Improve escaping of byte, byte str, and c str proc-macro literals

[dtolnay]

This PR changes the behavior of proc_macro::Literal::byte_character (#115268), byte_string, and c_string (#119750) to improve their choice of escape sequences. 3 categories of changes are made:

1. Never use \x00. Always prefer \0, which is supported in all the same places.

2. Never escape \' inside double quotes and \" inside single quotes.

3. Never use \x for valid UTF-8 in literals that permit \u.

The second commit adds tests covering these cases, asserting the old behavior.

The third commit implements the behavior change and simultaneously updates the tests to assert the new behavior.

[rustbot]

r? @fee1-dead

rustbot has assigned @fee1-dead.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[petrochenkov]

I remember we talked about something related in #95343, but don't remember details.
cc #60495 as well.

[fee1-dead]

Given the FCP here and that this is mostly an improvement / not a bug fix I'd say we should wait until it gets stabilized (on beta?)

[dtolnay]

If anything, having an official non-core::str::Utf8Chunks implementation checked in is an advantage because I'll need to use that in proc-macro2. Proc-macro2 is not going to be able to adopt Utf8Chunks as soon as it goes into beta.

[fee1-dead]

I don't think we should add burden of maintainability to a first-party library to ease the burden of maintainability of a third-party library.

[dtolnay]

@rustbot ready

The new implementation compiles with stable Rust 1.79, which contains the stabilization of Utf8Chunks.

[fee1-dead]

LGTM, thanks!

[fee1-dead]

@bors r+

[bors]

📌 Commit 7ddc89e has been approved by fee1-dead

It is now in the queue for this repository.