feat(expr): switch to fancy-regex crate & update the original version

[xzhseh]

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

What's changed and what's your intention?

resolve #12119

Checklist

• I have written necessary rustdoc comments
• I have added necessary unit tests and integration tests
• 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

The fancy-regex adds support for advanced regex features like back-reference & positive/negative lookahead, may need to update the documentation according to this.

[xzhseh]

The expression (regex) in fancy-regex will be essentially parsed to a ExprTree when building by RegexBuilder, thus we do not need to explicitly specify -i flag, but to integrate it with the input pattern.
Also, we support back-reference and positive/negative lookahead in the regex now, which means the following query should be valid at present:

query T
select regexp_replace('foobarbaz', 'a(?=r)', 'X');
----
foobXrbaz

[xzhseh]

Part of Regex::new_options is as below:

fn new_options(options: RegexOptions) -> Result<Regex> {
      let raw_tree = Expr::parse_tree(&options.pattern)?;

      // wrapper to search for re at arbitrary start position,
      // and to capture the match bounds
      let tree = ExprTree {
          expr: Expr::Concat(vec![
              Expr::Repeat {
                  child: Box::new(Expr::Any { newline: true }),
                  lo: 0,
                  hi: usize::MAX,
                  greedy: false,
              },
              Expr::Group(Box::new(raw_tree.expr)),
          ]),
          ..raw_tree
      };
...
}

[codecov]

Codecov Report

Merging #12329 (591ac5d) into main (0032145) will decrease coverage by 0.01%.
The diff coverage is 0.00%.

@@            Coverage Diff             @@
##             main   #12329      +/-   ##
==========================================
- Coverage   69.86%   69.86%   -0.01%     
==========================================
  Files        1417     1417              
  Lines      235501   235505       +4     
==========================================
- Hits       164541   164536       -5     
- Misses      70960    70969       +9     

Flag	Coverage Δ
rust	69.86% <0.00%> (-0.01%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files Changed	Coverage Δ
src/expr/src/table_function/regexp_matches.rs	32.00% <0.00%> (-1.34%)	⬇️
src/expr/src/vector_op/regexp.rs	17.24% <0.00%> (-0.19%)	⬇️

... and 4 files with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[TennyZhuang]

Are there some inputs can make the expression panic?

[TennyZhuang]

Test cases LGTM

[wangrunji0408]

LGTM

[xzhseh]

Are there some inputs can make the expression panic?

Yes, like if the input pattern can not be parsed when trying to RegexBuilder::build(), an Error will be returned.
And in the current implementation we just simply call unwrap() on it, which will cause panic in this case.
We could definitely make a wrapper to handle this if necessary.

The below is all the Error that may be triggered during the backtracking regex approach.

/// An error as the result of parsing, compiling or running a regex.
#[derive(Debug)]
pub enum Error {
    /// An error as a result of parsing a regex pattern, with the position where the error occurred
    ParseError(ParseErrorPosition, ParseError),
    /// An error as a result of compiling a regex
    CompileError(CompileError),
    /// An error as a result of running a regex
    RuntimeError(RuntimeError),

    /// This enum may grow additional variants, so this makes sure clients don't count on exhaustive
    /// matching. Otherwise, adding a new variant could break existing code.
    #[doc(hidden)]
    __Nonexhaustive,
}