Rustdoc: ignore deref-inherited static methods

[aochagavia]

Fixes#24575
Fixes#25673

r? @alexcrichton

[aochagavia]

I am not sure on how to test this. At this moment I only know that it compiles...

[alexcrichton]

Could you modify the existing tests for #19190 in conjunction with the @count directive to ensure that none of these show up in the various cases tested by #19190? You should be able to just modify the contents of the src/test/rustdoc folder.

Also, could you upload a copy of the docs so they could be manually verified as well?

[aochagavia]

It seems that my changes aren't entirely correct. Hiding static methods works if the target type is a primitive (in this case i32), but doesn't if it is user-defined. I have checked the logic of the code (mainly the following functions: build_deref_target_impls, build_deref_impls, build_impls, build_impls_helper, build_impl) and I am unable to find the problem. Are there any other functions which could interact with this feature?

btw, I used @!has instead of @count to check that the static methods aren't shown. Is that ok?

[alexcrichton]

I think you may want to modify the rendering function instead of the inlining functions as local Deref target impl blocks are not inlined (they're already in the AST) so they're not pruned already.

Also yeah using @!has is fine, I didn't even know it existed!

[aochagavia]

Thanks for your feedback! I have updated the PR and uploaded the documentation to http://aochagavia.github.io/rust/index.html

[alexcrichton]

@bors: r+ 2c79fb5

Thanks!

[alexcrichton]

@bors: r-

er actually, could you squash the commits and a more descriptive commit message about this fix? Unfortunately Fix #XXXXX isn't too descriptive :(

[bors]

☔ The latest upstream changes (presumably #23606) made this pull request unmergeable. Please resolve the merge conflicts.

[sfackler]

Ping?

[aochagavia]

Sorry, I won't have time to resolve the merge conflicts and test the patch again in the following weeks. Feel free to pick it up, though.

[bluss]

There's another tricky case, all types that deref to slice also show the method fn into_vec(self: Box<[T]>) -> Vec<T>. Is it possible to filter out that as well?

[aochagavia]

Finally I found time to rebase this. I think it would be better to create an issue for the case @bluss pointed out, since I will not be able to tackle it anytime soon.

@alexcrichton r+?

[bluss]

that's fine, I'm happy with every improvement to rustdoc, so thanks!

[bluss]

Also fixes dup #25673

[alexcrichton]

@bors: r+ 8703883

Thanks!

[bors]

⌛ Testing commit 8703883 with merge b7408c1...

[bors]

💔 Test failed - auto-linux-64-opt

[alexcrichton]

@bors: retry

On Tue, May 26, 2015 at 6:58 AM, bors notifications@github.com wrote:

[image: 💔] Test failed - auto-linux-64-opt
http://buildbot.rust-lang.org/builders/auto-linux-64-opt/builds/5088

—
Reply to this email directly or view it on GitHub
#24657 (comment).

[bors]

⌛ Testing commit 8703883 with merge 1742a01...

[bors]

☀️ Test successful - auto-linux-32-nopt-t, auto-linux-32-opt, auto-linux-64-nopt-t, auto-linux-64-opt, auto-linux-64-x-android-t, auto-mac-32-opt, auto-mac-64-nopt-t, auto-mac-64-opt, auto-win-gnu-32-nopt-t, auto-win-gnu-32-opt, auto-win-gnu-64-nopt-t, auto-win-gnu-64-opt