Skip to content
Snippets Groups Projects

impl_standard_builder: Test the Deserialize impl and have it generate ::builder

Merged impl_standard_builder: Test the Deserialize impl and have it generate ::builder
Diziet/arti:builder-default-3 into main
1 unresolved thread
Merged Ian Jackson requested to merge Diziet/arti:builder-default-3 into main
1 unresolved thread

This detects bugs like fixed in !502 (merged), and replaces quite a lot of handwritten code.

Merge request reports

Loading
Loading

Activity

Filter activity
  • Approvals
  • Assignees & reviewers
  • Comments (from bots)
  • Comments (from users)
  • Commits & branches
  • Edits
  • Labels
  • Lock status
  • Mentions
  • Merge request status
  • Tracking
crates/tor-config/src/lib.rs
140 //
141 // The implementation munches fake "trait bounds" off the RHS.
142 // We're going to add at least one more option.
108 143 #[macro_export]
109 macro_rules! impl_standard_builder { {
110 $Config:ty
111 } => {
112 $crate::paste!{
144 macro_rules! impl_standard_builder {
145 {
146 $Config:ty $(: $($options:tt)* )?
147 } => { $crate::impl_standard_builder!{
148 @ ( try_deserialize ) $Config : $( $( $options )* )?
149 } };
150 {
151 @ ( $($try_deserialize:ident)? ) $Config:ty : $(+)? !Deserialize $( $options:tt )*
  • Nick Mathewson
    Nick Mathewson @nickm ·
    Owner

    I don't understand the purpose of $(+)? here.

    Also, will this work if !Deserialize is not the first of $(options)?

  • Ian Jackson
    Ian Jackson @Diziet ·
    Author Maintainer

    The idea is to eat a syntax like FooConfig: !Deserialize + !OtherThing. So we have to match and discard a +. And, yes. Well, when there are other options they will be munched in any order.

    I have added some comments to explain the macro code. I hope this is helpful

  • Please register or sign in to reply
  • Nick Mathewson
    Nick Mathewson @nickm ·
    Owner

    I had a question about some of the macro-ness. Otherwise, it looks fine.

  • Ian Jackson added 1 commit

    added 1 commit

    • 8f724ad7 - impl_standard_builder: Better comments explaining the parser

    Compare with previous version

  • Ian Jackson
    Ian Jackson @Diziet ·
    Author Maintainer

    Pipeline failure looks spurious, especially given that all that I changed is comments. Retrying.

  • Nick Mathewson merged

    merged

  • Nick Mathewson mentioned in commit f482a5bd

    mentioned in commit f482a5bd

    • Please register or sign in to reply