Migrate to windows-rs 0.37 #102
Conversation
windows-rs 0.36 introduced its own Event type, which ends up in the glob imports and would collide without this rename.
|
@str4d Please update the test harness as you did for the example. I'm aware that our CI doesn't actually run the Windows tests at the moment, but I tried to do so, and they didn't build. |
This is in preparation for windows-rs 0.32 which moves to a trait-based `implement` macro. Closes AccessKit#104.
|
Ahh, I missed this because |
|
Okay, this should now be working, and also resolves #104. |
|
@str4d Just wondering if you've had a chance to look at the two pending code review comments I left shortly after you opened the PR. |
|
I don't see any unresolved issues on this PR. Are they in a review that hasn't been published? |
|
The only action items I'm aware of were about the |
|
Hello, Version 0.38.0 was recently published. Is it worth directly upgrading to it? By reading the changelog, it seems to me that this would just be a matter of updating Thanks again @str4d for your contribution. |
| enum FragmentRootResult { | ||
| This, | ||
| This(PlatformNode), |
There was a problem hiding this comment.
The spirit of the old implementation can be maintained by having PlatformNode derive Clone, then using self.clone().into() instead of self.into(). Please either do this or remove the enum altogether; it's safe to do the latter, since in your current implementation, both match arms do the same thing.
| @@ -668,6 +683,8 @@ macro_rules! properties { | |||
| macro_rules! patterns { | |||
| ($(($base_pattern_id:ident, $is_supported:ident, ( | |||
| $(($base_property_id:ident, $getter:ident, $com_type:ident)),* | |||
| ), ( | |||
| $($extra_trait_method:item),* | |||
There was a problem hiding this comment.
Hmm. Putting whole function definitions inside a macro invocation seems likely to be unfriendly to IDEs and other tools.
Did you verify that Rust doesn't allow multiple impl blocks for a trait as it does for the struct itself?
|
@str4d I'm really sorry; I wrote my review comments on May 21, but since this is the first time I've used the GitHub code review feature as a reviewer, I didn't realize I had to submit the review after adding the comments. |
|
@str4d Just wanted to confirm that you received the review comments that I belatedly submitted on Sunday. |
|
I've decided that the two review comments I posted aren't blockers. I'll go ahead and merge this, then address those issues myself when I have a chance. |
This reverts commit a3eaec6.
No description provided.