Hytak

5 ms·

1.
by Hytak 2y ago
What I think is that you are confusing the schema_version and user_version pragmas. user_version is what you would use to keep track of which migrations have run and it can be safely updated (or not updated) without corrupting the database.
2.
by Hytak 2y ago
manually updating the schema_version can lead to database corruption https://www.sqlite.org/pragma.html#pragma_schema_version If you are willing to risk corrupting the database, then yes you can trick rust-query
3.
by Hytak 2y ago
Author here: I just wanted to point out that rust-query does in fact read the schema from the database and compares it to the expected schema. There is no checksumming or comparison of migrations. It is checked efficiently that the schema h
4.
by Hytak 2y ago
sqlite has `STRICT` tables that are statically typed
5.
by Hytak 2y ago
You (and many other commenters) are right that rust-query currently requires downtime to do migrations. For many applications this is fine, but it would still be nice to support zero-downtime migrations Your argument that the application sh
6.
by Hytak 2y ago
Sending row IDs to your frontend has two potential problems: - Row IDs can be reused when the row is deleted. So when the frontend sends a message to the backend to modify that data, it might accidentally modify a different row that was cre
7.
by Hytak 2y ago
Hi, migrations are 1 select statement + `n` insert statement for `n` rows right now. This might be improved to insert in batches in the future without changing the API.
8.
by Hytak 2y ago
rust-query manages migrations and reads the schema from the database to check that it matches what was defined in the application. If at any point the database schema doesn't match the expected schema, then rust-query will panic with a