Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fix(server): error in pre handler triggers after event #1500

Merged
merged 2 commits into from Sep 19, 2017

Conversation

retrohacker
Copy link
Member

Currently, aborting a middleware chain inside a pre handler will cause restify to drop the chain on the floor. This results in the after event not firing.

An effect of this is that after handlers do not get access to requests rejected by throttles.

This PR fixes that.

@@ -2957,3 +2957,22 @@ test('calling next twice should throw', function (t) {
t.ifError(err);
});
});

test('aborting pre should still call after', function (t) {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we put this at/around L2312 to group it with the other pre handler test that is testing next(false)? And maybe rename the test to next(err) should still trigger after event to be a little more precise since "abort" doesn't really mean much in context of existing restify docs or nomenclature.

lib/server.js Outdated
self._finishReqResCycle(req, res);
// Like with regular handlers, if we are provided an error, we
// should abort the middleware chain and fire after events.
if (err !== undefined) {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This behavior is slightly inconsistent compared to how its implemented in the use chain, where we are explicitly checking for false or err. The result is that calling return next({}) could trigger an early exit in pre where it wouldn't in use. I don't mind revisiting what nexting with a POJO means, but probably worth tackling in a separate PR?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yup, was on the fence about this. The explicitness bit us in the butt here 😄 so I erred on the side of assuming everything was a cause to abort the chain. Though I guess if someone were to do next(true) or next(null) this would drop the chain which may not be desirable.

@DonutEspresso
Copy link
Member

Awesome - great work tracking this down. 🎉

Copy link
Member

@yunong yunong left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good, I'm just confused about dropping a code fragment.

return;
}

if (!err) {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

How come this got dropped? I think if you remove this there is the edge case where the "err" object is not false or a function?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants