Allow underscores in number literals

[wackbyte]

Allows basically the same things as Rust.
The underscores are removed during codegen.
Closes #7.

[codecov]

Codecov Report

Merging #20 (23a47fb) into main (61f026e) will increase coverage by 0.03%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main      #20      +/-   ##
==========================================
+ Coverage   76.76%   76.79%   +0.03%     
==========================================
  Files         105      105              
  Lines        6076     6086      +10     
==========================================
+ Hits         4664     4674      +10     
  Misses       1412     1412              

Impacted Files	Coverage Δ
crates/ditto-checker/src/typechecker/pre_ast.rs	100.00% <100.00%> (ø)
...rates/ditto-checker/src/typechecker/tests/float.rs	100.00% <100.00%> (ø)
crates/ditto-checker/src/typechecker/tests/int.rs	100.00% <100.00%> (ø)
crates/ditto-codegen-js/src/convert.rs	97.33% <100.00%> (ø)
crates/ditto-cst/src/parser/expression.rs	100.00% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 61f026e...23a47fb. Read the comment docs.

[jmackie]

Woah thanks so much @wackbyte, I really wasn't expecting any contributions this soon ❤️

Interesting that you removed the underscores during the JavaScript codegen. My first thought was to remove them when converting from CST to AST - that way the removal can happen in one place rather than in all the code generators (there will eventually be more than one...). But then again, maybe it makes sense to handle it on a per-codegen basis?

Not a big deal at all, happy to merge this PR as it is, but would be keen to hear your thoughts!

[wackbyte]

Doing it during AST conversion does make more sense, actually. I think I tried to do it but wasn't able to find where it was in the code 😅

[jmackie]

Doing it during AST conversion does make more sense, actually. I think I tried to do it but wasn't able to find where it was in the code 😅

Ahh that's my bad. The conversion from CST to an intermediate/pre-AST happens here, then there's another conversion (as part of type-checking) that happens here.

I'm happy to make the change if you want? Let me know 🙂