Description:
With a fresh pull of 4.1.8 this morning (latest changeset:
1.2158 04/12/10 02:36:45 heikki@hundin.mysql.fi +1 -0
dict0dict.c: Fix the bug that the character 0xA0 that EMS MySQL Manager
...)
I ran the test suite on my PC (Linux, x86, gcc 3.3).
It passed for the default options and with '--ps-protocol', but had failures
with '--embedded-server' which I now report individually.
For 'type_enum', the report is:
-------------------------------------------------------
*** r/type_enum.result 2004-12-07 21:10:31.000000000 +0300
--- r/type_enum.reject 2004-12-10 14:37:16.000000000 +0300
***************
*** 1731,1739 ****
alter table t1 add c enum ('Y','N') CHARACTER SET utf8 COLLATE utf8_bin;
select * from t1;
Catalog Database Table Table_alias Column Column_alias Name Type Length Max length Is_null Flags Decimals Charsetnr
! def test t1 t1 a a 254 3 1 Y 384 0 8
! def test t1 t1 b b 254 9 0 Y 2176 0 8
! def test t1 t1 c c 254 3 0 Y 384 0 8
a b c
Y NULL NULL
drop table t1;
--- 1731,1739 ----
alter table t1 add c enum ('Y','N') CHARACTER SET utf8 COLLATE utf8_bin;
select * from t1;
Catalog Database Table Table_alias Column Column_alias Name Type Length Max length Is_null Flags Decimals Charsetnr
! def test t1 t1 a a 254 3 1 Y 384 0 83
! def test t1 t1 b b 254 9 0 Y 2176 0 83
! def test t1 t1 c c 254 3 0 Y 384 0 83
a b c
Y NULL NULL
drop table t1;
-------------------------------------------------------
This is identical to the reports I got on _all_ of our build platforms on which
I tried the 'embedded' tests:
'aix43', 'barney', 'buildqnx', 'cane', 'hpux11'(32), 'intelxeon', 'pegasos1',
'powermacg4', 'powermacg5', 'sol9x86', 'sunfire100b', 'sunfire100c'.
How to repeat:
Run the test suite with '--embedded-server'.