Make scanner accept based floating point literals

[richcarl]

Similar to based float literals in Ada and hex float literals in C 99, using Erlang's Base#Literal notation (inspired by Ada), for example, 2#0.10101#e8, or 16#fefe.fefe#e-16. The # before the exponent is required. The exponent, like the base, is always a decimal number.

These can be useful for code generators and other cases where you want to preserve an exact bit representation of a float.

Currently missing: documentation, io formatting, list_to_float/2 implementation.

Based on #9104, which improves pinpointing of errors in numerical literals.

[github-actions]

CT Test Results

    2 files     96 suites   1h 13m 13s ⏱️
2 173 tests 2 126 ✅ 47 💤 0 ❌
2 536 runs  2 487 ✅ 49 💤 0 ❌

Results for commit 0afd8d0.

♻️ This comment has been updated with latest results.

To speed up review, make sure that you have read Contributing to Erlang/OTP and that all checks pass.

See the TESTING and DEVELOPMENT HowTo guides for details about how to run test locally.

Artifacts

• Complete CT logs (Download Logs)
• HTML Documentation (Download HTML Docs)
• Windows Installer

// Erlang/OTP Github Action Bot

[richcarl]

Rebased on #9104 after squashing.

[bjorng]

Looks good to me. We do need an EEP for any changes to the language.

[richcarl]

I'm seeing now that in Ada, which is where Erlang got its <Base> # <Digits> notation from, they have a slightly different notation for the exponent separator in a based numeral. Instead of p or P as in C99, they use #E. For example, 16#f.ff#E2 rather than 16#f.ffp2. It might be better to keep following the Ada style rather than creating a mix.

[bjorng]

If you write an EEP soon, I think it is still possible that we could include this improvement in OTP 28.

[richcarl]

If you write an EEP soon, I think it is still possible that we could include this improvement in OTP 28.

I was waiting to hear which form the OTP team favours. The Ada style, or the frankenstyle?

[bjorng]

I was waiting to hear which form the OTP team favours.

OK. I'll try to find out what the team thinks. I've posted a question in our internal forums.

Personally I think we should go for the Ada style.

[richcarl]

Personally I think we should go for the Ada style.

I agree; the # before the exponent makes it clear that the next letter is never part of the mantissa. The C99 form relies on p not being in the allowed a-f, but with the Ada form, Erlang could just accept any base that it already allows for integers.

[richcarl]

EEP PR: erlang/eep#71

[bjorng]

I've got two comments so far; both prefer the Ada style.

[richcarl]

I've got two comments so far; both prefer the Ada style.

Good; that's what I put in the EEP draft. Just need to update the code to match.

[richcarl]

The code now uses Ada style, i.e. 16#ff.fe#e+6.