[Feature Request] Ability for --delete-existing-rows and --skip-schema-files flags when using restore-dump

[orware]

Currently, when making use of the restore-dump sub-command there is not an easy way for the *-schema.sql files to be skipped entirely, which would be separate from the recent --overwrite-tables flag that was recently added in, which primarily allows for existing tables to be dropped and then created again, which does not work if the branch is a production one.

A database dump generated by pscale database dump ... will generally include several files ending in -schema.sql so making use of the proposed new --skip-schema-files flag would simply add in a file filter excluding those files from being processed when making use of restore-dump.

To test a similar effect, you can simply manually remove any files ending in -schema.sql from your database dump folder and this would allow only the files containing the rows to insert to be processed.

Separately, the proposed --delete-existing-rows flag would go hand in hand with this option since it would help automate the loop for users where a DELETE query must be repeated with a LIMIT clause until the table is empty that we have as an alternative to using TRUNCATE when working with production branches. Additionally, this flag could possibly be made aware of whether a branch is a development or production one and use TRUNCATE or the DELETE loop with LIMIT clause depending on what's more appropriate.

Both of the requests above used together would help enable users to use restore-dump against a production branch and make it easier to restore a database dump from a primary production branch into a secondary staging / development branch that happens to also be a production one.

Example usage:

pscale database restore-dump <DATABASE_NAME> <BRANCH_NAME> --dir="/path/to/database/dump" --delete-existing-rows --skip-schema-files --org=<ORGANIZATION_NAME>

There is some risk with exposing flags like the ones proposed here for users, but used appropriately and with care, it should provide useful functionality simplifying the work users may have to go through in order to restore data from a database dump into a production branch if they were available.