CREATE TABLE t1 (s1 CHAR(5) CHARACTER SET latin1 COLLATE latin1_german2_ci);
INSERT INTO t1 VALUES ('Ü');
INSERT INTO t1 VALUES ('ue');
SELECT DISTINCT s1 FROM t1;

The above returned two rows in error.
Now it returns one row, in latin1_german2_ci:  Ü == ue

produced an error. This led to server crash in some cases, e.g.:
create table t7 (s1 char);
select * from t7
where concat(s1 collate latin1_general_ci,s1 collate latin1_swedish_ci)
= 'AA';

into bar.mysql.r18.ru:/usr/home/bar/mysql-4.1

into mysql.com:/home/mysql_src/mysql-4.1

which are fixed now.
In 4.1 there are 3 calls; 2 of them will be fixed when there is a merge;
here I fix the 3rd, new one. 

into sanja.is.com.ua:/home/bell/mysql/bk/work-unknown_table-4.1

into mysql.com:/home/kostja/mysql/mysql-4.1-root

into sanja.is.com.ua:/home/bell/mysql/bk/work-qc_ci-4.1

  Use the default character set instead of binary. This is to make "mysqlshow" -> "cat'n'paste database name" -> "mysqldump dbname" circle working.

  Upper/lower case translation bug for letter 't' fix.

into myvenu.com:/home/venu/work/sql/dev-4.1

by MAX/MIN in some cases:
SELECT coercibility(max(s1)) from t1;

Subselect collation and derivation was not processed
correctly:

create table a select (select s1 from t1);
select * from t1 where s1 = (select s2 from t1);

  Revise 4.1 Windows-install instructions.

(fixed BUG#943)

CREATE TABLE t1 (s1 CHAR(5) COLLATE latin1_german1_ci,
                 s2 CHAR(5) COLLATE latin1_swedish_ci);
SELECT * FROM t1 WHERE s1 = s2;

count HAVING clause elements with select list elements, because agregate function can be present in it (BUG#922)
removerd unused loop_id

into deer.(none):/home/hf/work/mysql-4.1.clean

  Logging to logging@openlogging.org accepted
README.1st:
  new file - Readme for Windows distribution

into deer.(none):/home/hf/work/mysql-4.1.clean

bugs #701 and #783 - old .frm files were read in incorrectly in 4.1 (each field was assigned a charset, BLOBs were converted to TEXT)

into eagle.mysql.r18.ru:/home/vva/work/BUG_663/mysql-4.1

into serg.mylan:/usr/home/serg/Abk/mysql-4.1

into mysql.com:/home/mysql_src/mysql-4.1

have in fact no effect, because in 4.0 the slave always runs with --skip-innodb.
In 4.1, mysqld accepts option --innodb which can override --skip-innodb.
So I add this option to the test.
When the updated test (using InnoDB tables) from 4.0 will be merged
into 4.1, with this good .opt file, we'll have a meaningful test in 4.1.
Until then, the test in 4.1 will pass but be useless.
And in 4.0, it will stay useless forever.
("it" = the part of the test which tests replication of foreign_key_checks)

into teton.kitebird.com:/home/paul/mysql-4.1

Proposed fix for bug#909

The latter is misleading because differs from the syntax used for GRANT/REVOKE.

Fix up test result affected by error message edits.