Issue metadata
Sign in to add a comment
|
chained promise requesting a resource not resolving as in Chrome51
Reported by
alberto....@gmail.com,
Jul 4 2016
|
||||||||||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/52.0.2743.60 Safari/537.36 Steps to reproduce the problem: 1. Run plunker that uses AngularJS and chained 2 promises. http://plnkr.co/edit/DF7AcC We can see that 2 HTTP request and responses to get sample.json file are resolved correctly. We can see that the problem is not-reproducible in HTTPS or with .html files. What is the expected behavior? In Chrome 51, it executes the 2nd nested promise, printing "Done!" in the console log. What went wrong? In Chrome 52, it does not resolve the promise correctly on a first execution, but it will work correctly on future executions if something is modified on the plunkR. Did this work before? Yes Chrome 51 and before. Chrome version: 52.0.2743.60 Channel: beta OS Version: 6.1 (Windows 7, Windows Server 2008 R2) Flash Version:
,
Jul 7 2016
,
Jul 7 2016
,
Jul 11 2016
I can't reproduce it using plain XHR, so I have reported it in Angular.js codebase. plain XHR example: http://plnkr.co/edit/RJlyNW?p=preview Angular.js related issue: https://github.com/angular/angular.js/issues/14865 Anyway, I feel there's something that have changed on Chrome52 that is reproducing this regression issue, so if I were you I won't discard it immediatelly.
,
Aug 2 2016
I'm having a lot of trouble reproducing this bug, even though I was bitten very badly by it in a Backbone app. I'm sure that, prior to the weekend, I was able to reproduce using Alberto's Angular plunker, but I can't now. My version, using only jQuery: http://plnkr.co/edit/IECSdYMoKEv1sjtFdBVv?p=preview
,
Aug 3 2016
Not reproducible with Chrome 52. alberto, can you reproduce the issue with recording the network activities via the devtools? If so, providing the screenshot of the devtools network tab would be helpful. Thanks,
,
Aug 3 2016
Hi, As you can see, on a first run devTools_nok shows a 200 response and a 304 response, and the second one is not calling the handler. In the second image, after running the plunkr again, we can see 2 304 responses, both of them returning ok. So I feel that the combination of calling XHR returning 304 from inside the function where it has already returned 200 is the problem. I have used Version 53.0.2785.34 beta-m (64-bit)
,
Aug 3 2016
Thank you for reporting! Probably this has the same root cause as Issue 633696 , because both issues are related to initiating in an XHR's onload another XHR to the same URL and 304 responses.
,
Aug 12 2016
|
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 Deleted