feat(sql-udf): better hint display for invalid sql udf definition

[xzhseh]

I hereby agree to the terms of the RisingWave Labs, Inc. Contributor License Agreement.

What's changed and what's your intention?

resolve #14853.

BEFORE

As shown in the above issue.

AFTER

P.S. IMO the cost of constructing Regex when error definition is found is acceptable to display a better hint for user.

Note that the current hint works for invalid column / parameter, function hint will be supported in subsequent PR.

Checklist

• I have written necessary rustdoc comments
• I have added necessary unit tests and integration tests
• I have added test labels as necessary. See details.
• I have added fuzzing tests or opened an issue to track them. (Optional, recommended for new SQL features Sqlsmith: Sql feature generation #7934).
• My PR contains breaking changes. (If it deprecates some features, please create a tracking issue to remove them in the future).
• All checks passed in ./risedev check (or alias, ./risedev c)
• My PR changes performance-critical code. (Please run macro/micro-benchmarks and show the results.)

• My PR contains critical fixes that are necessary to be merged into the latest release. (Please check out the details)

Documentation

• My PR needs documentation updates. (Please use the Release note section below to summarize the impact on users)

Release note

If this PR includes changes that directly affect users or other significant modifications relevant to the community, kindly draft a release note to provide a concise summary of these changes. Please prioritize highlighting the impact these changes will have on users.

[BugenZhao]

The idea is great and it's definitely an improvement to the user experience.

Actually I'm wondering if this could be applied to other statements as well. I believe the ultimate solution could be attaching a span to each AST item during parsing. Databend had a good practice on this (but it could require a quite large amount of work to refactor):

https://medium.com/@databend/the-new-databend-sql-planner-e1b1f028a594

[xzhseh]

The idea is great and it's definitely an improvement to the user experience.

Actually I'm wondering if this could be applied to other statements as well. I believe the ultimate solution could be attaching a span to each AST item during parsing. Databend had a good practice on this:

https://medium.com/@databend/the-new-databend-sql-planner-e1b1f028a594

Indeed, let me do some further investigations on this and see if we can come up with an ultimate solution. (And I can work on this constantly 😄)

[wangrunji0408]

The idea is great and it's definitely an improvement to the user experience.
Actually I'm wondering if this could be applied to other statements as well. I believe the ultimate solution could be attaching a span to each AST item during parsing. Databend had a good practice on this:
https://medium.com/@databend/the-new-databend-sql-planner-e1b1f028a594

Indeed, let me do some further investigations on this and see if we can come up with an ultimate solution. (And I can work on this constantly 😄)

We can learn from syn on how to gracefully attach span to each AST node.

[BugenZhao]

We can learn from syn on how to gracefully attach span to each AST node.

Could you elaborate more? Do you mean the span in the proc_macro crate?

[wangrunji0408]

Generally LGTM!

[wangrunji0408]

We can learn from syn on how to gracefully attach span to each AST node.

Could you elaborate more? Do you mean the span in the proc_macro crate?

IIUC, in syn, each AST node is recursively composed by a series of tokens (as leaf nodes of the AST). Each token provides its own span. And the span of each AST node can be computed as the join of all children's by derive macro.

However, adding span to sqlparser is still a big project. Someone has proposed it several years ago but it has not been landed until now.