Reference count TokenStream in fallback mode #335
Merged
+22
−14
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The data structure for TokenStream currently in proc-macro2 is a bad fit for the API we inherit from libproc_macro. Specifically this problematic part:
This function, along with the lack of by-reference iterator on TokenStream, means that the only possible way to parse tokens ends up performing quadratically many token clones. For example if you start with a TokenStream representing
( ( ( ( a ) ) ) )
, parsing it involves:( ( ( a ) ) )
( ( a ) )
( a )
a
.Changing to
Rc<Vec<TokenTree>>
makes a large difference in syn's benchmarks.Before:
After:
Rustc also uses a similar
Rc<Vec>
data structure in its own TokenStream implementation:https://github.com/rust-lang/rust/blob/1.62.1/compiler/rustc_ast/src/tokenstream.rs#L305-L306