Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

@edgarrmondragon: Thank you for GH-97, it succeeds without ado. Do you have any idea why this patch fails, that aims to actually leverage the improvements from your pen?

@edgarrmondragon: Thank you for GH-97, it succeeds without ado. Do you have any idea why this patch fails, that aims to actually leverage the improvements from your pen?

You might wanna update the reference for the sample taps:

i.e. set rev to the v0.53 branch or at least a more recent commit

Oh right, thank you so much for reminding me about this. 🙇
I've added 765c80b to bump to sdk v0.53.

Now, I probably need to tackle this to complete the update:

FAILED target_cratedb/tests/test_standard_target.py::test_new_array_column - 
sqlalchemy.exc.UnsupportedCompilationError: Compiler <sqlalchemy_cratedb.compiler.CrateTypeCompiler object at 0x7f0f89825f40> can't render element of type JSONB

Hm. It looks like our pick_best_sql_type is no longer called, that's why it picks the one from PostgresConnector. This is why we submitted the other patch, but apparently, it doesn't resolve the problem completely yet?

Edit: I see that this method is now located in JSONSchemaToPostgres, so I probably need to transfer a bit of model refactoring?

It looks like overriding pick_best_sql_type is no longer needed, but -- of course -- a custom variant of _handle_array_type. After providing a variant for CrateDB with dee44c5, the test suite succeeds.

Thank you again for your swift replies so I was able resolve relevant problems easily. 🙇
I've just released v0.2.0 that builds upon your recent improvements.