Commit graph

19 commits

Author SHA1 Message Date
Ramil Kalimullin
cf6c7262d0 Fix for bug#35383: binlog playback and replication breaks
due to name_const substitution

Problem:
"In general, statements executed within a stored procedure
are written to the binary log using the same rules that
would apply were the statements to be executed in standalone
fashion. Some special care is taken when logging procedure
statements because statement execution within procedures
is not quite the same as in non-procedure context".

For example, each reference to a local variable in SP's
statements is replaced by NAME_CONST(var_name, var_value).
Queries like
"CREATE TABLE ... SELECT FUNC(local_var ..."
are logged as
"CREATE TABLE ... SELECT FUNC(NAME_CONST("local_var", var_value) ..."
that leads to differrent field names and
might result in "Incorrect column name" if var_value is long enough.

Fix: in 5.x we'll issue a warning in such a case.
In 6.0 we should get rid of NAME_CONST().

Note: this issue and change should be described in the documentation
("Binary Logging of Stored Programs").
2009-03-25 20:48:10 +04:00
Ramil Kalimullin
dc6a5ff899 Fix for bug#39182: Binary log producing incompatible character set query
from stored procedure. 

Problem: we replace all references to local variables in stored procedures     
with NAME_CONST(name, value) logging to the binary log. However, if the
value's collation differs we might get an 'illegal mix of collation'           
error as we don't pass the collation to the function.

Fix: pass the value's collation to NAME_CONST().

Note: actually we should pass to NAME_CONST() the value's derivation as well.
It's impossible without the parser modifying. Now we always set the 
derivation to DERIVATION_IMPLICIT, the same as local variables have.
2008-10-01 14:48:47 +05:00
davi@mysql.com/endora.local
24b9abf36d Bug#33798 prepared statements improperly handle large unsigned ints
The unsignedness of large integer user variables was not being
properly preserved when feeded to prepared statements. This was
happening because the unsigned flags wasn't being updated when
converting the user variable is converted to a parameter.

The solution is to copy the unsigned flag when converting the
user variable to a parameter and take the unsigned flag into
account when converting the integer to a string.
2008-02-08 08:55:55 -02:00
tnurnberg@mysql.com/sin.intern.azundris.com
3dfc83f1a1 Bug#22540: Incorrect value in column End_log_pos of SHOW BINLOG EVENTS using InnoDB
dollin' it up for Guilhem ;) -- test streamlined,
better comments, faster code, add'l assert.
2007-07-09 08:11:38 +02:00
tnurnberg@sin.intern.azundris.com
2fbd3a2eca Bug#22540: Incorrect value in column End_log_pos of SHOW BINLOG EVENTS using InnoDB
minor fixes to appease pushbuild.
2007-07-02 07:13:40 +02:00
tnurnberg@sin.intern.azundris.com
1ac88a2b41 Bug#22540: Incorrect value in column End_log_pos of SHOW BINLOG EVENTS using InnoDB
previous correction didn't. make sure "tail" is fixed up
when filling cache several times; rework formulae.
2007-06-30 02:30:42 +02:00
tnurnberg@sin.intern.azundris.com
25af844f6f Bug #22540: Incorrect value in column End_log_pos of SHOW BINLOG EVENTS using InnoDB
fix binlog-writing so that end_log_pos is given correctly even
within transactions for both SHOW BINLOG and SHOW MASTER STATUS,
that is as absolute values (from log start) rather than relative
values (from transaction's start).
---
Merge tnurnberg@bk-internal.mysql.com:/home/bk/mysql-5.0-maint
into  sin.intern.azundris.com:/home/tnurnberg/22540/50-22540
2007-06-25 11:34:23 +02:00
tsmith@quadxeon.mysql.com
891116551e binlog.result, binlog.test:
Post-merge fix: replace xid=* with XID to isolate from number of transactions
2007-06-15 23:57:53 +02:00
tsmith@quadxeon.mysql.com
87fdad4ce8 Merge tsmith@bk-internal.mysql.com:/home/bk/mysql-5.0-build
into  quadxeon.mysql.com:/benchmarks/ext3/TOSAVE/tsmith/bk/maint/jun14/50
2007-06-14 23:23:30 +02:00
msvensson@pilot.(none)
9ed0e74db3 Bug#28369 rpl test cases fail with binlog disabled 2007-06-08 07:16:36 +02:00
msvensson@pilot.(none)
1afb1c8ddf Bug#28369 rpl test cases fail with binlog disabled
- Turn on binlog only for tests that need it
 - Skip those tests if --skip-binlog is used
2007-06-07 19:18:01 +02:00
df@pippilotta.erinye.com
6e7585a011 fix test case 2007-05-31 14:05:07 +02:00
df@pippilotta.erinye.com
779a055ed0 some test cases cannot run with --skip-log-bin 2007-05-31 09:34:24 +02:00
monty@mysql.com
dd1d40f8f8 Fixed new introduced bug in binlog.test with --ps-protocol 2006-02-25 13:24:18 +02:00
monty@mysql.com
82b77cdd90 Fixes to embedded server to be able to run tests with it
(Needed for "list of pushes" web page and autopush)
2006-02-24 18:34:15 +02:00
dlenev@brandersnatch.localdomain
d8d507926d Fixed binlog.test failure which occured after patch implementing
WL#874 "Extended LOAD DATA" was pushed.
2005-03-16 23:12:27 +03:00
serg@serg.mylan
b8288584e9 innodb-dependent test moved from rpl_rotate_logs.test to binlog.test 2005-02-23 20:35:59 +01:00
serg@serg.mylan
383c6e312a don't log BEGIN in auto-commit mode
correct end_log_pos for Xid_log_event
2005-02-23 18:26:49 +01:00
serg@serg.mylan
da13ab086d write "COMMIT" into binlog if there's no Xid at the end of transaction
new tests
2005-02-23 14:55:16 +01:00