rustc_privacy: Fix bugs in SanePrivacyVisitor #29726
Conversation
|
(rust_highfive has picked a reviewer for you, use r? to override) |
|
I'm not actually sure if the part of SanePrivacyVisitor ensuring that there's no |
| hir::ItemTrait(..) | hir::ItemDefaultImpl(..) | | ||
| hir::ItemConst(..) | hir::ItemStatic(..) | hir::ItemStruct(..) | | ||
| hir::ItemFn(..) | hir::ItemMod(..) | hir::ItemTy(..) | | ||
| hir::ItemExternCrate(_) | hir::ItemUse(_) => {} |
There was a problem hiding this comment.
I personally like to always have exhaustive matches wherever possible, e.g. it's easy to skim over this list and say "ah yeah pub is allowed on each of these". Perhaps this part could stay?
|
Looks good to me! (just a minor nit) Scheduling a crater run just to double check |
alexcrichton
added
the
S-waiting-on-crater
|
Crater reports one regression, but I believe it's spurious, so otherwise this should be good to go. |
alexcrichton
removed
the
S-waiting-on-crater
|
Updated with exhaustive matching. |
alexcrichton
added
the
relnotes
- Check privacy sanity in all blocks, not only function bodies - Check all fields, not only named - Check all impl items, not only methods - Check default impls - Move the sanity check in the beginning of privacy checking, so others could rely on it Technically it's a [breaking-change], but I expect no breakage because, well, it's *sane* privacy visitor, if code is broken it must be insane by definition!
Technically it's a [breaking-change], but I expect no breakage because, well, it's sane privacy visitor, if code is broken it must be insane by definition!