Remove lots of generics from ty::print

[Nilstrieb]

All of these generics mostly resolve to the same thing, which means we can remove them, greatly simplifying the types involved in pretty printing and unlocking another simplification (that is not performed in this PR): Using &mut self instead of passing self through the return type.

cc @eddyb you probably know why it's like this, just checking in and making sure I didn't do anything bad

r? oli-obk

[eddyb]

cc @eddyb you probably know why it's like this, just checking in and making sure I didn't do anything bad

There's an arguably good reason for it, trouble is that it was kind of mostly made irrelevant during the refactor itself. Keeping the excessive generics was mostly a matter of not having to rewrite everything yet again.

(the reason is building heterogenous ASTs for "pretty-printing" - similar to what @michaelwoerister's initial implementation of a Rust demangler did, but for the "mangling" direction)

I will say that replacing T -> T with &mut T, or worse, {fmt,io}::Write-like &mut self, is probably more relevant long-term than having just one choice of T, but honestly all of this stuff has outlived its usefulness and any future APIs would much easier copy leaf functionality than refactor this API.

(that's the other thing that became clear during that refactor: the trickier aspects didn't need to be tied too strongly to pretty-printing in order to be reusable, and in fact most of the deduplication/sharing ended up being tied to the ty::print::pretty variation, not the more general ty::print)

---

In fewer words: yes, the overgeneric ty::print APIs can be removed as a historical mistake (and barely an improvement over the ppaux they replaced), sorry if people had any doubts about that (tbh I should've opened an issue about what was learned during that refactor and how it can be cleaned up).

[Nilstrieb]

dropped the dependency on #116813 since that PR is proving difficult, this can be reviewed now.

[compiler-errors]

great work @Nilstrieb

@bors r+

[bors]

📌 Commit 6fc6a6d has been approved by compiler-errors

It is now in the queue for this repository.

[bors]

⌛ Testing commit 6fc6a6d with merge e8b8c78...

[bors]

☀️ Test successful - checks-actions
Approved by: compiler-errors
Pushing e8b8c78 to master...

[rust-timer]

Finished benchmarking commit (e8b8c78): comparison URL.

Overall result: ✅ improvements - no action needed

@rustbot label: -perf-regression

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.3%	[-0.3%, -0.3%]	1
All ❌✅ (primary)	-	-	0

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
Regressions ❌ (secondary)	3.8%	[3.8%, 3.8%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-1.4%	[-1.6%, -1.1%]	2
All ❌✅ (primary)	-	-	0

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)	-	-	0
Improvements ✅ (secondary)	-3.0%	[-4.5%, -2.3%]	11
All ❌✅ (primary)	-	-	0

Binary size

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

Bootstrap: 631.187s -> 631.223s (0.01%)
Artifact size: 304.03 MiB -> 304.04 MiB (0.00%)