Fix teardown in SQLite File spec

[Akirathan]

Pull Request Description

Fixes teardown of SQLite spec. There used to be only connection.close, but we also have to call connection.drop_table for every created table.

This causes problems only in [SQLite File] tests. These are backed by a sqlite file and some tables are persistent in that table. It is possible, that before tests are run, this file is non-empty and contains garbage from previous runs.

Important Notes

Should fix https://github.com/enso-org/enso/actions/runs/7724599547/job/21057063900#step:10:7243 that was triggered when Table_Tests were run on a non-clean runner.

Checklist

Please ensure that the following checklist has been satisfied before submitting the PR:

• The documentation has been updated, if necessary.
• Screenshots/screencasts have been attached, if there are any visual changes. For interactive or animated visual changes, a screencast is preferred.
• All code follows the
  Scala,
  Java,
  and
  Rust
  style guides. In case you are using a language not listed above, follow the Rust style guide.
• All code has been tested:
  • Unit tests have been written where possible.
  • If GUI codebase was changed, the GUI was tested when built using ./run ide build.

[radeusgd]

This causes problems only in [SQLite File] tests. These are backed by a sqlite file and some tables are persistent in that table. It is possible, that before tests are run, this file is non-empty and contains garbage from previous runs.

While cleaning up seems the right thing to do, for this particular case it seems that a more robust solution would be to ensure that file.delete_if_exists is called before the first SQLite File test is run - that would ensure that even if a previous run did not clean something up (e.g. because it failed fatally with OOM or was terminated by a signal), the new run still gets a clean working environment. Cleaning afterwards is good and recommended, but ensuring the initial environemnt is clean seems even better.

Can we add a delete_if_exists to init?

[Akirathan]

Can we add a delete_if_exists to init?

@radeusgd Sure, good idea. Added in f4163b7

[Akirathan]

Deleting the SQLite file before a connection is made is not a good idea. It causes failures like https://github.com/enso-org/enso/actions/runs/7741741366/job/21109332981?pr=8918#step:8:15352 for tests that create multiple connections. Let's revert f4163b7 and merge the PR. We will address other issues related to teardowns once some more transient failures occur.

GitHub

Fix teardown in SQLite File spec · enso-org/enso@f5587b6

Hybrid visual and textual functional programming. Contribute to enso-org/enso development by creating an account on GitHub.

[radeusgd]

Deleting the SQLite file before a connection is made is not a good idea. It causes failures like enso-org/enso/actions/runs/7741741366/job/21109332981?pr=8918#step:8:15352 for tests that create multiple connections. Let's revert f4163b7 and merge the PR. We will address other issues related to teardowns once some more transient failures occur.

GitHub**Fix teardown in SQLite File spec · enso-org/enso@f5587b6**Hybrid visual and textual functional programming. Contribute to enso-org/enso development by creating an account on GitHub.

I still think it is a good idea, as argued before. It just has to be done at the proper place - it should be done once before all SQLite tests run, not on every connection being established.