Update(Snowflake): Update tests for concat string function

[fivetran-amrutabhimsenayachit]

Added tests string function: concat

Refer:

• https://docs.snowflake.com/en/sql-reference/functions/concat

• https://docs.snowflake.com/en/sql-reference/functions/concat_ws

[Copilot]

Pull Request Overview

This PR adds test coverage for Snowflake's string concatenation functions (CONCAT and CONCAT_WS) to verify their return type annotations in the optimizer.

Key Changes

• Added 5 new test cases for CONCAT and CONCAT_WS functions in Snowflake dialect
• Tests verify that both functions return VARCHAR type regardless of input parameters

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[georgesittas]

Given Snowflake's docs say:

The input expressions must all be strings, or all be binary values.

Let's make sure we test the binary values as well.

[fivetran-amrutabhimsenayachit]

Given Snowflake's docs say:

The input expressions must all be strings, or all be binary values.

Let's make sure we test the binary values as well.

The base Dialect class has exp.Concat in its TYPE_TO_EXPRESSIONS[VARCHAR] mapping . Thus, CONCAT always returns VARCHAR, regardless of input types BINARY or NULL values.
We will have to override this logic for both CONCAT and CONCAT_WS methods for snowflake.

[georgesittas]

@fivetran-amrutabhimsenayachit have you created a virtual environment for Python? Seems like there may be a mismatch between the ruff version SQLGlot installs vs what you have locally. Try the following:

$ python -m venv .venv
$ source .venv/bin/activate
$ make install-dev
$ make style

This should fix any formatting issues.

The base Dialect class has exp.Concat in its TYPE_TO_EXPRESSIONS[VARCHAR] mapping . Thus, CONCAT always returns VARCHAR, regardless of input types BINARY or NULL values.
We will have to override this logic for both CONCAT and CONCAT_WS methods for snowflake.

Overriding makes sense, but with a quick look I feel like the current implementations are quite complicated. What about using _annotate_by_arg for both of them?

[fivetran-amrutabhimsenayachit]

have you created a virtual environment for Python
@georgesittas , No, I do not have virtual environment for python. Seems like the new functions added have not been formatted properly. I did not run the make style before commit. Thanks for these commands, will try these.

[georgesittas]

I think this was added by mistake– I got rid of it in main. Let's git rebase off of it.

[georgesittas]

I believe we can simply do:

exp.ConcatWs: lambda self, e: self._annotate_by_args(e, "expressions")

Don't think that we need _annotate_concat_ws. Lemme know if I missed something.