Support float-like tuple indices in offset_of!()

[est31]

Supports invocations like offset_of!((((), ()), ()), 0.0). This 0.0 gets tokenized as float literal, so it has to be broken up again.

The code that did the breaking up was returning a finished Expr, while we need a Ident, so this PR splits up the parse_expr_tuple_field_access_float function into:

• a function that breaks up the float literal (similar to TokenKind::break_two_token_op, but we do access the parser during this splitting operation, so we keep it as an inherent function on the parser)
• and a function that constructs an Expr from it

The former we can then re-use in offset_of parsing. The edge cases especially involving whitespaces are tricky so this adds a bunch of new tests as well.

fixes #112204

[rustbot]

r? @petrochenkov

(rustbot has picked a reviewer for you, use r? to override)

[est31]

Note that this isn't supported by the offset_of!() macro's declaration:

pub macro offset_of($Container:ty, $($fields:tt).+ $(,)?) {
    builtin # offset_of($Container, $($fields).+)
}

Does it make sense to change it to:

pub macro offset_of($Container:ty, $($fields:tt)+) {
    builtin # offset_of($Container, $($fields)+)
}

and then let the builtin syntax deal with it?

This pattern is supported for normal tuple field access.

[est31]

Hmm yeah that's a bit complicated to do because it means that the builtin offset_of needs to get way better when it comes to errors... maybe it would be good for a follow up PR?

[petrochenkov]

If the macro arguments weren't somewhat important as documentation, I'd rather go with

macro offset_of($($tt:tt)+) { ... }

and do all the work in the parser.

[petrochenkov]

@bors r+

[bors]

📌 Commit 9fb266b has been approved by petrochenkov

It is now in the queue for this repository.

[bors]

⌛ Testing commit 9fb266b with merge 43062c4...

[bors]

☀️ Test successful - checks-actions
Approved by: petrochenkov
Pushing 43062c4 to master...

[rust-timer]

Finished benchmarking commit (43062c4): comparison URL.

Overall result: no relevant changes - no action needed

@rustbot label: -perf-regression

Instruction count

This benchmark run did not return any relevant results for this metric.

Max RSS (memory usage)

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.9%	[0.4%, 1.3%]	2
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.9%	[0.4%, 1.3%]	2

Cycles

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-2.4%	[-2.4%, -2.4%]	1
Improvements ✅ (secondary)	-2.1%	[-2.1%, -2.1%]	1
All ❌✅ (primary)	-2.4%	[-2.4%, -2.4%]	1

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 649.669s -> 649.217s (-0.07%)