple of node copy functions. I think this was a
preliminary attempt to get rules to work.
1996-11-13 20:56:15 +00:00
17befd6c69
Use new utils/version.c instead of backend/utils/init/magic.c.
1996-11-12 06:47:10 +00:00
a157098f7c
Fixed double pfree on opaque.
1996-11-11 14:02:10 +00:00
2bdded3540
Various cleanups from D'Arcy
1996-11-11 12:14:45 +00:00
6913c8b4a4
Add an abort() call to ExceptionalCondition so that is Assert is
...
called, it dumps core...
ABORT_ON_ASSERT must be defined, as I don't know if this is the correct
way to do this...
1996-11-11 11:49:40 +00:00
9fc64aefe1
change include c.h to postgres.h
1996-11-11 11:29:15 +00:00
13d44b4aac
Warning removal cleanup
1996-11-11 04:54:54 +00:00
60915ed2c2
Change gmake to $(MAKE) so backend compiles on system with no "gmake".
1996-11-10 04:09:51 +00:00
77f8dff617
prototype cleanup
1996-11-10 03:30:46 +00:00
08029facb3
Recognize dependencies more reliably.
1996-11-10 03:13:59 +00:00
aaeef4d17d
All external function definitions now have prototypes that are checked.
1996-11-10 03:06:38 +00:00
bf5cbbf789
Quiet compiler warnings.
1996-11-10 02:27:15 +00:00
9f1b6bf7db
Add #include <unistd.h> to quiet compiler warning about missing getopt() declaratoin.
1996-11-10 01:37:48 +00:00
0e5ab3655c
Remove #include <regex.h> so it compiles on systems with GNU regex library.
1996-11-10 01:20:44 +00:00
d7806dc5ee
Make it find inet_aton.h so backend compiles on Solaris.
1996-11-10 00:38:38 +00:00
a0990e1884
Makefile cleanup after reorganization
1996-11-09 06:24:51 +00:00
c9c0e111b8
More compile cleanups
1996-11-08 20:46:33 +00:00
7886d7351e
Fix perbyte_cpu thing
1996-11-08 06:24:58 +00:00
4b2b8592a0
Compile and warning cleanup
1996-11-08 06:02:30 +00:00
ad6ca0c3af
Another patch to quiet the compiler from D'Arcy
1996-11-08 00:59:10 +00:00
7d8b794a9a
Add some missing includes
...
From: Randy Terbush <randy@zyzzyva.com>
1996-11-08 00:56:17 +00:00
b38d565fc3
D'Arcy's recent cleanups
1996-11-08 00:46:14 +00:00
0020e8790d
Another directory that compiles with no errors, and few warnings
1996-11-06 10:32:10 +00:00
b6e640f829
Make sure that #include <postgres.h> is included in fmgrtab.c
1996-11-06 10:22:13 +00:00
58a0150f64
More files cleaned out
1996-11-06 09:29:26 +00:00
83d93d437b
Just a couple of small mods
1996-11-06 08:54:17 +00:00
a472a29bb4
I'm getting there, slowly :)
1996-11-06 08:48:33 +00:00
e232909157
Slowly getting faster to make sure it compiles...still needs cleaning
...
up, but one thing at a time :)
1996-11-06 08:27:16 +00:00
7385619f14
Clean it up so that it compiles
1996-11-06 08:21:43 +00:00
9e3d4c43a7
Another directory cleaned up
1996-11-06 07:31:26 +00:00
ce4c0ce1de
Some compile failure fixes from Keith Parks <emkxp01@mtcc.demon.co.uk>
1996-11-06 06:52:23 +00:00
e4725ee76a
Makes it compile...still needs work
1996-11-05 11:57:55 +00:00
225a224792
Missed a file...
...
access now compile with -Wall -Werror
1996-11-05 11:17:47 +00:00
03eb5a50e7
There, that cleans *that* out
1996-11-05 11:12:33 +00:00
050d148295
One more to go
1996-11-05 10:54:20 +00:00
500ce42f37
Almost done access...
1996-11-05 10:35:38 +00:00
4fa90e38d5
More cleanups
1996-11-05 10:02:06 +00:00
466f5fd729
More cleanups
1996-11-05 09:53:03 +00:00
4312192e6b
another directory cleaned of extraneous #includes
...
makefile fixed for fmgr.h
1996-11-05 09:40:25 +00:00
0cf60fb718
More cleaned out #include files
...
Makefile fixed for fmgr.h
1996-11-05 08:19:01 +00:00
1c3a7767ca
Re-add -I../.. for fmgr.h
...
Change #include "" to #include <>
Remove a few unused #includes
Make sure it compiles with -Wall -Werror
1996-11-05 07:42:46 +00:00
fd569a2308
Clean up a few of the #include files
1996-11-05 05:26:32 +00:00
48a7170a71
Added missing #includes pointed out by Keith Parks
1996-11-05 05:25:44 +00:00
18bbad7696
Remove OPENLINK define
1996-11-04 04:53:51 +00:00
0108fddf13
Remove NULL_PATCH define
1996-11-04 04:35:48 +00:00
8edbc3bde4
Remove ARRAY_PATCH define
1996-11-04 04:19:55 +00:00
e43dfad289
Remove ESCAPE_PATCH define
1996-11-04 04:05:16 +00:00
7e42f4814d
Removed PORTNAME_. Use OS version only.
1996-11-04 04:00:56 +00:00
510ebf75a1
Another first pass at cleaning up the #includes
1996-11-03 23:57:43 +00:00
d2aa40517a
Cleaned up include headers
1996-11-03 23:46:28 +00:00
4029fb52c6
first pass at cleaning up include files
1996-11-03 23:27:08 +00:00
ca112038ab
More include file cleanups
1996-11-03 22:58:26 +00:00
302b809793
Changes required so that access/* compiles cleanly...
...
cleaning up behind myself before *yawn* bed :)
1996-11-03 12:35:27 +00:00
71cd646a34
More include file cleanups
1996-11-03 12:11:19 +00:00
dd71c36933
That pretty much totally cleans up the includes files here.
1996-11-03 10:57:24 +00:00
5a58dabbbc
Include Makefile.global so fmgrtab.c compiles properly. Thanks Keith Parks.
1996-11-03 09:28:31 +00:00
8ac872b26d
Fix bogus format in two error messages.
1996-11-03 09:24:04 +00:00
66a518735f
Finish cleaning up -I lines
1996-11-03 09:07:03 +00:00
a7a0b3ba30
Keep the generated .h files in the local directory, like other object files,
...
rather than in src/include with normal header files. Necessary to make
dependancies work out so they don't keep getting rebuilt for no reason.
1996-11-03 09:05:30 +00:00
0c960e7fc0
More include file cleanups
1996-11-03 08:17:38 +00:00
aed384f12d
Quiet a compiler warning (see line 729+, retvalue)
1996-11-03 07:00:57 +00:00
dd5b78f1c1
Remove 'unused variable `open_errno''
1996-11-03 06:56:52 +00:00
c9002ecb21
Produce a clean compile of backend...
1996-11-03 06:54:38 +00:00
6fac4602ae
Clean up Makefile(s)
...
Add #include "postgres.h" as required
Remove #include "c.h" *and* "utils/elog.h" as applicable
1996-11-03 05:08:01 +00:00
66637f4a2f
Clean up Makefile
...
Add #include "postgres.h" as required
Remove #include "utils/elog.h"
1996-11-03 04:57:03 +00:00
c4edfb4667
Clean up Makefile, add #postgres.h as required
1996-11-03 04:51:51 +00:00
daa93cf5c5
Clean up Makefile, make sure that postmaster.c compiles
1996-11-03 04:48:27 +00:00
763adb5235
Convenience routine for checking superuser status.
1996-11-02 02:06:47 +00:00
675740a8f3
Add misc subdirectory.
1996-11-02 02:03:50 +00:00
fa608ad2a6
Fix bug in checking permissions on table being COPY'd.
1996-11-02 02:03:13 +00:00
d27c28fdc0
Allow only superuser to do backend copy.
1996-11-02 02:01:48 +00:00
a95674269f
More intelligent #include cleanups, as pointed out by Bryan...
...
Compiled with -Wall -Werror
1996-11-01 09:41:41 +00:00
ea94d639d0
Okay, so I was going backwards on my include cleanup...
...
Pointed out by Bryan (in a subtle way *grin*)
1996-11-01 09:12:18 +00:00
dd10d0de23
More chanes for bsdi_2_1 removal
1996-11-01 03:36:16 +00:00
3a4c2888a1
Removal of bsdi_2_1 port.
1996-11-01 01:47:23 +00:00
3ce9ccff7c
Fix for gmake clean.
1996-10-31 17:27:58 +00:00
b17c49e95d
Add #include "postgres.h"
1996-10-31 11:09:44 +00:00
76294e1d38
Cleanup Makefiles
...
Add #include "postgres.h"
and
#include <sys/types.h>
1996-10-31 10:59:42 +00:00
eb501f1cd3
More Makefile cleanups
...
Add #include "postgres.h" as required
1996-10-31 10:42:56 +00:00
ae1d931e88
clean up makefile
...
add #include "postgres.h"
1996-10-31 10:37:53 +00:00
ffaff3dcd4
Clean out makefile
...
add #include "postgres.h"
1996-10-31 10:26:33 +00:00
df6a9e633c
more removal of PORTNAME_*
1996-10-31 10:23:28 +00:00
4df1a41478
more removals of PORTNAME_*
1996-10-31 10:20:09 +00:00
bdd07411b4
remove use of PORTNAME_*
1996-10-31 10:17:09 +00:00
3df33180a1
add #include "postgres.h", as required by all .c files
1996-10-31 10:12:26 +00:00
3341750769
Require superuser privilege to create C function.
1996-10-31 09:08:10 +00:00
b13f5c25e2
More of the same...clean Makefile, add include files
1996-10-31 08:52:54 +00:00
ab2444032a
Same clean up of Makefile, and addition of include files
1996-10-31 08:32:23 +00:00
f578f8fb68
Cleaned out -I../.. -I../../include from Makefile
...
Added missing include files
1996-10-31 08:28:53 +00:00
367ade186f
Remove -I../.. and -I../../include from Makefile
...
Add some missing include files
1996-10-31 08:24:47 +00:00
ea04818002
Add a couple of missing includes
1996-10-31 08:09:47 +00:00
ba4ad2c71a
Added misc include files
1996-10-31 07:48:38 +00:00
14352a3ef6
.h files shouldn't include .h files, .c files should contain .h files...
...
remove postgres.h from fmgr.h creation
1996-10-31 07:37:57 +00:00
081fd70d92
Even compile generated .h files should be in ${SRCDIR}/include...
1996-10-31 07:25:15 +00:00
c59b3afbf1
Well, normally there wouldn't be anything in -I../.., and in a minute, there
...
won't be...
1996-10-31 07:22:07 +00:00
0a761375e9
Move these files to ${SRCDIR}/include
1996-10-31 07:09:33 +00:00
5565d32e10
There is no -I../.. include files...
1996-10-31 07:04:50 +00:00
3d7fe6c97c
Added needed include file.
1996-10-31 05:58:01 +00:00
34945055a8
Changed make to gmake.
...
Added needed include file.
1996-10-31 05:55:24 +00:00
4cf9d03ec7
Added prototypes missing from parser patch.
...
Added needed include file.
1996-10-31 05:54:27 +00:00
f1c95ed41e
Fix memset() call, variables being passed in wrong order.
...
Pointed out by: wieck@sapserv.debis.de
1996-10-30 21:18:22 +00:00
69c7f25bed
Fixes:
...
I found another bug in btree index. Looking at the code it seems that NULL
keys are never used to build or scan a btree index (see the explain commands
in the example). However this is not the case when a null key is retrieved
in an outer loop of a join select and used in an index scan of an inner loop.
This bug causes at least three kinds of problems:
1) the backend crashes when it tries to compare a text string with a null.
2) it is not possible to find tuples with null keys in a join.
3) null is considered equal to 0 when the datum is passed by value, see
the last query.
Submitted by: Massimo Dal Zotto <dz@cs.unitn.it>
1996-10-30 06:08:10 +00:00
f59a46a8c8
Parser Overhaul
1996-10-30 02:02:41 +00:00
1dfe4eaeb1
Fix small syntax error.
1996-10-30 00:36:59 +00:00
71c1d59d1a
Remove another snprintf() for Solaris.
1996-10-29 21:51:13 +00:00
63481f0b92
Use strncpy() and local buffers instead of snprintf(), since not everyone
...
has snprintf().
1996-10-28 09:05:29 +00:00
aa1eac7979
Make a local isblank() function, since not all standard C libraries have it.
...
For sparc_solaris, add inet_aton prototype, since it isn't in the header
files included for the standard C library functions.
1996-10-28 09:03:50 +00:00
b13ffd34b3
Add our own copy of inet_aton() for sparc_solaris, which doesn't have it in
...
the standard C library.
1996-10-28 09:00:54 +00:00
b0d6f0aa63
Simplify make files, add full dependencies.
1996-10-27 09:55:05 +00:00
1e39d14ff3
Correct #if statement so it compiles on Solaris. Thanks Keith Parks.
1996-10-26 05:03:24 +00:00
87b48ff032
D'Arcy's cleanups
1996-10-26 04:15:05 +00:00
716d506a81
Make sure the btree patch gets into 2.0 as well...
...
Still submitted by: Massimo Dal Zotto <dz@cs.unitn.it>
1996-10-25 09:55:36 +00:00
ec86e4b95b
Use EUROPEAN_DATES instead of EUROPEAN_STYLE
1996-10-25 06:02:52 +00:00
3010f46d2a
E Use EUROPEAN_DATES from config.h instead of -D option
1996-10-25 06:00:49 +00:00
f6f8ca4970
Fixes: Growing backend when using nested function calls
...
Submitted by: wieck@sapserv.debis.de (Jan Wieck)
1996-10-24 07:59:46 +00:00
20b4c46e26
Fixes: Errors when PQexec() in backend creates temp
...
relations and transaction is aborted
Submitted by: wieck@sapserv.debis.de (Jan Wieck)
1996-10-24 07:55:54 +00:00
e152661200
Fixes:
...
It's bug in nodeAgg.c on lines 241, 242:
null_array = malloc(nagg);
for (i=0;i<nagg;i++)
null_array[i] = 'n';
oneTuple = heap_formtuple(tupType, tupValue, null_array);
- your query has not only aggregates but also 'group by-ed' fields and so
null_array should contain tupType->natts elements (tupType->natts > nagg in
your case).
Patch follows and it's very simple.
VAdim
1996-10-24 06:32:01 +00:00
c471d2bdeb
Take out the PERFECT_MMGR #ifdefs:
...
My guess is that the thing had bugs, and the pfree was commented out.
The thing is probabally free'ed anyway at the end, so it was not a bad
thing.
If it does cause a bug, it will generate an error when hit, so I say
unless someone else knows, let's remove it and run the regression test.
-Bruce
1996-10-24 06:30:40 +00:00
f36b2560a4
Major code cleanups from D'arcy (-Wall -Werror)
1996-10-23 07:42:13 +00:00
2db54b8e4f
Update "planner" directory to "optimizer" in include paths in this dead code.
1996-10-23 07:14:43 +00:00
82b83f6f71
Cosmetic changes to ordering of #include files
1996-10-21 11:49:41 +00:00
9a7fda57ee
-Wall'd
1996-10-21 09:37:26 +00:00
aa907f7d0c
-Wall'd
1996-10-21 08:31:23 +00:00
8385d557d1
-Wall'd ... I missed a subdir under access :(
1996-10-21 07:38:20 +00:00
f47ffc9a4e
Make line 159:bp = NULL, to quiet compiler
...
Suggested by: Randy Terbush <randy@zyzzyva.com>
1996-10-21 07:18:08 +00:00
c9f8bad182
-Wall'd
...
That finishes access/*
1996-10-21 07:15:18 +00:00
b446f38510
-Wall'd
1996-10-21 05:59:49 +00:00
ede7e6b045
-Wall'd
1996-10-21 05:45:21 +00:00
0452227989
-Wall Cleanup of gist subdirectory
1996-10-21 05:14:02 +00:00
95874fa034
-Wall cleanup of gist subdirectory:
...
- cleaned out unused variables
- added missing prototype headers
One outstanding -Werror failure:
/home/staff/scrappy/postgres/2.0/cvs/postgres95/src/backend/access/gist/giststra
t.c:117: warning: missing braces around initializer for `GISTEvaluationData.expr
ession'
1996-10-21 05:11:00 +00:00
d458a1c747
Other then:
...
indextuple.c:159: warning: `bp' might be used uninitialized in this function
this directory passes -Wall -Werror under FreeBSD
1996-10-20 22:04:49 +00:00
13cd836d62
#include file cleanup
1996-10-20 10:53:18 +00:00
e1220b7c21
More #include cleanups
...
Once access/* is cleaned out, will redo using -Wall on compile to
make sure that all prototyping is correct
1996-10-20 09:27:24 +00:00
5a0b450c78
First pass at fixing my own mistakes
...
Mainly...fix up the includes I removed, as well as prototypes
Pointed out by D'Arcy
1996-10-20 08:32:11 +00:00
dd007d4c39
Another directory cleaned out
1996-10-20 06:56:04 +00:00
3cb7ff8db2
Okay...another directory cleaned out and compilable
1996-10-20 06:34:30 +00:00
5eee45ea05
Add HAVE_Cplusplus= true to BSD44_derived Makefile.inc, as I know that
...
FreeBSD/NetBSD both ship with gcc 2.7.x
1996-10-19 06:36:52 +00:00
0eac5b0ed5
Welp, another subdirectory cleaned out of redundant/unused #include
...
files
1996-10-19 06:27:33 +00:00
e77f64af62
There...that pretty much cleans up redundant/unused #includes in
...
access/common...how many more directories to go? :)
1996-10-19 04:51:44 +00:00
d7dd05db55
Another cleaned up
1996-10-19 03:39:04 +00:00
5dedad99a1
One at a time...another one is complete...
1996-10-19 03:28:20 +00:00
69b42eb1a3
another one with #include file clean'd up
1996-10-19 03:12:50 +00:00
9834913cd0
All #include's removed from *.h files, so cleaning up the .c #includes...
...
First file of, what...1000's?
1996-10-18 19:01:16 +00:00
9caf84bc00
#include "postgres.h" is a standard include for all .c files, auto-including
...
config.h, c.h *and* elog.h...added elog.h
1996-10-18 18:28:33 +00:00
bc50067426
1. Reap deleted/unused tuples on page-base
...
2. Reap unused tuples too
3. Reap empty pages
4. Check if a page is initialized, initialize it if not
and reap it
5. Binary search in list of reapped pages/tids to check
is the heap' tid pointed by a index' tuple on this list
(it's mu-u-uch faster)
1996-10-18 08:13:36 +00:00
327d900b51
Move #define PageIsNew (true if page is not initialize) to bufpage.h
1996-10-18 07:43:43 +00:00
94e825145d
Document more #ifdef's into config.h
...
Get rid of ESCAPE_PATCH ifdef, as its on by default, and there is no
apparent reason for turning it off...it fixes a bug
1996-10-18 05:59:17 +00:00
1c00e68fad
Moved definition of MAXPATHLEN to config.h
1996-10-18 05:50:04 +00:00
927e5887a0
Added a #define NEED_SIG_JMP to config.h
...
Removed #ifdef's for OPENLINK_PATCH. We enable it by default, and nobody
*has* to enable the functionality, but no reason to retain the "old code"
1996-10-18 05:47:12 +00:00
2ecb7e5e2c
Moved '#define *DEBUG' defines to config.h, and document where first
...
found.
Document any '#ifdef' segments found in config.h
1996-10-18 05:21:27 +00:00
6e420677dd
Moved '#define *DEBUG' defines to config.h, and document where first
...
found.
Document any '#ifdef' segments found in config.h
Remove gistold.c, since it was older copy of gist.c
1996-10-18 05:21:09 +00:00
6c8465f69b
#define cleanup: moved #define IPORTAL_DEBUG to config.h
1996-10-18 04:54:47 +00:00
2d8ffead24
Moved all the CFLAG -D's from the makefiles to config.h towards better
...
centralized configurations/porting...
1996-10-18 04:49:09 +00:00
d16bc9ce37
Centralize more of the defines from backend/port under config.h
...
Further ports should *really* avoid backend/port and make use of
config.h
1996-10-18 00:33:41 +00:00
a839456c41
Move "port specific" #ifdefs out of here and into config.h
...
- created a HAVE_TZSET define for this purpose
1996-10-17 23:59:45 +00:00
83cb729887
More closing of relations left open by parser.
1996-10-14 03:53:53 +00:00
81f2f81960
Removed const warning.
1996-10-13 18:39:41 +00:00
54612f5141
D'Arcy change.
1996-10-13 18:38:51 +00:00
baf9a6e24c
Added include needed for recent addition.
...
I hope everyone has sys/socket.h.
1996-10-13 18:38:04 +00:00
e6cacf9211
Added include needed for recent addition.
1996-10-13 18:37:19 +00:00
86be8677a9
Added consistent include file handling for MAXHOSTNAMELEN.
...
If I have introduced any syntax errors with this, I will patch them in
the morning, and Marc will have a good laugh.
1996-10-13 04:50:27 +00:00
abb1b3e770
I checked the alter table code, and started suspecting the relation
...
cache. I found if I manually added a line to flush the whole relation
cache, the assert error disappeared. Looking through the code, I found
that the relation cache is flushed at the end of each query if the
reference count is zero for the relation. However, printf's showed that
the rd_relcnt(reference count) for the accessed query was not returning
to zero after each query.
It turns out the parser was doing a heap_ropen in parser/analyze.c to
get information about the table's columns, but was not doing a
heap_close.
This was causing the query after the ALTER TABLE ADD to see the old
table structure, and the executor's assert was reporting the problem.
1996-10-13 04:26:39 +00:00
bef3c89a1c
Fix prototypes so postmaster.c will compile.
1996-10-13 04:01:05 +00:00
785234d6ca
New host-based authentication -- send error message when authentication fails
1996-10-12 07:48:49 +00:00
4b5c977782
New host-based authentication with ident
1996-10-12 07:47:12 +00:00
57026d6009
New host-based authentication -- install pg_hba.conf instead of pg_hba
1996-10-12 07:44:39 +00:00
09bb369d23
Separate general purpose functions from portal functions so they may be
...
used in test drivers.
1996-10-11 09:47:14 +00:00
81cda65f74
Add a MIPS/NetBSD port...
1996-10-11 03:56:41 +00:00
b7559f94cd
This change should have no practical effect but it is the more
...
correct way to do this. Theoretically you could have a NULL
pointer that isn't represented internally as all 0 bits. This
guarantees that it convert correctly.
Submitted by: darcy@druid.com (D'Arcy J.M. Cain)
1996-10-11 03:25:00 +00:00
2663dfd94e
I have written some patches to the postgres lock manager which allow the
...
use of long term cooperative locks managed by the user applications.
Submitted by: Massimo Dal Zotto <dz@cs.unitn.it>
1996-10-11 03:22:59 +00:00
ca5f6dba0f
Comment cleanup.
1996-10-10 00:18:52 +00:00
a409f40ea8
Modified Assert to be more selective.
1996-10-07 19:53:15 +00:00
3dc0c8e4e5
Fixed a missed change from Carsten Heyl <heyl@nads.de>
1996-10-07 07:18:34 +00:00
de466eb8f4
Mostly adding "const" keyword and making some functions static.
...
Submitted by: D'Arcy Cain
1996-10-07 03:30:40 +00:00
257b4d090c
Change new assert so it generates assert message rather than SIGSEG.
1996-10-07 02:50:19 +00:00
47279960a0
Fix assert to allow zero. OK'ed by Bryan.
1996-10-05 20:31:38 +00:00
a433f22c24
Fix args to match prototype.
1996-10-05 20:30:31 +00:00
4feb696be7
Here the fix for the first assertion failure I had which killed
...
my postmaster 1.07.
It's really simple, the loop dealing with all sockets
can't handle more than one ready socket :-)
A simple logic error dealing with lists.
OR IS THERE ANY REASON FOR SETTING curr TO 0?
Submitted by: Carsten Heyl <Heyl@nads.de>
1996-10-04 20:32:07 +00:00
f8009e1786
The remainder of D'Arcy's changes, most notibly the usage of SIGNAL_ARGS
1996-10-04 20:17:11 +00:00
c61d606a15
Added dlfcn.h to dl.c in BSD44_derived port
1996-10-03 22:41:42 +00:00
aefb294e13
Added restriction to tables vacuum'able.
...
Added NOTICE messages for bad table names.
1996-10-03 20:11:41 +00:00
f5316239b8
Moved them.
1996-10-03 17:27:01 +00:00
267ef0a642
Allow vacuum to run for only one table. Fixes allocation bug.
1996-10-03 04:19:29 +00:00
8f7756d796
palloc() finfo memory. From Kurt.
1996-09-26 16:29:05 +00:00
adc98c0b35
Fix cluster finfo problem Kurt reported.
1996-09-26 15:43:35 +00:00
2baf2150e9
Fixed YACC depenencies and 'make clean' now cleans totally.
1996-09-26 15:40:45 +00:00
f8a84c947a
This should cause 'make clean' to clean up bki files. Looks like a typo.
1996-09-26 04:22:49 +00:00
deaaecdf2e
Prevents possible packet loss from SIGCHLD.
1996-09-26 03:17:44 +00:00
0d83b86720
Changed gawk to awk.
1996-09-25 19:24:58 +00:00
9c6135fb15
From: Bruce Momjian <maillist@candle.pha.pa.us>
...
To: Postgres95-development <pg95-dev@ki.net>
Subject: [PG95-DEV] postgres.h patch
This removes the parameters from the func_ptr prototype in postgres.h in
2.0.
1996-09-23 08:38:45 +00:00
d00f621d7b
diff -ru ../../../orig/postgres95/src/backend/utils/adt/varlena.c
...
./utils/adt/varlena.c
- --- ../../../orig/postgres95/src/backend/utils/adt/varlena.c Mon Jul 22
23:56:04 1996
+++ ./utils/adt/varlena.c Tue Sep 17 15:12:55 1996
@@ -166,7 +166,7 @@
if (inputText == NULL)
return(NULL);
- - len = strlen(inputText) + VARHDRSZ;
+ len = strlen(inputText) + VARHDRSZ + 1 /* terminating 0 */;
result = (struct varlena *) palloc(len);
VARSIZE(result) = len;
memmove(VARDATA(result), inputText, len - VARHDRSZ);
Submitted by: skimo@breughel.ufsia.ac.be (Sven Verdoolaege)
1996-09-23 08:31:01 +00:00
fb7f8e7655
diff -ru ../../../orig/postgres95/src/backend/libpq/be-fsstubs.c
...
./libpq/be-fsstubs.c
- --- ../../../orig/postgres95/src/backend/libpq/be-fsstubs.c Tue Jul 9
08:21:30 1996
+++ ./libpq/be-fsstubs.c Tue Sep 17 15:30:09 1996
@@ -241,7 +241,7 @@
*/
fd = open(VARDATA(filename), O_RDONLY, 0666);
if (fd < 0) { /* error */
- - elog(WARN, "lo_import: can't open unix file\"%s\"\n", filename);
+ elog(WARN, "lo_import: can't open unix file \"%s\"\n",
VARDATA(filename));
}
/*
Submited by: skimo@breughel.ufsia.ac.be (Sven Verdoolaege)
1996-09-23 08:29:52 +00:00
ff4d0d435e
Here's a patch that I discussed recently on pg95-dev that changes the
...
way one creates a database system. Parts that were in "make install"
are not either in "make all" or initdb. Nothing goes in the PGDATA
directory besides user data. Creating multiple database systems is
easier.
In addition to applying the patch, it is necessary to move the file
libpq/pg_hba to backend/libpq/pg_hba.sample.
Submitted by: Bryan Henderson <bryanh@giraffe.netgate.net>
1996-09-23 08:24:13 +00:00
e7c3adcd94
*** src/backend/storage/file/fd.c.orig Thu Sep 12 17:17:21 1996
...
--- src/backend/storage/file/fd.c Thu Sep 12 17:23:38 1996
***************
*** 262,268 ****
Delete(file);
/* save the seek position */
! fileP->seekPos = lseek(fileP->fd, 0L, SEEK_CUR);
Assert( fileP->seekPos != -1);
/* if we have written to the file, sync it */
--- 262,268 ----
Delete(file);
/* save the seek position */
! fileP->seekPos = (long) lseek(fileP->fd, 0L, SEEK_CUR);
Assert( fileP->seekPos != -1);
/* if we have written to the file, sync it */
Submitted by: Randy Terbush <randy@zyzzyva.com>
1996-09-22 01:30:52 +00:00
ee420c049f
|This fixes the bug introduced yesterday that causes "not found" errors
...
|with gram.c and parser.h.
|
|--
|Bryan Henderson Phone 408-227-6803
|San Jose, California
|
1996-09-21 06:29:09 +00:00
83298ee312
Nothing patched...just testing something with CVS...
1996-09-21 06:27:11 +00:00
1a003fbcc2
Various patches from Bryan that *should* clean up the compile problems
...
ppl are seeing with v2.0
1996-09-20 08:34:39 +00:00
3c237e7347
I have two small patches which correct some very obscure bug in the parser
...
of the array constants and in one of the loadable modules I posted some time
ago.
Submitted by: Massimo Dal Zotto <dz@cs.unitn.it>
1996-09-19 20:05:59 +00:00
8594ade7b0
Another small change. Added the void keyword to an empty arg list.
...
Submitted by: darcy@druid.druid.com (D'Arcy J.M. Cain)
1996-09-19 20:02:53 +00:00
b19b90d706
Just a small change to flesh out the func_ptr typedef declaration.
...
Submitted by: darcy@druid.druid.com (D'Arcy J.M. Cain)
1996-09-19 20:01:40 +00:00
3023dc6af0
A few changes to cleanup the code.
...
- Added the header access/heapam.h.
- Changed all instances of "length" to "data_length" to quiet
the compiler.
- initialized a few variables. The compiler couldn't see that
the code guaranteed that these would be initialized before
being dereferenced. If anyone wants to check my work follow
the usage of these variables and make sure that this true
and wasn't actually a bug in the original code.
- added a missing break statement to a default case. This
was a benign error but bad style.
- layed out heap_sysattrlen differently. I think this way
makes the structure of the code crystal clear. There should
be no actual difference in the actual behaviour of the code.
Submitted by: darcy@druid.druid.com (D'Arcy J.M. Cain)
1996-09-19 20:00:37 +00:00
ea8adfc435
Added querylimit patch...
...
Submitted by: Massimo Dal Zotto <dz@cs.unitn.it>
1996-09-19 19:57:18 +00:00
aceac3a927
Fix for pg_log bug
...
Submitted by: "Vadim B. Mikheev" <vadim@sable.krasnoyarsk.su>
1996-09-19 19:50:48 +00:00
40aade2c06
At Andrew's suggestion, upgrade the Version numbers to reflect the
...
current state of development...namely, we are on 2.0
NOTE:
BTW, the is also a check in postmaster which won't let you use an older
version of the database by checking the version number. The version number
of a database is in data/PG_VERSION (a plain ASCII file).
- Andrew
1996-09-16 06:11:44 +00:00
e3b41d40fa
More code cleanups from "Kurt J. Lidl" <lidl@va.pubnix.com>
1996-09-16 05:40:31 +00:00
906651f663
There is a bug in the function executor. The backend crashes while trying to
...
execute an sql function containing an utility command (create, notify, ...).
The bug is part in the planner, which returns a number of plans different
than the number of commands if there are utility commands in the query, and
in part in the function executor which assumes that all commands are normal
query commands and causes a SIGSEGV trying to execute commands without plan.
Submitted by: Massimo Dal Zotto <dz@cs.unitn.it>
1996-09-16 05:36:38 +00:00
475493130d
|Subject: Postgres patch: Assert attribute type match
...
|
|Here's a patch for Version 2 only. It just adds an Assert to catch some
|inconsistencies in the catalog classes.
|
|--
|Bryan Henderson Phone 408-227-6803
|San Jose, California
|
1996-09-16 05:33:20 +00:00
796f78998e
Fixes:
...
The problem is that the function arguments are not considered as possible key
candidates for index scan and so only a sequential scan is possible inside
the body of a function. I have therefore made some patches to the optimizer
so that indices are now used also by functions. I have also moved the plan
debug message from pg_eval to pg_plan so that it is printed also for plans
genereated for function execution. I had also to add an index rescan to the
executor because it ignored the parameters set in the execution state, they
were flagged as runtime variables in ExecInitIndexScan but then never used
by the executor so that the scan were always done with any key=1. Very odd.
This means that an index rescan is now done twice for each function execution
which uses an index, the first time when the index scan is initialized and
the second when the actual function arguments are finally available for the
execution. I don't know what is the cost of an double index scan but I
suppose it is anyway less than the cost of a full sequential scan, at leat
for large tables. This is my patch, you must also add -DINDEXSCAN_PATCH in
Makefile.global to enable the changes.
Submitted by: Massimo Dal Zotto <dz@cs.unitn.it>
1996-09-10 06:48:52 +00:00
f2f53aee0f
Fixes:
...
The comparison routines for text and char data type give incorrect results
if the input data contains characters greater than 127. As these routines
perform the comparison using signed char variables all character codes
greater than 127 are interpreted as less than 0. These codes are used to
encode the iso8859 char sets.
The other text-like data types seem to work as expected as they use unsigned
chars in comparisons.
Submitted by: Massimo Dal Zotto <dz@cs.unitn.it>
1996-09-10 06:41:38 +00:00
1ba34d91fc
Cleaned up a spurious '-' ...
...
Pointed out by: ernst.molitor@uni-bonn.de
1996-09-10 06:23:46 +00:00
4978d3f4bb
modifications required to reflect centralized include files
1996-08-28 22:50:24 +00:00
870be9fa8e
Clean up th ecompile process by centralizing the include files
...
- code compile tested, but due to a yet unresolved problem with
parse.h's creation, compile not completed...
1996-08-28 07:27:54 +00:00
907c884fe8
Went back one directory too far for the -I include
1996-08-28 02:18:00 +00:00
f634c14c9e
Path to pg_proc.h to create fmgrtab.c change to include/catalog
1996-08-28 02:13:10 +00:00
ca405ae4bf
Moved the include files to src/include/regex
1996-08-28 01:55:44 +00:00
b8a0bb68c9
Should finish cleaning out the machine.h includes
1996-08-27 22:21:34 +00:00
6196646291
Remove include of machine.h
1996-08-27 22:20:28 +00:00
3c0f8ed4ae
Remove all traces of machine.h and redundant calls to c.h where
...
postgres.h already pulled in (postgres.h includes c.h)
1996-08-27 22:15:17 +00:00
eadf5dc754
#include "postgres.h" exists in most .c files in system, so adding
...
#include "config.h" here will (should?) ensure that any platform
dependencies defined in config.h should be reflected in all .c files...
1996-08-27 22:00:21 +00:00
9247b29228
The use of include files is a mess...alot of redundancy, it seems...
...
First Step: Centralize them under on src/include hierarchy
1996-08-27 21:49:14 +00:00
e1f31a2bb6
added #include "config.h" for ESCAPE_PATCH define
1996-08-27 07:42:29 +00:00
321b8c80fd
At this rate, maybe next year sometime I'll get this done...
...
Goals: reduce the difficulty of porting from platform to platform,
release to release, but moving as much as possible into config.h
1996-08-27 07:32:33 +00:00
c028568217
first pass...move some of the "Port" dependencies to src/include/config.h
1996-08-27 06:55:28 +00:00
51b3f2d371
Create support for a "common" include directory for the source
...
tree, instead of having include files all over the place...
Immediate goal...a 'config.h' file so that we can make #ifdef's
being used throughout the code more a rarity as far as porting
is concerned
1996-08-27 06:10:50 +00:00
bfc308d196
Damn, we really need to clean up this "include file" dilemna...
...
include files *everywhere* ;(
1996-08-26 23:04:07 +00:00
491b9b89c4
The patch that is applied at the end of the email makes sure that these
...
conditions are always met. The patch can be applied to any version
of Postgres95 from 1.02 to 1.05. After applying the patch, queries
using indices on bpchar and varchar fields should (hopefully ;-) )
always return the same tuple set regardless to the fact whether
indices are used or not.
Submitted by: Gerhard Reithofer <tbr_laa@AON.AT>
1996-08-26 20:38:52 +00:00
Bruce Momjian
Bryan Henderson
Marc G. Fournier
Vadim B. Mikheev