refactor(parser): introduce a compactible layer for parser v2 to call v1

[TennyZhuang]

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

What's changed and what's your intention?

• Introduce ParseV1 as a compatible trait for parser v2
• Write expr_case as an example

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]

👀 May I know the relationship versus #17019?

[TennyZhuang]

👀 May I know the relationship versus #17019?

The PR is based on that.

#17019 replace the underlying input stream of parser v2 from &[TokenWithLocation] to parser_v1::Parser<'a>.

The PR introduce a compatible trait, to allow S: TokenStream export underlying parser_v1::Parser, and call methods on it.

Based on the PR, we can refactor the recursive terms step by step. E.g., we create a fake v2 expr which calls v1 internally, then we can write a real v2 expr_case based on the fake expr.

[wangrunji0408]

Rest LGTM!

[wangrunji0408]

In terms of semantics, this function should start parsing from CASE and the caller should not consume CASE before calling it.

Suggested change

	pub fn expr_case<S>(input: &mut S) -> PResult<Expr>
	where
	S: TokenStream,
	{
	let parse = (
	opt(expr),
	pub fn expr_case<S>(input: &mut S) -> PResult<Expr>
	where
	S: TokenStream,
	{
	let parse = (
	Keyword::CASE,
	opt(expr),

[TennyZhuang]

Unfortunately we have to following the v1's behavior.

Another problem is dispatch! is much more efficient than alt, even in v2, using the style can achieve best performance.

[TennyZhuang]

The only thing I can do is rename it to expr_case_suffix or expr_case_rest, but seems too verbose.

[wangrunji0408]

Okay let's correct this later.

dispatch! can also be used with peek(keyword) which doesn't consume the token. 😄