Changes look good. Test succeeds with fix and fails without.
I have two questions:
- In Bug#33567, Peter says that "users wouldn't enter the same name
twice except in error, so this should be an error." Have you
discussed your solution (to just ignore duplicates) with Peter?
- Is it difficult to test that the backup image is no longer
larger with duplication as reported in BUG#34480?
--
Øystein
Rafal Somla wrote:
> #At file:///ext/mysql/bzr/backup/bug36933/
>
> 2717 Rafal Somla 2008-10-24
> BUG#36933, BUG#34480, BUG#33567 (Backup: snapshot error + crash
if double backup)
>
> Before: When user specified the same database twice in the
BACKUP DATABASE statement,
> it was included twice in the backup image. This lead to a host
of problems.
>
> After: BACKUP DATABASE will ignore duplicates in the database
list. Each database,
> together with its objects, will be included only once in the
produced backup image.
> modified:
> mysql-test/suite/backup/r/backup.result
> mysql-test/suite/backup/t/backup.test
> sql/backup/backup_info.cc
> sql/backup/image_info.cc
> sql/backup/image_info.h
>
> per-file messages:
> mysql-test/suite/backup/t/backup.test
> - Add primary key to t1_blob table to cover the scenario
described in BUG#36933 report.
> - In one of the BACKUP commands, repeat twice a database
containing table with a primary key and a blob.
> sql/backup/backup_info.cc
> When iterating over the list of databases specified for backup,
ignore repeated names.
> sql/backup/image_info.cc
> Mark has_db() method as const.
> sql/backup/image_info.h
> Mark has_db() method as const.
