RFC: Add recognize_till

[9999years]

This is a variant of repeat_till that recognizes the repeated portion of the input instead of accumulating it.

See: “Recognizing repeat_till0 to () without the suffix”.

I have found this combinator useful in my own code for recognizing ambiguous delimiters.

This code needs to be cleaned up before being merged (we'd probably want a Range argument and an infinite loop check, for example), so consider this a proof of concept. If you're on board with the idea, let me know what the next steps are. Otherwise, feel free to close this PR. Thanks!

[coveralls]

Pull Request Test Coverage Report for Build 9523540454

Details

• 0 of 17 (0.0%) changed or added relevant lines in 1 file are covered.
• 2 unchanged lines in 1 file lost coverage.
• Overall coverage increased (+0.2%) to 41.848%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/combinator/multi.rs	0	17	0.0%

Files with Coverage Reduction	New Missed Lines	%
src/macros/seq.rs	2	85.71%

Totals
Change from base Build 9407425121:	0.2%
Covered Lines:	1309
Relevant Lines:	3128

---

💛 - Coveralls

[epage]

Moved to draft as I take a fairly strict policy to close (or move to draft) PRs that have not formerly been accepted

• This keeps the solution discussion centralized in the Issue and the implementation discussion in the PR
• This focuses us on the solution discussion without getting distracted by the implementation discussion
• I have found that PRs put (intended or not) social pressure on maintainers to "move things forward" because someone has already invested time into this. I tend to see maintainers either give in and burn out or become apathetic to contributions.

[9999years]

@epage That's OK by me. Would you like me to open an issue for recognizing the repeated part of a repeat_till invocation?

[epage]

You are welcome to create an issue from that discussion.