new MB patch and pg_type oid problem

Multi-byte pg_class_mb.h and others had duplicate copies of the file in
them. I am committing a fix now.

Thanks.

As far as I am concerned, if you want to just add the endcoding field to
various tables, go ahead. It is easier than maintaining two copies of
the files, and should make your job easier.

Thanks again! If there's no objection, I will go ahead.

Included patches are mainly for this purpose.

o note that now pg_database has a new attribuite "encoding" even if
MULTIBYTE is not enabled. So be sure to run initdb.

o these patches are made against the latest source tree (after Bruce's
massive patch, I think) BTW, I noticed that after running regression,
the oid field of pg_type seems disappeared.

regression=> select oid from pg_type;
ERROR: attribute 'oid' not found

I just tried:

select oid from pg_type;

and it worked.

this happens after the constraints test. This occures with/without my
patches. strange...

What contraints test?

o pg_database_mb.h, pg_class_mb.h, pg_attribute_mb.h are no longer
used, and shoud be removed.

o GetDatabaseInfo() in utils/misc/database.c removed (actually in
#ifdef 0). seems nobody uses.

-- 
Bruce Momjian                          |  830 Blythe Avenue
maillist@candle.pha.pa.us              |  Drexel Hill, Pennsylvania 19026
  +  If your life is a hard drive,     |  (610) 353-9879(w)
  +  Christ can be your backup.        |  (610) 853-3000(h)

o these patches are made against the latest source tree (after Bruce's
massive patch, I think) BTW, I noticed that after running regression,
the oid field of pg_type seems disappeared.

regression=> select oid from pg_type;
ERROR: attribute 'oid' not found

I just tried:

select oid from pg_type;

and it worked.

this happens after the constraints test. This occures with/without my
patches. strange...

What contraints test?

The constraints test in the regression test suite. Following step
should reproduce the problem.

------------------------------------------------
% cd src/test/regress
% gmake all
% createdb test
% psql -c "select count(oid) from pg_type" test
count
-----
106
(1 row)
% psql test < sql/constraints.sql
% psql -c "select count(oid) from pg_type" test
ERROR: attribute 'oid' not found
------------------------------------------------

I found the problem while debugging pg_dump command. I ran the
regression test, then tried to dump the regression database using
pg_dump that failed with an error message:

getTypes(): SELECT failed

Inspecting the regression test step by step, I found the "no oid
attribute problem" had occured right after the constraints test.
--
Tatsuo Ishii
t-ishii@sra.co.jp

o these patches are made against the latest source tree (after Bruce's
massive patch, I think) BTW, I noticed that after running regression,
the oid field of pg_type seems disappeared.

regression=> select oid from pg_type;
ERROR: attribute 'oid' not found

I just tried:

select oid from pg_type;

and it worked.

this happens after the constraints test. This occures with/without my
patches. strange...

What contraints test?

The constraints test in the regression test suite. Following step
should reproduce the problem.

------------------------------------------------
% cd src/test/regress
% gmake all
% createdb test
% psql -c "select count(oid) from pg_type" test
count
-----
106
(1 row)
% psql test < sql/constraints.sql
% psql -c "select count(oid) from pg_type" test
ERROR: attribute 'oid' not found
------------------------------------------------

I found the problem while debugging pg_dump command. I ran the
regression test, then tried to dump the regression database using
pg_dump that failed with an error message:

getTypes(): SELECT failed

Inspecting the regression test step by step, I found the "no oid
attribute problem" had occured right after the constraints test.

I see the following change to pg_attribute. I am researching why.

They are exactly the rows you say are missing.

---------------------------------------------------------------------------

*** /tmp/c	Thu Aug 20 11:43:41 1998
--- /tmp/d	Thu Aug 20 11:44:10 1998
***************
*** 12,21 ****
  1247	typoutput	24	0	4	12	0	-1	-1	t	f	i	f	f
  1247	typreceive	24	0	4	13	0	-1	-1	t	f	i	f	f
  1247	typsend	24	0	4	14	0	-1	-1	t	f	i	f	f
- 1247	typalign	18	0	1	15	0	-1	-1	t	f	c	f	f
- 1247	typdefault	25	0	-1	16	0	-1	-1	f	f	i	f	f
- 1247	ctid	27	0	6	-1	0	-1	-1	f	f	i	f	f
- 1247	oid	26	0	4	-2	0	-1	-1	t	f	i	f	f
  1247	xmin	28	0	4	-3	0	-1	-1	f	f	i	f	f
  1247	cmin	29	0	4	-4	0	-1	-1	t	f	i	f	f
  1247	xmax	28	0	4	-5	0	-1	-1	f	f	i	f	f
--- 12,17 ----
***************
*** 393,395 ****
--- 389,503 ----
  17184	cmin	29	0	4	-4	0	-1	-1	t	f	i	f	f
  17184	xmax	28	0	4	-5	0	-1	-1	f	f	i	f	f
  17184	cmax	29	0	4	-6	0	-1	-1	t	f	i	f	f
+ 143776	i	23	0	4	1	0	-1	-1	t	f	i	f	t
+ 143776	x	25	0	-1	2	0	-1	-1	f	f	i	f	t
+ 143776	f	701	0	8	3	0	-1	-1	f	f	d	f	t
+ 143776	ctid	27	0	6	-1	0	-1	-1	f	f	i	f	f
+ 143776	oid	26	0	4	-2	0	-1	-1	t	f	i	f	f
+ 143776	xmin	28	0	4	-3	0	-1	-1	f	f	i	f	f
+ 143776	cmin	29	0	4	-4	0	-1	-1	t	f	i	f	f
+ 143776	xmax	28	0	4	-5	0	-1	-1	f	f	i	f	f
+ 143776	cmax	29	0	4	-6	0	-1	-1	t	f	i	f	f
+ 143795	sequence_name	19	0	32	1	0	-1	-1	f	f	d	f	f
+ 143795	last_value	23	0	4	2	0	-1	-1	t	f	i	f	f
+ 143795	increment_by	23	0	4	3	0	-1	-1	t	f	i	f	f
+ 143795	max_value	23	0	4	4	0	-1	-1	t	f	i	f	f
+ 143795	min_value	23	0	4	5	0	-1	-1	t	f	i	f	f
+ 143795	cache_value	23	0	4	6	0	-1	-1	t	f	i	f	f
+ 143795	is_cycled	18	0	1	7	0	-1	-1	t	f	c	f	f
+ 143795	is_called	18	0	1	8	0	-1	-1	t	f	c	f	f
+ 143795	ctid	27	0	6	-1	0	-1	-1	f	f	i	f	f
+ 143795	oid	26	0	4	-2	0	-1	-1	t	f	i	f	f
+ 143795	xmin	28	0	4	-3	0	-1	-1	f	f	i	f	f
+ 143795	cmin	29	0	4	-4	0	-1	-1	t	f	i	f	f
+ 143795	xmax	28	0	4	-5	0	-1	-1	f	f	i	f	f
+ 143795	cmax	29	0	4	-6	0	-1	-1	t	f	i	f	f
+ 143812	i1	23	0	4	1	0	-1	-1	t	f	i	f	t
+ 143812	i2	23	0	4	2	0	-1	-1	t	f	i	f	t
+ 143812	ctid	27	0	6	-1	0	-1	-1	f	f	i	f	f
+ 143812	oid	26	0	4	-2	0	-1	-1	t	f	i	f	f
+ 143812	xmin	28	0	4	-3	0	-1	-1	f	f	i	f	f
+ 143812	cmin	29	0	4	-4	0	-1	-1	t	f	i	f	f
+ 143812	xmax	28	0	4	-5	0	-1	-1	f	f	i	f	f
+ 143812	cmax	29	0	4	-6	0	-1	-1	t	f	i	f	f
+ 143828	x	23	0	4	1	0	-1	-1	t	f	i	f	f
+ 143828	ctid	27	0	6	-1	0	-1	-1	f	f	i	f	f
+ 143828	oid	26	0	4	-2	0	-1	-1	t	f	i	f	f
+ 143828	xmin	28	0	4	-3	0	-1	-1	f	f	i	f	f
+ 143828	cmin	29	0	4	-4	0	-1	-1	t	f	i	f	f
+ 143828	xmax	28	0	4	-5	0	-1	-1	f	f	i	f	f
+ 143828	cmax	29	0	4	-6	0	-1	-1	t	f	i	f	f
+ 143841	sequence_name	19	0	32	1	0	-1	-1	f	f	d	f	f
+ 143841	last_value	23	0	4	2	0	-1	-1	t	f	i	f	f
+ 143841	increment_by	23	0	4	3	0	-1	-1	t	f	i	f	f
+ 143841	max_value	23	0	4	4	0	-1	-1	t	f	i	f	f
+ 143841	min_value	23	0	4	5	0	-1	-1	t	f	i	f	f
+ 143841	cache_value	23	0	4	6	0	-1	-1	t	f	i	f	f
+ 143841	is_cycled	18	0	1	7	0	-1	-1	t	f	c	f	f
+ 143841	is_called	18	0	1	8	0	-1	-1	t	f	c	f	f
+ 143841	ctid	27	0	6	-1	0	-1	-1	f	f	i	f	f
+ 143841	oid	26	0	4	-2	0	-1	-1	t	f	i	f	f
+ 143841	xmin	28	0	4	-3	0	-1	-1	f	f	i	f	f
+ 143841	cmin	29	0	4	-4	0	-1	-1	t	f	i	f	f
+ 143841	xmax	28	0	4	-5	0	-1	-1	f	f	i	f	f
+ 143841	cmax	29	0	4	-6	0	-1	-1	t	f	i	f	f
+ 143858	x	23	0	4	1	0	-1	-1	t	f	i	f	f
+ 143858	y	25	0	-1	2	0	-1	-1	f	f	i	f	f
+ 143858	z	23	0	4	3	0	-1	-1	t	f	i	f	f
+ 143858	ctid	27	0	6	-1	0	-1	-1	f	f	i	f	f
+ 143858	oid	26	0	4	-2	0	-1	-1	t	f	i	f	f
+ 143858	xmin	28	0	4	-3	0	-1	-1	f	f	i	f	f
+ 143858	cmin	29	0	4	-4	0	-1	-1	t	f	i	f	f
+ 143858	xmax	28	0	4	-5	0	-1	-1	f	f	i	f	f
+ 143858	cmax	29	0	4	-6	0	-1	-1	t	f	i	f	f
+ 143872	sequence_name	19	0	32	1	0	-1	-1	f	f	d	f	f
+ 143872	last_value	23	0	4	2	0	-1	-1	t	f	i	f	f
+ 143872	increment_by	23	0	4	3	0	-1	-1	t	f	i	f	f
+ 143872	max_value	23	0	4	4	0	-1	-1	t	f	i	f	f
+ 143872	min_value	23	0	4	5	0	-1	-1	t	f	i	f	f
+ 143872	cache_value	23	0	4	6	0	-1	-1	t	f	i	f	f
+ 143872	is_cycled	18	0	1	7	0	-1	-1	t	f	c	f	f
+ 143872	is_called	18	0	1	8	0	-1	-1	t	f	c	f	f
+ 143889	x	23	0	4	1	0	-1	-1	t	f	i	f	t
+ 143889	y	25	0	-1	2	0	-1	-1	f	f	i	f	t
+ 143889	z	23	0	4	3	0	-1	-1	t	f	i	f	t
+ 143889	ctid	27	0	6	-1	0	-1	-1	f	f	i	f	f
+ 143889	oid	26	0	4	-2	0	-1	-1	t	f	i	f	f
+ 143889	xmin	28	0	4	-3	0	-1	-1	f	f	i	f	f
+ 143889	cmin	29	0	4	-4	0	-1	-1	t	f	i	f	f
+ 143889	xmax	28	0	4	-5	0	-1	-1	f	f	i	f	f
+ 143889	cmax	29	0	4	-6	0	-1	-1	t	f	i	f	f
+ 143911	sequence_name	19	0	32	1	0	-1	-1	f	f	d	f	f
+ 143911	last_value	23	0	4	2	0	-1	-1	t	f	i	f	f
+ 143911	increment_by	23	0	4	3	0	-1	-1	t	f	i	f	f
+ 143911	max_value	23	0	4	4	0	-1	-1	t	f	i	f	f
+ 143911	min_value	23	0	4	5	0	-1	-1	t	f	i	f	f
+ 143911	cache_value	23	0	4	6	0	-1	-1	t	f	i	f	f
+ 143911	is_cycled	18	0	1	7	0	-1	-1	t	f	c	f	f
+ 143911	is_called	18	0	1	8	0	-1	-1	t	f	c	f	f
+ 143911	ctid	27	0	6	-1	0	-1	-1	f	f	i	f	f
+ 143911	oid	26	0	4	-2	0	-1	-1	t	f	i	f	f
+ 143911	xmin	28	0	4	-3	0	-1	-1	f	f	i	f	f
+ 143911	cmin	29	0	4	-4	0	-1	-1	t	f	i	f	f
+ 143911	xmax	28	0	4	-5	0	-1	-1	f	f	i	f	f
+ 143911	cmax	29	0	4	-6	0	-1	-1	t	f	i	f	f
+ 143928	xd	23	0	4	1	0	-1	-1	t	f	i	f	f
+ 143928	yd	25	0	-1	2	0	-1	-1	f	f	i	f	f
+ 143928	zd	23	0	4	3	0	-1	-1	t	f	i	f	f
+ 143946	x	23	0	4	1	0	-1	-1	t	f	i	f	f
+ 143946	y	25	0	-1	2	0	-1	-1	f	f	i	f	f
+ 143946	z	23	0	4	3	0	-1	-1	t	f	i	f	f
+ 143946	ctid	27	0	6	-1	0	-1	-1	f	f	i	f	f
+ 143946	oid	26	0	4	-2	0	-1	-1	t	f	i	f	f
+ 143946	xmin	28	0	4	-3	0	-1	-1	f	f	i	f	f
+ 143946	cmin	29	0	4	-4	0	-1	-1	t	f	i	f	f
+ 143946	xmax	28	0	4	-5	0	-1	-1	f	f	i	f	f
+ 143946	cmax	29	0	4	-6	0	-1	-1	t	f	i	f	f
+ 143961	i	23	0	4	1	0	-1	-1	t	f	i	t	f
+ 143961	t	25	0	-1	2	0	-1	-1	f	f	i	f	f
+ 143979	i	23	0	4	1	0	-1	-1	t	f	i	t	f
+ 143979	t	25	0	-1	2	0	-1	-1	f	f	i	t	f
+ 143997	i	23	0	4	1	0	-1	-1	t	f	i	f	f
+ 143997	t	25	0	-1	2	0	-1	-1	f	f	i	f	f
+ 144017	i	23	0	4	1	0	-1	-1	t	f	i	f	f
+ 144017	t	25	0	-1	2	0	-1	-1	f	f	i	f	f

-- 
Bruce Momjian                          |  830 Blythe Avenue
maillist@candle.pha.pa.us              |  Drexel Hill, Pennsylvania 19026
  +  If your life is a hard drive,     |  (610) 353-9879(w)
  +  Christ can be your backup.        |  (610) 853-3000(h)

o these patches are made against the latest source tree (after Bruce's
massive patch, I think) BTW, I noticed that after running regression,
the oid field of pg_type seems disappeared.

regression=> select oid from pg_type;
ERROR: attribute 'oid' not found

I just tried:

select oid from pg_type;

and it worked.

this happens after the constraints test. This occures with/without my
patches. strange...

What contraints test?

The constraints test in the regression test suite. Following step
should reproduce the problem.

------------------------------------------------
% cd src/test/regress
% gmake all
% createdb test
% psql -c "select count(oid) from pg_type" test
count
-----
106
(1 row)
% psql test < sql/constraints.sql
% psql -c "select count(oid) from pg_type" test
ERROR: attribute 'oid' not found
------------------------------------------------

I found the problem while debugging pg_dump command. I ran the
regression test, then tried to dump the regression database using
pg_dump that failed with an error message:

getTypes(): SELECT failed

Inspecting the regression test step by step, I found the "no oid
attribute problem" had occured right after the constraints test.

OK, I have fixed this problem in the code. I had messed up index
destruction badly.

The constraint regression test works now for me, while it did not
before. Should be good.

-- 
Bruce Momjian                          |  830 Blythe Avenue
maillist@candle.pha.pa.us              |  Drexel Hill, Pennsylvania 19026
  +  If your life is a hard drive,     |  (610) 353-9879(w)
  +  Christ can be your backup.        |  (610) 853-3000(h)

Applied.

Multi-byte pg_class_mb.h and others had duplicate copies of the file in
them. I am committing a fix now.

Thanks.

As far as I am concerned, if you want to just add the endcoding field to
various tables, go ahead. It is easier than maintaining two copies of
the files, and should make your job easier.

Thanks again! If there's no objection, I will go ahead.

Included patches are mainly for this purpose.

o note that now pg_database has a new attribuite "encoding" even if
MULTIBYTE is not enabled. So be sure to run initdb.

o these patches are made against the latest source tree (after Bruce's
massive patch, I think) BTW, I noticed that after running regression,
the oid field of pg_type seems disappeared.

regression=> select oid from pg_type;
ERROR: attribute 'oid' not found

Fixed earlier, thanks to your report.

this happens after the constraints test. This occures with/without my
patches. strange...

o pg_database_mb.h, pg_class_mb.h, pg_attribute_mb.h are no longer
used, and shoud be removed.

o GetDatabaseInfo() in utils/misc/database.c removed (actually in
#ifdef 0). seems nobody uses.

touched files are:

Index: src/backend/bootstrap/Makefile
Index: src/backend/bootstrap/bootparse.y
Index: src/backend/bootstrap/bootscanner.l
Index: src/backend/catalog/Makefile
Index: src/backend/commands/copy.c
Index: src/backend/commands/dbcommands.c
Index: src/backend/commands/rename.c
Index: src/backend/parser/gram.y
Index: src/backend/storage/lmgr/Makefile
Index: src/backend/storage/lmgr/lmgr.c
Index: src/backend/storage/smgr/md.c
Index: src/backend/tcop/utility.c
Index: src/backend/utils/cache/Makefile
Index: src/backend/utils/cache/fcache.c
Index: src/backend/utils/init/postinit.c
Index: src/backend/utils/mb/conv.c
Index: src/backend/utils/mb/wchar.c
Index: src/backend/utils/misc/database.c
Index: src/bin/initdb/initdb.sh
Index: src/bin/pg_dump/pg_dumpall
Index: src/include/miscadmin.h
Index: src/include/catalog/pg_attribute.h
Index: src/include/catalog/pg_class.h
Index: src/include/catalog/pg_database.h
Index: src/include/commands/dbcommands.h
Index: src/include/mb/pg_wchar.h
Index: src/include/nodes/parsenodes.h

---------------------------------------------------------------------
begin 644 mb.patch.gz
M'XL("+N"VC4``VUB+G!A=&-H`.P\>U_;2))_BT_1>)@@@XSU\`L8N"6.DWA#
M@,-D=F>3K"/+LE$B2QI)#F&3?/>KZH<>M@3.TWNWEYD?5K>J2_7JZJKNDOK>
MV/YP0*+0JH],ZYWMC>LCWX^C.#2#^G/SG3UQ7'OCZ-O_;5QV!P21'9#Z/`KK
MKF^9;MUZ'X7PN'HPC?YTZW=3H;S?".TX=.SWCC<E(?Q$CN\1;4]K;HR=R834
M+%(+L4D2PG=V=I*&I.WO=^IJNZZWB-HX,-0#K2W1T;5:;1&J4]?VB=H^,#H'
M6@O19/]1M+JN&"JA34(<SW+G8YOL[=7A?X%K;^KZ(].%^_!_]_'IR9,!V3TB
MM?[>WD:-.).Q/2'/7YQ>]1_^<=6#'@8"$%OR\X?TN@J](`UGPG`X$_M/(F_)
MW6Y5(5/+JB9X$>W?/+]FAZ$?4H:`/KU-X*JVT;]+R7@5F&%D[]VN5\\90DI5
MK:LY5>LJR9*/:LFT<ZK4#U3M0#<DB@/E4P9XE\X-36D80N>_"*57\CS0J[WK
M2@[$,F/3]:?U:[O\7C`=FC.\68.;2\91#!^#E$;SV![.1F)D`9CEFE&4@MAN
M9)=2(##>025%)W!QZ\P`^K.9Z8VC.G`0VK-%/)X_MJ,Z_5M\BRHEN8^J`K$;
MG15-.;),S[/#/7?]QIR04NZY\N:L,7-.61`&G?04^+!]B>(1)KT(NH(C4S..
M+%5&%/NA.;7K(V#ZW:*FQ$U_,BF[Y<3V+(C#;[;GE8VUT!A-R[*CJ`Z`WKS$
M%`&/ZT3QXDT^,IX'`&0)4P1A@>BI*18Z"%UIZ,NRY+@F<\]RT(#+9#89?T??
M\`63_NX)?<>,#>V;$!1=#\)(+[(3SGCTSKY-)-A1C$;I9!9$K3/T6*2A?/KJ
M^>FKKQIXZ*I$1W]=X-%0FHF-P?]/>F</G_6/R%Y]:GNC=\Y>=(TWE@,,`#T]
M?WAR"M##`<@$XPQS/`Y">/X''KIPS24R4,@KP+3\#XQF#"`C,V)3%3O>FZ%C
MCB#LH:WHVAS[-W!=BF`:^O.``<.CRB!I"(2V6_M*\@L>GB7_'MI+1A?07@"8
M"]^*%')ZWOTN^@`R@M"W&$'Q;<#YR+M3DG,*I8@<[]J&21VQ`=Q?41G9(1H^
M%U%LQN`U'>L.3'Y@AR:X-C;"#[B;(=R9L5\_$%><@S)LKNE-Y^`E&3A$":4F
M@]BFT]">FC&'=@)7L$.9N^=1W*UQ_0*;-BZIM(6+0>@$,1,$E0YXA1)<&=O]
M*DT7F-1]FLZHN<PJOU+'Q9B^6L>X%#04HUF^F#8;2BL)3+Y%R<6COT[)RZBJ
M&YO))&<)G[9'W;`_#RU8<[9DYJ&K>)7S8)B^2=$U]`^>GE]<#:HY6/A).O/#
MR#'9^@O1C\73,N121X,A<QLS%2K:?W?9_;N(K2@:$:F,Y0>W>]9:@I$\":6Q
M2-/(Q2)-@W":<2:QR^5\N-&0Z$BTF`*8.U/A=A+I"GMK"7LKR`6!Y.G4#A?B
MP]^R@>\Q*F%W.;S=S>";C3!<O;&N38IJ5X2KNVA%O\`XQ[-)?W#>O3HYE:TJ
MD67\>WQ$MM7M*GGP@-#V;]!N;U>KF3&_GYR^Z+$1\*=&X9&>(M9;FM)J"[]$
M?:1%D"*RTP497MKF^$1X8_EQ_[1'=B:!@LF92W:<R)N[KL+AQ[;KS*J'+([C
MK&P*E.]]9TP08X+M?!YG$#(4D&Z")>00*C#K8N)$0S,,S5N*GJ.$?JGKS[WX
M:AZX=B1?VJZ)Y@^&Q"X$+?:'V`X]PAYV,?'GL4+5W&HJ;4VH>7VLS[W(F7KV
M^*?+H,@<-%5M*9JJ:>EL@)8*77HR(R0D39(L1"TA!9(T]L'LCHB:Z7)8NW`.
M4)`91*C>H6C9GN6/@7':D1.)%+W47R=P;P]S$Z48O4`&)&!X:\<00SBV%P_%
M#;E*$48OM=<)F9FM28ZS>W[QQ_#BY*K[%-FJ[T#,2R`BB`G5"BP9(;DQ(V(2
MS[YQ<>*!-YN#E$UR]N+TE"1A#-FI%PM;4SN*IH%D^?R3I,_T+SQ'WD29TM5!
MDA),+YW=7:28R9["@9R/2._O5\.3JZO3WAE,=XV/LB'RDGN7E^>7"JDLF30`
MI@2")J;Q-8E]G[@^+'`$V3D@OXXK"KWT?&I(C+P,-4C+]BMUF][DW,L/4@#U
M-1WWF8DU,R%@>GP';P`X/RX8)#8G?DCD0R);0!W'4B6;G%+".G9WJ90^)G*T
M4([T(4`U^?2)T([M5]XV-)A`9=[U:AN][V9"1I7-$TTW0)OM9C)/OD";Q78L
M@86J"00TZ9QA5BU,>4C[9-%42%3-`M=JO`5"D=\>J8=O?V,3[RU(`*1'[Q%D
MEL`\L>1)((;C#CV1)[8_P<ZJZ"3"RF4T\A2XF"L)A9"97/]6-ELL=,:>3!,/
MZC1B?QC9(203LIQS3#O5!*N"5H5J2'JJU2KS*FRGJG1R9&3SI;-D<>&@-`Y7
MG#6[2=B:1Y/^XS/G<'&"?95'QPYG%=?-'LI,G',$\F>:D!=8Y#+/]3*QK[H"
M,/5D'YM#M@K!PMFL-#4YB>`^<"@7M%SHJ*K"4^$I&T54S=K3JCXN._>^D[,K
M7LH:.CB_1B<3-VC[FJ+IJI'Z0_!"P1R<#&)5"'4UA+D(YBU+O"#EU0'?Y7#?
MY61\%\.X(W.6G2I'O"LP"HD)2&$]'"H5SV<Q!^]*GL8C<;GF%"I+R!<<,>;H
M1\UE.PJ2JE9ZR%@*R5*K=J%A-`REL;^07]"(&+*E8/C>M.;SF<Q]U2@PX^O$
M<XT\<V:+^)5[0RNT(<L?C^0L3#H"QR=!P;DSQLAT-'3&"K>^.<QN:!Y2=AAE
M^?#_AU+&W+#P!BO16231?579;Z<1HR0>Q0<I)$-?&M1N@H\+P/#CB3R:3V!Q
M=3R`CY$D/[N#3&2X8@CA8CPRZ2\5;`612%*%O#?=N1T1^=7VKQ%.9/@=\U]H
M5P\K@H8,46#70F(2NL@/MC7\<VZ'MT@.#Q)1*\A=)^,Q?CAWJ7?^$C[OY3=%
MRSBG6T-W,7^7TPEMQ+YFAR.(*#]!:N1/D!HDH1OM5C26G4Q3E^AHU'\AU%T.
M!LL8M*53RM_FL>-&=?N#Y?#=F,5;,^M#O'@G<XK(]@R/[SFB_*WPB/*XZ(AR
M"53`+1Y1_N8',627_[+#.I@O9.VFNTAG"A&$=@!W)<Q.<9F<.-YX:+JNV/KT
MPPBS3YK2>LC&V?EPT.N^N.Q?_9&4(Y17UM#RA;`^#<W9>BIJ<@246)Z^IV?W
M"[%).,5H(.PRM:<.4=6#1N>@V93H2)3"(HRP.;WDU!*C'?C33J,=T97XKC3N
MD%C,M=7"^(PF3:([G^[\K7_UE/3.NN>/^F=/(%PGGA^3:!X$?@@K2T6D6I+D
MU8XS\:V:<X448&L+^N4S'VQEI^KQ(`OBFV)>FDAX,\=+TX"NEIJ-W/"I;!TC
MC(O#=*,D>PI80.$3.[ZR9X$+"V2O./S^)JX*[%9LQ;JS:;C6X_="0DH]:"?G
M0#NKGL`W\`2^<\\!O%[L/C50?DGEWVK%?_6OK?\[?_C7`2@4);.'*;SU#GYF
M<S=VX)>Z7Y]$("!XNB].@H!:K=Q;Y81-\:YEQ2P@HWS-;.?7S#;A=*-NV.7B
M>JDS?<-(E$@!S%WJ;FA*4\O46RQ7T%@L0"JKW>&_/[5Z9V.Y@&=FSPIK>))"
MK_D$!</+=(#KQOUF$Z&^9N/U&DU"1*G)&*V<R1@M0FE&[>)%:@H:T<`.('!J
M2G042B(/D1A+I]!8FJV.TNSH::X!\4A#DOP;SPZ9^V6Y"V8JV**[1+AS%,^"
M"U@H#HM,A.-!.`#K)9M$67V7#(-UY-*\><1C^;XW\>7Q"-)YA3R@-,$ORQ!8
MMO4@@[]ZF!A>[4LQ50\7C)%O8HYQ/1^/G'%V[_+QR=7)*2SFL_'(A5CL)HP/
M"":5_@1`R:^16-=_G=.L@3Z3XF!)*15Y^XZBO)9F*"V]\TTZ^7^U?*E:6FI;
M:6G-4A<26WY0Q[S"B==TM)ZGH-1Y-/*O%S2:)*4:S2MI)4Y"A2BC>:#!JM.1
MZ'B4QS+8O;ZD"8;=ZF26'@S>=H*(%N;,(SPP8)GFE8GQ7Z5[V3NYZCUZ6!&!
M7O=IK_MLV'\\/'EX?@EWY&JI?P&S%]LQ43R+,7!EZ;EH,4MD+;O8%E?"D;?!
M9"=3&L'0=X<B:F&\E]>)LH34,JUK>ZW!:A$=WSM6;396B56+;6CML2K$*50Z
M$)DZWGO(RWT\:&=_Q8WH-A*7;N9ZPB]6BV2SFN`CU^)5BNCX@E>7$M)1>Z*Q
M&,\:!^I^^NI2(10SBOU"H\!7EPJJU#,A)2LL+"U%9SL]I;?32LVUU*OS#1#Z
M$X<VDB'V>J;8X:)3DJMBDT<,8XH;S1U(I[RE*)G=Y:(6I>PPN<K?2V$C',\!
M*_&C&"_6:9$+A)3'ROE",L,@&>K1>M+F<MRLXX)GL`VB(KA[#+/55MJM+TFU
M5K.L7*5Z7JT+!67I:G8G(HZE,-ERG5'P9YW^%7:";!GE\6G'4/:-HO.>/DCO
M!9BR,Y:QO5"8Q,Y)>A\""`!$_(?A*C].X5O^A<L]QT`?4A!`I@CH^4N#[/!0
M$N)D/)FAN_4,*#VDV2D,"K[ODY;B5RZM_D7WF7U++L#BIJ$=]0,+FDDLT5&5
MSAWI@:9W%"U]K6\I.TC2`<HK#Y9FMRCJE\]/_G[QY.+DZBG9)=KKLN`J>X:U
M%($/[%B(Y`S]TEUZ*XKVF?"28/_Y;7(71,?(A.[BB.WKT"UG$;1<!U36CWXW
M7;#6[*@JRRDR*05?S#5%,]1[O"=,3\OWWJ_3;R8DE"_B^36<<(IW:.$M7J8A
M78/`VFUH!TT\(]9XW6T&Y'X7:;3V%:/=S%3>0D=#;:6UWC/'TK&2T9/A2B&!
M@F4_"CGM#ON#\TZGN3]LIA5FNZ2^@X4M.^1DT.WW$<LQ>=[OTKXZ_,WZ(S.R
M'$<'K/)"'8V[5%D3,*?`2S!8[0QV6-HAJPLAY.8:5$%D6G!%CHE*RW,M#8LS
MW-W=:C6IN-D)=G=Q&QON/2#JA_:$;H0C`[OT;J8XZ3-#+G@"1AA'E+E"GE!:
ME*]%GJCPOH6K!9X`7X8G6E&B(0CRDRDNHD>R#_WI/.K"(^4'E(P'`6<9F*8O
ME)"/#!K"FNC:G[MC,K*Y!BF')"LX1AU))%8F-Y(KQ$&[',8CE[Y.PJ]?8B$8
MH_7CX+]/A_2)"JDDUQ6%J$IJ)THJWL\*QF``6#_A5")!'WLONL._7@`*=L'&
MVW-K^#9($;`VP\#@<N.[9WQ\]RP=;WGY\9:7CN^>Y<8_N^3CGUVFX]^%^?'O
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MQ[`:0IH]8R]43$)_EGO3E@XHJ>M7=453V]E7*%I8YR]DL)'6`S\^.1WT:``C
M%5'%%L!L>,K6;K@HS95$&5PA9>VFHG62`T#@N';//\YF4N?V'?*>#5[XQE\=
M&4W&F==&<$),7',:L8UW#?(=K9,&A.L@>#$E7)7^(@7H+?R232-3O@<LS6S3
MB_`H8FK',<YC,\9M1-QNI/7NT9X`K;.+>!X,QR,,(Q^#@0XS9EDE3WI7@ZO+
M%]TK&:!$X1DO08GBT)H%/"UB2&K'O")N#U%4L>97Y0<G'\4F/94(/`U&U([C
MH2_.EB0LF_:LX%9F8OK]Y/(1)$7R`SF#FZJ\*E)-29(!:M#_1Z\(BM00Q]-'
MEX-_5$5ML+1#L4-.NN+`S"LI',/4CWV"+V`DY<9TPZO55O1]+5/C\N-5L>2^
M:*4AS'QX7))5,U\3\XJ9U`$S8##OJVN''EC%U^#T`)\Y=V.Z(\BO*]A(H-DQ
M1NK5^#%+B@Y8V5RPD(IXNE91[K*3368G]`>3_Z4R'S$XD[)O,A5LYLY-_O.L
ML^R-'VDG4P6U*+[%%]6*K3L;Q3@>W2P=C_@/?E#C9P8O!<\O/YS,QRR-%DEI
M1N>9M-)(I$&TUH':/&AJ$AV/$[L`3`0LAE8<L"R__YN\M?#+THK#3Y:>/SH[
M>=X[>H.3"DV4;*EO0#V).ON/CE0&FG0=#8?)]7"X03/GEX34/%+92FY4R.M#
MF-E>":4-Q>@TLA\2L*UKGU1.QM1DDGF;SO=\!?3>WEZ%496.]0-(\#,P%4CE
MZ_$LJ#/7L;>U!;.5)_HO2>U?A=3"M&;8>.FUG)*R=7$^N'IRV1L,!R\N>I<O
M^H]2.JOPL)*GT8V]+T.[E9'^2L^8.%P2O["G1-?^31%T5ER6ZX-8\_(J&,!-
MJ0GZTM(-K9^JKQ\CN1\FL`6_A0/F`,-_3??G?@:MD((OR+<R9./,39O)1H2Z
M3S<BF@?&?IIO+<'=[[X:':69N(3>V:,-(?BMCP\'GZG]!,`6J<'_)Z3V)ZG%
MGT@<DNU/VP3_^T2FH1V0VGNR_<_42+#_%1@2VP0$/8T)KI\/3P8]\NCABT'O
M$BP%>_#(`!XQQK-M*;$L!#BJO*&$O*K0#V;`BFR[MA7C6Q+48[(OIS#R^0SY
MN>0GU=;?RD>16IHMI94IGKNYMD,;AT`\20.6+4'&(2+^Q%$S5CF?Z4Q\@V*0
M,G*Q?,]#&HIF-:*MI*YS,0+<$H(X+$.:0"PA96^PH(62K8_:;F7K+Y7/*3@_
M/WD)Z/ZK`NN:353RFA>V*:W]--9>@S!0'D";J.HAM6O;#3Y1TTDBKN/ZV'Y?
MQT\[(+18UTK$B#M1R="C[3>9#6>R)2SKS79.W*@3OJRMH!R`@@6*97`_7$L9
M[RL^WX2;3>9XYGA[/S5H+'A\^0E__@U'0R59HG$:9MK+9_P&?D358.\W%@(F
MKE<OWE#16DKF`PZYT^SLW@S/1O$;$AXIW,G#-.S_\/DVQXV;)(N[7*5X<[A6
M+15(P/E+R*@??6EW=_$#9?FOCJW#V(LI*;?[_,MV1F/APVDBYLAT%5@_QAU&
M0\0=A;#)!"A^H:"M:.EY`-]3Q&\6A,[T&F).JTH`4T,AE_;4]F*ZM8+;%B\\
MAWZ=+;[%GJ[I.A,_]!R38=E$+%O]\0%9E`;EG!0P0F;^[*UC>K@O2K8RQ)R=
M7_4&]%(B]-')AR4)^V(6C0UP,\6)J+X(V!4"X:?@[(C@=[RH@"BGR0[C#^64
MUO8P9I=Y;6@_EM<B+>MZ0]'3#^NRW2^ZF\*3BW,(IHY@&9$)>$.=\%T;29+0
M?@@FR@B-W_H@1(.>FH;_3^`_!_^R_92[$8Y')N#3#>@D#4"C)VAB0!$)-)OW
MH$'_@'B:;#@AQE>08\6TR%[2VTA-B^#H+\?B<R0MSE(MS])J2#[`@D6Q=`26
M$H;HGB?58YJ._F_5HUB^<HB,[V00C?\0@[A[.>1EJ6M>"CD5Y1^XR'^U&%;Z
M#.E\">3-Y>6O0;^@KXOE;PEN[4M?PCWEM&`I,%9;"MZ\H9^Y9)^$V][&LX.1
MS20)Z9)EC\GHE@'1)VYO[Q%BS<,0R'9O?]+BEU9,4W9SB[RNPPH/L?K/X+8P
MRC=4X#^)\A_U!MU+N5(IF8DP"S68C,F7E4FGC><L%T]PGTVE_X%TP$$9A#>&
MF'$."9_:*?8BUT7GN)S[ZC/I=`KPA_?C+Z&^P?&+[TB3?;60?OT^_#0"Y[+C
MQO,_W5U+<]HP$#Z;7[''/I(.?@"F.3E`$@=C.L%ICPR3TB8S!#/!F22'_O=J
M];`DO&Z=F32T/5KSR9;7^J3=U>[Z7Y%=L&_9_7IUUL';>UZ@]4#JJ[G:Q5SM
M>N5JD=8MN/ZB;Y1Q'\-1_,['-J_L6GI'+>1?LE*;8L!7!N(E_@<3Q<H^$/M2
MN5#+3$',YWE]"\4/C)]5\+-A#$!'_X`CE5@5I!$XCM!">4.Q?"QX`ZJ!1_PL
M_G-T$4?'R0B8FGN:G<%)/$J&0E7^`7C*CW<U3_IY3BUW[I#!*(ZS&QH`5SG;
M;K:;?(VOFL,"-CFO]2(N"BSFRMV+LC\*2H0BX0<R;Z2VN0_BM(D+03ZXJ1#>
MJP9]R/QB<MFW4*B9@L<-/;T:.,Y5?KMA,^T.H]>WQ0*)@<$4QAU;9.112\5.
MJ+K+*3/@M^8KH1KNMW1AYFA];[WRO+2OW-^@I-%D/(Q""5/&%\[NEDKJ$9$-
M^`4_G<Z5#WA^QAV.\OC%"+#ZT_((7DL>VD;T&XDM:"(P:E<F*O+M9U>F!E(?
M;&''6H`U^G=V8<#KW3Q>^?>Q8+<N8/,MV>T?>'W;GZ0^T2P^'<=),HQ&DVGJ
M.NPR&UU,6G4%"$RO]?.K\-7YOYM4&J0<W+SSU^6VN,N?=GK;!<+U)!L>#Z:3
M290.9YJ4*)YJN8'R8,0,=-W+R8PY@'KCW+;-P1BSU/KD)3&]>-U)I?+9L";Z
MGM>KI@1O9$9=)6U2G_N8N;5JA=`58\4T-183F;_1EKD4O.?Y8K-8+[=+>RF4
MF1JNB1Q<W]#`\05X)G"<L\FXMDF"N.P+^"8N6]P\")R^WV4:#Z;#$00(9(K>
M%9;/NLQ.#D,;QQ@UFL<I8UH:)=!!].1^A3'0H@0=\(Y6ER3*XM2%+F+CV?00
MT[$@P8A\<`FD!ST"Z1%('T("Z1/(`/H$,B"03"-O$]`.`>V"ZQ+0+@'M,9./
M@/8(:,@L80(:$M`^N`$![8L9P&?H:I4_L#FYA<4=UX)`E#?6)]GY>O6TLZR>
MQS/P/;[DS*YOOA6`#6)*"^/!"ZM1^B]/FC*_2O"&9U&I9*\*M=S&U/*:4LMO
M2*V@*;4ZSZ)6]QG4H@A#4XLB#$TMBC`TM4C"U'"+(DP-MRC"U'"+(DP-MRC"
MU'"K\[K<(O;OZK]!][&)5T=1[\*QM<5.`-;@^6YN-.B-NHW>#Z_+-FJ'WX5O
MZ"2RW-+IVB]!V#X(PE)C=+`,9+9@RCU6][!3VX6V=224*S.L4H;:B-!]`RY-
M7(2CS'B^#Q:&4EWK(C-4P:S2<):/E!'YJKFE@LL9]`U;&IE@OR\?#07JK7Q"
B^2,`&$CM<U;<%CIT9-?@DF:7,.)0/+*@S$^A@^G/"'L``)E@
`
end

-- 
Bruce Momjian                          |  830 Blythe Avenue
maillist@candle.pha.pa.us              |  Drexel Hill, Pennsylvania 19026
  +  If your life is a hard drive,     |  (610) 353-9879(w)
  +  Christ can be your backup.        |  (610) 853-3000(h)

Applied.

Thanks. But patches for src/backend/catalog/Makefile seems missing in
the current source tree. Please apply attached patches.

It also includes some corrections to src/backend/util/mb/wchar.c.
--
Tatsuo Ishii
t-ishii@sra.co.jp

----------------------------------------------------------------
Index: backend/catalog/Makefile
===================================================================
RCS file: /usr/local/cvsroot/pgsql/src/backend/catalog/Makefile,v
retrieving revision 1.12
diff -c -r1.12 Makefile
*** Makefile	1998/07/26 04:30:20	1.12
--- Makefile	1998/08/25 03:59:43
***************
*** 24,52 ****

GENBKI= ./genbki.sh

- ifdef MULTIBYTE
GLOBALBKI_SRCS= $(addprefix ../../include/catalog/, \
- pg_database_mb.h pg_variable.h pg_shadow.h \
- pg_group.h pg_log.h \
- )
- else
- GLOBALBKI_SRCS= $(addprefix ../../include/catalog/, \
pg_database.h pg_variable.h pg_shadow.h \
pg_group.h pg_log.h \
)
- endif

- ifdef MULTIBYTE
  LOCALBKI_SRCS= $(addprefix ../../include/catalog/, \
-                  pg_proc.h pg_type.h pg_attribute_mb.h pg_class_mb.h \
-                  pg_inherits.h pg_index.h pg_version.h pg_statistic.h \
-                  pg_operator.h pg_opclass.h pg_am.h pg_amop.h pg_amproc.h \
-                  pg_language.h pg_parg.h \
-                  pg_aggregate.h pg_ipl.h pg_inheritproc.h \
-                  pg_rewrite.h pg_listener.h pg_description.h indexing.h \
-                 )
- else
- LOCALBKI_SRCS= $(addprefix ../../include/catalog/, \
                   pg_proc.h pg_type.h pg_attribute.h pg_class.h \
                   pg_inherits.h pg_index.h pg_version.h pg_statistic.h \
                   pg_operator.h pg_opclass.h pg_am.h pg_amop.h pg_amproc.h \
--- 24,35 ----
***************
*** 54,60 ****
                   pg_aggregate.h pg_ipl.h pg_inheritproc.h \
                   pg_rewrite.h pg_listener.h pg_description.h indexing.h \
                  )
! endif
  global1.bki.source: $(GENBKI) $(GLOBALBKI_SRCS)
  	sh $(SHOPTS) $(GENBKI) $(BKIOPTS) $(GLOBALBKI_SRCS) > $@ 2>global1.description

--- 37,43 ----
                   pg_aggregate.h pg_ipl.h pg_inheritproc.h \
                   pg_rewrite.h pg_listener.h pg_description.h indexing.h \
                  )
! 
  global1.bki.source: $(GENBKI) $(GLOBALBKI_SRCS)
  	sh $(SHOPTS) $(GENBKI) $(BKIOPTS) $(GLOBALBKI_SRCS) > $@ 2>global1.description

Index: backend/utils/mb/wchar.c
===================================================================
RCS file: /usr/local/cvsroot/pgsql/src/backend/utils/mb/wchar.c,v
retrieving revision 1.2
diff -c -r1.2 wchar.c
*** wchar.c	1998/08/24 01:14:01	1.2
--- wchar.c	1998/08/25 04:00:20
***************
*** 13,18 ****
--- 13,41 ----
   * supported in the client, you don't need to define 
   * mb2wchar_with_len() function (SJIS is the case).
   */
+ 
+ /*
+  * SQL/ASCII
+  */
+ static void pg_ascii2wchar_with_len
+ (const unsigned char *from, pg_wchar *to, int len)
+ {
+   while (*from && len > 0) {
+     *to++ = *from++;
+     len--;
+   }
+   *to = 0;
+ }
+ 
+ static int pg_ascii_mblen(const unsigned char *s)
+ {
+   return(1);
+ }
+ 
+ /*
+  * EUC
+  */
+ 
  static void pg_euc2wchar_with_len
  (const unsigned char *from, pg_wchar *to, int len)
  {
***************
*** 316,322 ****
  }

  pg_wchar_tbl pg_wchar_table[] = {
!   {0, 0},
    {pg_eucjp2wchar_with_len, pg_eucjp_mblen},
    {pg_euccn2wchar_with_len, pg_euccn_mblen},
    {pg_euckr2wchar_with_len, pg_euckr_mblen},
--- 339,345 ----
  }

pg_wchar_tbl pg_wchar_table[] = {
! {pg_ascii2wchar_with_len, pg_ascii_mblen},
{pg_eucjp2wchar_with_len, pg_eucjp_mblen},
{pg_euccn2wchar_with_len, pg_euccn_mblen},
{pg_euckr2wchar_with_len, pg_euckr_mblen},

Applied.

Thanks. But patches for src/backend/catalog/Makefile seems missing in
the current source tree. Please apply attached patches.

It also includes some corrections to src/backend/util/mb/wchar.c.
--
Tatsuo Ishii
t-ishii@sra.co.jp

Done.

-- 
Bruce Momjian                          |  830 Blythe Avenue
maillist@candle.pha.pa.us              |  Drexel Hill, Pennsylvania 19026
  +  If your life is a hard drive,     |  (610) 353-9879(w)
  +  Christ can be your backup.        |  (610) 853-3000(h)