-
Notifications
You must be signed in to change notification settings - Fork 4
Sakilla restore fix #16
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: master
Are you sure you want to change the base?
Conversation
…endencies among MetaSql objects more efficient This might fix order of objects during `Sakilla` schema restore process
@rultor merge, tests= |
This comment has been minimized.
This comment has been minimized.
@rocket-3 Oops, I failed. You can see the full log here (spent 4min)
|
There is no View Further, there will be a more detailed investigation. |
In schema Need to implement fetching aggregate functions next. Note: there could be window functions also |
Fix DBAdapterPostgres errors on single object fetching functions Removed some unnecessary messages parts from console output
OK, aggregate functions handling on Postgress is implemented |
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
@rocket-3 Oops, I failed. You can see the full log here (spent 3min)
|
Oops! Not the same error I have locally, but seems like there is no way changing table columns withoud dropping views... Trying again with clean target database. |
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
@rocket-3 Oops, I failed. You can see the full log here (spent 3min)
|
OK, got the same error I had locally Now the current picture:
|
…o DbObjectNameInSqlPresenceTest.java with test Add dropping existing in files and database both MetaFunction, MetaProcedure, MetaView, MetaTrigger before actually dropping unneeded and restoring changed and added objects
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
@rocket-3 Oops, I failed. You can see the full log here (spent 3min)
|
Fix drop/restore procedure in Postgres Fix drop/restore function/aggregate function in Postgres Fix drop trigger in Postgres
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
@rocket-3 Oops, I failed. You can see the full log here (spent 7min)
|
Seems like sorting (ordering mechanism) is OK, but... In The restore order is As you can see, when dropping constraints before restore, dbgit uses target schema dependencies instead of source...
|
… database related to .dbgit repo Add IT extension CharsDbGitIgnoreWithTypes.java to write .dbignore file with custom type extensions and without .csv (for faster execution of tests when data non needed at the moment) Enhance DbGitIntegrationTestBasic.dbToDbRestoreWorksWithCustomTypes with dropping target schema (for having the same result of each execution everywhere) Add some logging on dropping table constraints to be able to see the actual drop order and actual tables dependencies at the moment
…en ordering instead of _target_) Add `Cactoos` lib to project for EO approach learning
This comment has been minimized.
This comment has been minimized.
OK, now need to specify columns explicitly on CREATE TABLE next |
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
@rocket-3 Oops, I failed. You can see the full log here (spent 18min)
|
… when network or database is temporary unavailable
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
@rocket-3 Oops, I failed. You can see the full log here (spent 45min)
|
There are a bug in DbAdapterPostgres, that can't detect PK on a column properly, when the column is an FK at the same time. This causes emtpy delete statements on restrore data of table with such columns. That's why we see duplicate key error in log. |
…stomTypes' Change scenario to cover 'empty_db -> pagilla <-> dvdrental' sequences of restore
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
@rocket-3 Oops, I failed. You can see the full log here (spent 29min)
|
DBRestoreTableDataPostgres messages cleanup
The reason of this error is when we find 'affected' tables in database and add them to update list when it's not needed. This worked well until there was no case with checkout from schema with partitioned table A to schema with non-partitioned version on A. |
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
@rocket-3 Oops, I failed. You can see the full log here (spent 13min)
|
Seems like something's changing existing and restore tables descriptions, so they have all their fields not PK. BackupAdapter does this some way. Need investigate. |
Alter heuristic dependencies searching mechanism in order to find dependencies among MetaSql objects more efficient
This might fix order of objects during
Sakilla
schema restore process