Hi, here are the patches to enhance existing MB handling. This time

I have implemented a framework of encoding translation between the
backend and the frontend. Also I have added a new variable setting
command:

SET CLIENT_ENCODING TO 'encoding';

Other features include:
	Latin1 support more 8 bit cleaness

See doc/README.mb for more details. Note that the pacthes are
against May 30 snapshot.

Tatsuo Ishii

1 files changed, 58 insertions, 7 deletions

diff --git a/src/test/regress/sql/mule_internal.sql b/src/test/regress/sql/mule_internal.sql
index 6d07906ff7..2e381f0f7e 100644
--- a/src/test/regress/sql/mule_internal.sql
+++ b/src/test/regress/sql/mule_internal.sql
@@ -5,17 +5,68 @@ create index ��ג�������ђ��index2 on ��ג�������ђ�� using hash (�ʬ�����������
 insert into ��ג�������ђ�� values('������Ԓ�咡������ǒ�������ג�쒥�','���A01���');
 insert into ��ג�������ђ�� values('������Ԓ�咡���������钥Ւ����Ò�����','�ʬB10���');
 insert into ��ג�������ђ�� values('������Ԓ�咡������ג�풥���钥ޒ��','���Z01���');
-insert into ��ג�������ђ�� values('����ԑ�ԑʾ���','���A01���');
-insert into ��ג�������ђ�� values('����ԑͼ���','���B01���');
-insert into ��ג�������ђ�� values('����ԑ�̑��Ա','���Z01���');
 vacuum ��ג�������ђ��;
 select * from ��ג�������ђ��;
 select * from ��ג�������ђ�� where �ʬ������������ = '���Z01���';
-select * from ��ג�������ђ�� where �ʬ������������ ~ 'Z01';
-select * from ��ג�������ђ�� where �ʬ������������ ~* 'z01';
+select * from ��ג�������ђ�� where �ʬ������������ ~* '���z01���';
 select * from ��ג�������ђ�� where �ʬ������������ like '_Z01_';
 select * from ��ג�������ђ�� where �ʬ������������ like '_Z%';
 select * from ��ג�������ђ�� where ��ђ�� ~ '������Ԓ�咡����[��ǒ��]';
 select * from ��ג�������ђ�� where ��ђ�� ~* '������Ԓ�咡����[��ǒ��]';
-select * from ��ג�������ђ�� where ��ђ�� ~ '�����[��ԑͼ]';
-select * from ��ג�������ђ�� where ��ђ�� ~* '�����[��ԑͼ]';
+select *,character_length(��ђ��) from ��ג�������ђ��;
+select *,octet_length(��ђ��) from ��ג�������ђ��;
+select *,position('���' in ��ђ��) from ��ג�������ђ��;
+select *,substring(��ђ�� from 10 for 4) from ��ג�������ђ��;
+drop table ��Ƒ�㑻�������;
+create table ��Ƒ�㑻�������(������ text, ��֑����� varchar, ����ע1A char(16));
+create index ��Ƒ�㑻�������index1 on ��Ƒ�㑻������� using btree(������);
+create index ��Ƒ�㑻�������index2 on ��Ƒ�㑻������� using btree(��֑�����);
+insert into ��Ƒ�㑻������� values('����ԑ�ԑʾ���','���A01���');
+insert into ��Ƒ�㑻������� values('����ԑͼ���','���B01���');
+insert into ��Ƒ�㑻������� values('����ԑ�̑��Ա','���Z01���');
+vacuum ��Ƒ�㑻�������;
+select * from ��Ƒ�㑻�������;
+select * from ��Ƒ�㑻������� where ��֑����� = '���Z01���';
+select * from ��Ƒ�㑻������� where ��֑����� ~* '���z01���';
+select * from ��Ƒ�㑻������� where ��֑����� like '_Z01_';
+select * from ��Ƒ�㑻������� where ��֑����� like '_Z%';
+select * from ��Ƒ�㑻������� where ������ ~ '�����[��ԑͼ]';
+select * from ��Ƒ�㑻������� where ������ ~* '�����[��ԑͼ]';
+select *,character_length(������) from ��Ƒ�㑻�������;
+select *,octet_length(������) from ��Ƒ�㑻�������;
+select *,position('���' in ������) from ��Ƒ�㑻�������;
+select *,substring(������ from 3 for 4) from ��Ƒ�㑻�������;
+drop table �ͪ�ߩ�Ѧ��듾�;
+create table �ͪ�ߩ�Ѧ��듾� (��듾� text, ��“׾��ړ�� varchar, ����1A��󓱸 char(16));
+create index �ͪ�ߩ�Ѧ��듾�index1 on �ͪ�ߩ�Ѧ��듾� using btree (��듾�);
+create index �ͪ�ߩ�Ѧ��듾�index2 on �ͪ�ߩ�Ѧ��듾� using hash (��“׾��ړ��);
+insert into �ͪ�ߩ�Ѧ��듾� values('��ēǻ��͓�𓽺��Ó�����', '�ѦA01�߾');
+insert into �ͪ�ߩ�Ѧ��듾� values('��ēǻ��͓�ד����ȓ��', '���B10���');
+insert into �ͪ�ߩ�Ѧ��듾� values('��ēǻ��͓����Γ�ד�����', '���Z01���');
+vacuum �ͪ�ߩ�Ѧ��듾�;
+select * from �ͪ�ߩ�Ѧ��듾�;
+select * from �ͪ�ߩ�Ѧ��듾� where ��“׾��ړ�� = '���Z01���';
+select * from �ͪ�ߩ�Ѧ��듾� where ��“׾��ړ�� ~* '���z01���';
+select * from �ͪ�ߩ�Ѧ��듾� where ��“׾��ړ�� like '_Z01_';
+select * from �ͪ�ߩ�Ѧ��듾� where ��“׾��ړ�� like '_Z%';
+select * from �ͪ�ߩ�Ѧ��듾� where ��듾� ~ '��ēǻ���[����]';
+select * from �ͪ�ߩ�Ѧ��듾� where ��듾� ~* '��ēǻ���[����]';
+select *,character_length(��듾�) from �ͪ�ߩ�Ѧ��듾�;
+select *,octet_length(��듾�) from �ͪ�ߩ�Ѧ��듾�;
+select *,position('���' in ��듾�) from �ͪ�ߩ�Ѧ��듾�;
+select *,substring(��듾� from 3 for 4) from �ͪ�ߩ�Ѧ��듾�;
+drop table test;
+create table test (t text);
+insert into test values('ENGLISH');
+insert into test values('FRAN��AIS');
+insert into test values('ESPA��OL');
+insert into test values('��SLENSKA');
+insert into test values('ENGLISH FRAN��AIS ESPA��OL ��SLENSKA');
+vacuum test;
+select * from test;
+select * from test where t = 'ESPA��OL';
+select * from test where t ~* 'espa��ol';
+select *,character_length(t) from test;
+select *,octet_length(t) from test;
+select *,position('L' in t) from test;
+select *,substring(t from 3 for 4) from test;