[Merged by Bors] - Implement ResolveLocale helper

[NorbertGarfield]

This Pull Request implements ResolveLocale abstract method. It is required for further InitializeDateTimeFormat development.

It changes the following:

• Adds several helpers to operate with locale extensions
• Adds DefaultLocale placeholder
• Implements BestAvailableLocale and locale matchers
• Implements UnicodeExtensionsComponents
• Introduces testing

[codecov]

Codecov Report

Merging #2036 (6a923dc) into main (20f72d1) will increase coverage by 0.14%.
The diff coverage is 63.00%.

@@            Coverage Diff             @@
##             main    #2036      +/-   ##
==========================================
+ Coverage   44.16%   44.31%   +0.14%     
==========================================
  Files         212      212              
  Lines       18812    18912     +100     
==========================================
+ Hits         8309     8380      +71     
- Misses      10503    10532      +29     

Impacted Files	Coverage Δ
boa_engine/src/builtins/intl/mod.rs	56.00% <63.00%> (+28.00%)	⬆️
boa_engine/src/realm.rs	38.46% <0.00%> (-7.70%)	⬇️
boa_engine/src/syntax/lexer/template.rs	36.36% <0.00%> (-2.77%)	⬇️
boa_engine/src/environments/compile.rs	40.62% <0.00%> (-1.05%)	⬇️
boa_engine/src/builtins/string/mod.rs	63.55% <0.00%> (+0.07%)	⬆️
boa_engine/src/builtins/number/mod.rs	76.66% <0.00%> (+0.08%)	⬆️
boa_engine/src/object/internal_methods/mod.rs	68.39% <0.00%> (+0.43%)	⬆️
boa_engine/src/string.rs	59.72% <0.00%> (+0.69%)	⬆️
boa_engine/src/builtins/function/mod.rs	30.76% <0.00%> (+0.85%)	⬆️
...ntax/parser/expression/left_hand_side/arguments.rs	37.50% <0.00%> (+2.50%)	⬆️
... and 2 more

---

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 20f72d1...6a923dc. Read the comment docs.

[raskad]

Thanks for the contribution.

I commented on some things but I have some more general comments here.

Instead of working with Strings, it would probably be better to use JsStrings. We plan to switch to UTF-16 strings and these would need to be converted too.

There are some data structures in this code that the spec defines as Records. For example I commented on the return value of ResolveLocale. These are currently either JsValues or JsObjects. I think it would be much better to have them as distinct struct types.

Also available_locales and requested_locales are JsValues but you convert them to objects and then use them as array-like objects. The spec specifies to use a List, so it would be probably better to use a Vec<SomeType> instead.

[NorbertGarfield]

@raskad do I need to change '&str' to '&JsString' as well?

[raskad]

@raskad do I need to change '&str' to '&JsString' as well?

In theory I think yes. But think we can probably fix that later. Currently the most correct thing we could do is to use core::str::encode_utf16 everywhere. For example any (String|str).len() usage is in theory wrong, because it uses utf8. We would need to use (String|str).encode_utf16().count() to get the utf16 length.

But I think this is probably not a top priority for this right now. Let's first get some working code, get some 262 tests to pass and then we can go from there ;)

[raskad]

Just got my one comment left. Really nice work. Thank you for taking this on!

[raskad]

Thank you, really nice work! I'm really exited for the first intl test to pass ;)

[Razican]

Looks very good, just some documentation and Debug derives missing, along with some coding standards improvements.

[HalidOdat]

Looks good to me! :)

[Razican]

It looks great now!

[Razican]

bors r+

[bors]

Pull request successfully merged into main.

Build succeeded:

• Clippy
• Documentation
• Examples
• Rustfmt
• Tests on Linux
• Tests on MacOS
• Tests on Windows