Add string format tools to library.

[toinehartman]

This PR adds many generic tools that can be used to process/format strings.

TODO

• Extensively document (once we figured out which ones we will actually keep)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 47%. Comparing base (875fe06) to head (8d8de3e).

Additional details and impacted files

@@           Coverage Diff           @@
##              main   #2373   +/-   ##
=======================================
  Coverage       47%     47%           
- Complexity    6560    6571   +11     
=======================================
  Files          780     780           
  Lines        64398   64398           
  Branches      9628    9628           
=======================================
+ Hits         30509   30517    +8     
+ Misses       31569   31555   -14     
- Partials      2320    2326    +6     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[rodinaarssen]

Cool! Just a few remarks and questions.

[DavyLandman]

Pinging @jurgenvinju since he has in the past done some more stuff around formatting & indentation.

I have the feeling we're missing something in this PR, a feature we already have in rascal. (aside from the fact I don't really like all these regexps and char loops)

[DavyLandman]

what if the string contains multiple lines?

[toinehartman]

I think that should be an explicit exception (invalid argument or similar)

[DavyLandman]

this almost looks like pattern matching on strings? (which we only have reasonable support over)

for example:

rascal>visit("abcd") { case str m : println(m); }
abcd
bcd
cd
d

come to think of it, this whole function smells like an parsing automata. Where we build a big state table of all the possible matches and then iterate through all the chars and count the matches based on their state.

In java this would be 20/30 lines, but in rascal we might be missing some primitives (as we don't have a character loop).

[toinehartman]

If we hard-code the set of newline characters (e.g. to all Unicode newline chars), we could write it as a grammar and use the parser generator. Downside (as we discussed) is that all (transitive) imports of this module will trigger generation of a parser. We could also move some of those to a specific Format module.