MySQL Bugs: #58632: Build failure in 5.5.7 when building without partitioning

Bug #58632	Build failure in 5.5.7 when building without partitioning
Submitted:	1 Dec 2010 13:49	Modified:	14 Sep 2011 17:50
Reporter:	Alexey Kopytov	Email Updates:
Status:	Can't repeat	Impact on me:	None
Category:	MySQL Server: Compiling	Severity:	S3 (Non-critical)
Version:	5.5.7-rc	OS:	Any
Assigned to:		CPU Architecture:	Any

Description:
I get the following build errors in 5.5.7 when building without partitioning:

/home/user/src//mysql-5.5.7-rc/sql/ha_partition.h: In member function ‘virtual void ha_partition::lock_auto_increment()’:
/home/user/src//mysql-5.5.7-rc/sql/ha_partition.h:942: error: ‘struct TABLE_SHARE’ has no member named ‘ha_part_data’
/home/user/src//mysql-5.5.7-rc/sql/ha_partition.h: In member function ‘virtual void ha_partition::unlock_auto_increment()’:
/home/user/src//mysql-5.5.7-rc/sql/ha_partition.h:955: error: ‘struct TABLE_SHARE’ has no member named ‘ha_part_data’
/home/user/src//mysql-5.5.7-rc/sql/ha_partition.h: In member function ‘virtual void ha_partition::set_auto_increment_if_higher(Field*)’:
/home/user/src//mysql-5.5.7-rc/sql/ha_partition.h:966: error: ‘struct TABLE_SHARE’ has no member named ‘ha_part_data’
/home/user/src//mysql-5.5.7-rc/sql/ha_partition.h:967: error: ‘struct TABLE_SHARE’ has no member named ‘ha_part_data’
make[2]: *** [sql/CMakeFiles/sql.dir/sql_partition_admin.cc.o] Error 1
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [sql/CMakeFiles/sql.dir/all] Error 2
make: *** [all] Error 2

The build command was:

cmake . -DCMAKE_BUILD_TYPE=Debug -DWITH_PARTITION_STORAGE_ENGINE=0; make

This is a regression from 5.5.6-rc.

How to repeat:
cmake . -DCMAKE_BUILD_TYPE=Debug -DWITH_PARTITION_STORAGE_ENGINE=0; make

Thank you for the bug report.

[ 98%] Building CXX object sql/CMakeFiles/sql.dir/sql_partition_admin.cc.o                                                                             
In file included from /home/miguel/bzr/5.5test/sql/sql_partition_admin.cc:21:0:                                                                        
/home/miguel/bzr/5.5test/sql/ha_partition.h: In member function ‘virtual void ha_partition::lock_auto_increment()’:
/home/miguel/bzr/5.5test/sql/ha_partition.h:938:20: error: ‘struct TABLE_SHARE’ has no member named ‘ha_part_data’
In file included from /home/miguel/bzr/5.5test/sql/sql_partition_admin.cc:21:0:
/home/miguel/bzr/5.5test/sql/ha_partition.h:942:45: error: ‘struct TABLE_SHARE’ has no member named ‘ha_part_data’
/home/miguel/bzr/5.5test/sql/ha_partition.h: In member function ‘virtual void ha_partition::unlock_auto_increment()’:
/home/miguel/bzr/5.5test/sql/ha_partition.h:947:20: error: ‘struct TABLE_SHARE’ has no member named ‘ha_part_data’
/home/miguel/bzr/5.5test/sql/ha_partition.h:955:47: error: ‘struct TABLE_SHARE’ has no member named ‘ha_part_data’
/home/miguel/bzr/5.5test/sql/ha_partition.h: In member function ‘virtual void ha_partition::set_auto_increment_if_higher(Field*)’:
/home/miguel/bzr/5.5test/sql/ha_partition.h:964:20: error: ‘struct TABLE_SHARE’ has no member named ‘ha_part_data’
/home/miguel/bzr/5.5test/sql/ha_partition.h:966:28: error: ‘struct TABLE_SHARE’ has no member named ‘ha_part_data’
/home/miguel/bzr/5.5test/sql/ha_partition.h:967:20: error: ‘struct TABLE_SHARE’ has no member named ‘ha_part_data’
make[2]: *** [sql/CMakeFiles/sql.dir/sql_partition_admin.cc.o] Error 1
make[1]: *** [sql/CMakeFiles/sql.dir/all] Error 2
make: *** [all] Error 2
[miguel@hegel 5.5test]$

I tried to reproduce it, using the commands given, and failed:
The build succeeded. Especially, the object module
"./sql/CMakeFiles/sql.dir/sql_partition_admin.cc.o"
was created.

Code version: 5.5.17
joerg@trift:/V55/work-5.5$ bzr version-info
revision-id: vasil.dimov@oracle.com-20110909071254-jb86bz87bx33zyje
date: 2011-09-09 10:12:54 +0300
build-date: 2011-09-14 19:47:01 +0200
revno: 3528
branch-nick: work-5.5

My environment: Linux (Debian 5.0), x86_64,
gcc (Debian 4.3.2-1.1) 4.3.2

As the OS is set to "Any", I guess this got fixed by some in-between code change.

If the problem still occurs, please specify the platform and other details and re-open the bug.

That said:
I don't understand why a problem in compiling some server code
on *any* platform is filed to the build team lead.
The developer(s) working on the respective files seem much more appropriate receivers IMNSHO.