Pullup ticket #3208 - requested by taca

databases/mysql51-client: security update
databases/mysql51-server: security update

Revisions pulled up:
- databases/mysql51-client/Makefile.common	1.10
- databases/mysql51-client/distinfo		1.6
- databases/mysql51-client/patches/patch-ap	1.2
- databases/mysql51-server/PLIST		1.7
- databases/mysql51-server/Makefile		1.6
- databases/mysql51-server/distinfo		1.8-1.9
- databases/mysql51-server/patches/patch-ae	1.2
- databases/mysql51-server/patches/patch-aj	1.2
- databases/mysql51-server/patches/patch-ak	1.2
- databases/mysql51-server/patches/patch-ay	1.3
- databases/mysql51-server/patches/patch-az	1.1
- databases/mysql51-server/patches/patch-ba	1.1
- databases/mysql51-server/patches/patch-bb	1.1
---
Module Name:	pkgsrc
Committed By:	taca
Date:		Wed Aug 11 23:18:04 UTC 2010

Modified Files:
	pkgsrc/databases/mysql51-client: Makefile.common distinfo
	pkgsrc/databases/mysql51-client/patches: patch-ap
	pkgsrc/databases/mysql51-server: PLIST distinfo
	pkgsrc/databases/mysql51-server/patches: patch-aj patch-ak
Removed Files:
	pkgsrc/databases/mysql51-server/patches: patch-ay

Log Message:
Update mysql51-client and mysql51-server to 5.1.49.

Please refer http://dev.mysql.com/doc/refman/5.1/en/news-5-1-49.html
for full changes.

* InnoDB Plugin has been upgraded to version 1.0.10. This version is
  considered of General Availability (GA) quality.

  In this release, the InnoDB Plugin is included in source and binary
  distributions, except RHEL3, RHEL4, SuSE 9 (x86, x86_64, ia64), and
  generic Linux RPM packages. It also does not work for FreeBSD 6 and
  HP-UX or for Linux on generic ia64.

Bugs fixed:

* Security Fix: After changing the values of the innodb_file_format or
  innodb_file_per_table configuration parameters, DDL statements could
  cause a server crash. (Bug#55039)

* Security Fix: Joins involving a table with with a unique SET column
  could cause a server crash. (Bug#54575)

* Security Fix: Incorrect handling of NULL arguments could lead to a
  crash for IN() or CASE operations when NULL arguments were either
  passed explicitly as arguments (for IN()) or implicitly generated by
  the WITH ROLLUP modifier (for IN() and CASE). (Bug#54477)

* Security Fix: A malformed argument to the BINLOG statement could
  result in Valgrind warnings or a server crash. (Bug#54393)

* Security Fix: Use of TEMPORARY InnoDB tables with nullable columns
  could cause a server crash. (Bug#54044)

* Security Fix: The server could crash if there were alternate reads
  from two indexes on a table using the HANDLER interface. (Bug#54007)

* Security Fix: Using EXPLAIN with queries of the form SELECT
  ... UNION ... ORDER BY (SELECT ... WHERE ...) could cause a server
  crash. (Bug#52711)

* Security Fix: LOAD DATA INFILE did not check for SQL errors and sent
  an OK packet even when errors were already reported. Also, an assert
  related to client-server protocol checking in debug servers
  sometimes was raised when it should not have been. (Bug#52512)
---
Module Name:	pkgsrc
Committed By:	taca
Date:		Fri Aug 13 14:25:35 UTC 2010

Modified Files:
	pkgsrc/databases/mysql51-server: Makefile distinfo
	pkgsrc/databases/mysql51-server/patches: patch-ae
Added Files:
	pkgsrc/databases/mysql51-server/patches: patch-ay patch-az patch-ba
	    patch-bb

Log Message:
Fix some my_time_t (long) v.s. time_t problem introduced MySQL 5.1.49.
This problem results mysqld to exit on start up.

		5.1/i386	5.1/amd64	5.99.38/i386	5.99.38/amd64
my_time_t	int32_t		int64_t		int32_t		int64_t
time_t		int32_t		int32_t		int64_t		int64_t

I confirmed to mysqld running on these four case except 5.99.38/i386.

Bump PKG_REVISION.

1 files changed, 25 insertions, 0 deletions

diff --git a/databases/mysql51-server/patches/patch-bb b/databases/mysql51-server/patches/patch-bb
new file mode 100644
index 00000000000..18939397bae
--- /dev/null
+++ b/databases/mysql51-server/patches/patch-bb
@@ -0,0 +1,25 @@
+$NetBSD: patch-bb,v 1.1.2.2 2010/08/19 16:42:18 tron Exp $
+
+* if long is int64_t and time_t is int32_t, "(time_t) MY_TIME_T_MAX"
+  results -1 and server_start_time is always parsed as exceeding limit.
+  So, use TIMESTAMP_MAX_VALUE which is INT_MAX32.
+* if time_t is int32_t, comparsion with TIMESTAMP_MAX_VALUE is always
+  true here.  So, limit the comparsion to other cases.
+
+--- sql/sql_class.h.orig	2010-07-09 12:34:57.000000000 +0000
++++ sql/sql_class.h
+@@ -2031,8 +2031,12 @@ public:
+   }
+   /*TODO: this will be obsolete when we have support for 64 bit my_time_t */
+   inline bool	is_valid_time() 
+-  { 
+-    return (start_time < (time_t) MY_TIME_T_MAX); 
++  {
++#if defined(TIME_T_UNSIGNE) || (SIZEOF_TIME_T > 4)
++    return (start_time < TIMESTAMP_MAX_VALUE); 
++#else
++    return 1;
++#endif
+   }
+   void set_time_after_lock()  { utime_after_lock= my_micro_time(); }
+   ulonglong current_utime()  { return my_micro_time(); }