Rework missing_docs_in_private_items
#14741
Conversation
|
r? @Alexendoo rustbot has assigned @Alexendoo. Use |
rustbot
added
the
S-waiting-on-review
This comment has been minimized.
This comment has been minimized.
|
Looks like vscode did something weird and disappeared the new test file. |
|
r? clippy |
|
Lintcheck changes for 41ee3b4
This comment will be updated if you push new changes |
* Don't search spans between items for comments. * Don't lint items inside bodies, automatically derived impls. * Don't lint items inside macro generated modules unless they are namable outside. * Delay `is_from_proc_macro` checks * Handle reexports when checking if an item is reachable from the crate root. * Don't lint the crate root.
There was a problem hiding this comment.
From the lintcheck result, it looks like the quality of the lint messages have regressed. For example, talking about "impl item" instead of "associated function" or "method" might make it less clear to the user.
For example, the compiler makes this distinction.
|
Is the distinction really needed here? The only time it might even be helpful is when you view the message without the span which isn't the how you would normally view warnings. Calling all items / trait items / impl items just "item" might be better than even making the current distinction. |
|
☔ The latest upstream changes (possibly 20ce69b) made this pull request unmergeable. Please resolve the merge conflicts. |
This is blocking #14724 as it triggers the debug assertions it adds.
The inter item span parsing was introduced to solve #12197. This is better handled by just not linting anything within bodies.
changelog: [
missing_docs_in_private_items]: Don't lint items in bodies and automatically derived implschangelog: [
missing_docs_in_private_items]: Better detect when things are accessible from the crate rootchangelog: [
missing_docs_in_private_items]: Lint unnameable items which are accessible outside the crate