Re: Creating complex track changes database - challenge!

[geoff hoffman <geoff@xxxxxxxx>]

I would personally do that separately: write a bash script & cron job that does a schema dump every hour, and (if there are any changes) commits any changes to your schema repository; then you can use Github or Bitbucket web hooks to do stuff with the changeset when it’s pushed. 

https://stackoverflow.com/questions/3878624/how-do-i-programmatically-determine-if-there-are-uncommitted-changes

https://stackoverflow.com/questions/24772591/check-if-git-has-changes-programmatically 

On Feb 26, 2018, at 9:36 AM, Łukasz Jarych <jaryszek@xxxxxxxxx> wrote:

Thank you geoff! 

I think that i will test http://www.liquibase.org/ this one. 

what about setting up trigger to metadata (structural table) to find if column was added for example? 

Best,

Jacek

2018-02-26 16:43 GMT+01:00 geoff hoffman <geoff@xxxxxxxx>:

There’s https://flywaydb.org/

and http://www.liquibase.org/

More: https://dbmstools.com/version-control-tools 

Also, if you know PHP, Laravel database migrations have worked great for us!

https://laravel.com/docs/5.6/migrations

On Feb 26, 2018, at 3:44 AM, Łukasz Jarych <jaryszek@xxxxxxxxx> wrote:

i would like to ask you for help with track changes to my database. 

I am new to PosgtreeSQL but i have to learn it quickly because of my boss. 

I have to:

1. Keep all changes within table including:

-adding rows

-deleting

-editing

2. Save table with specific state and recover specific state (so go back to previous table versions) including comparing tables.

3. Track all DLL and DML changes with possibility to ho back to previous version.