Issue with protocol less resources in nested iframes
Reported by
iceke...@gmail.com,
Aug 19
|
|||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/68.0.3440.106 Safari/537.36 Steps to reproduce the problem: 1. create an iframe inside an iframe through srcdoc or with src about:blank 2. insert a protocol less call inside innest iframe What is the expected behavior? I expect resources to be loaded with parent iframe or base href url protocol What went wrong? Here's an example with srcdoc: https://codepen.io/anon/pen/oPvwBz I've encountered this issue developing VPAID wrapper to be working with Google IMA which is unluckily requiring JS file without protocol. It's returning error since resource URL is prepended with about: protocol. It looks like issue was present in the past and it has been marked as fixed, but now it is still present. See: https://bugs.chromium.org/p/chromium/issues/detail?id=484930 https://bugs.chromium.org/p/chromium/issues/detail?id=496814 No issue with other browsers, since protocol is correctly assigned even when no base href is set. Did this work before? Yes 48 Chrome version: 68.0.3440.106 Channel: stable OS Version: 10.0 Flash Version:
,
Aug 20
icekemia@ Thanks for the issue. Tested this issue on Windows 10 on the reported version 68.0.3440.106 and the latest Canary 70.0.3527.0 by following the below steps. 1. Launched Chrome and navigated to the give codepen link: https://codepen.io/anon/pen/oPvwBz 2. Couldn't observe any resources loaded in the iframe and the same behavior is observed on Firefox. Attached is the screen shot for reference. Request you to check and confirm if anything is missed from our end in triaging the issue. Also request you to provide a screen cast of the steps followed to reproduce the issue which will help in better understanding. Thanks..
,
Aug 20
Hi, i also found an error with fullscreen video, this error occurs when the video is a child of an iframe inside an iframe. Example: https://jsfiddle.net/EMLilianaIturribarria/aq5dx8co/ Js Fiddle already wraps my iframe with an iframe. if you check the iframe src in isolation on clicking watch this will toggle fullscreen. however that doesnt work if it's inside an iframe.
,
Aug 21
@liliana, try adding allow="fullscreen" attribute to your iframe
,
Aug 21
Thank you for providing more feedback. Adding the requester to the cc list. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Aug 21
@icekemia, note we are still asking for confirmation of our test method in comment #2.
,
Aug 24
Reporter, I think https://codepen.io/anon/pen/oPvwBz doesn't represent the issue because it has <base> and window.location.protocol is unrelated to base URL computation. Would you provide another reproduction case please? http://jsfiddle.net/n0gdh5y4/5/ linked in Issue 484930 still works correctly.
,
Aug 27
,
Aug 29
Sorry if it tooks so long, our test tag had expired and we couldn't add real test. You can check the following testpage https://goo.gl/yoEBAP It displays fine on Firefox, while Chrome fails loading ima3.js If we edit vpaid_adapter.js to load ima3.js with protocol then everything works as it's supposed to.
,
Aug 29
Thank you for providing more feedback. Adding the requester to the cc list. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Aug 30
> You can check the following testpage https://goo.gl/yoEBAP I saw no problem with this page though it had four cross-site-script-in-document.write warnings about adfscript and bootstrap.js. I saw no console messages about ima3.js, and I couldn't find strings "vpaid_adapter.js" and "ima3.js" in all source files of the page.
,
Aug 30
There's no console message about ima3.js. If you check Network tab you might be able to find vpaid_adapter.js (only after pressing on play button) This file is calling ima3.js, you can use fiddler to respond with an edited version of the file. Il giorno gio 30 ago 2018 alle ore 03:29 tk… via monorail < monorail+v2.567034358@chromium.org> ha scritto:
,
Aug 30
Thank you for providing more feedback. Adding the requester to the cc list. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Aug 30
> might be able to find vpaid_adapter.js (only after pressing on play button) Oh, I see. I confirmed vpaid_adapter.js was loaded in the network tab. Is there a user-visible behavior difference except for the network tab? Firefox shows an ad video before "THE RICH MEDIA RIDE" video Chrome shows "THE RICH MEDIA RIDE" video without an ad Is this?
,
Aug 30
Exactly, since ima3.js doesn't load, adv video is not loading as well (and you can see timeout while waiting for event 'AdLoaded' error in console) Il giorno gio 30 ago 2018 alle ore 10:36 tk… via monorail < monorail+v2.567034358@chromium.org> ha scritto:
,
Aug 30
Thank you for providing more feedback. Adding the requester to the cc list. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Aug 30
Ok, thanks. So, the reproducing steps: 1. Open https://goo.gl/yoEBAP 2. Click the play button in the black rectangle 3. Observe Expected: Playing an advertisement video before "THE RICH MEDIA RIDE" video Actual: Playing "THE RICH MEDIA RIDE" video without any advertisement video TE, please do bisect since Chrome 48.
,
Aug 30
,
Aug 31
Able to reproduce the issue on Windows 10, Mac OS 10.13.3 and Ubuntu 17.10 on the latest Stable 68.0.3440.106 and the latest Canary 70.0.353.0 as per comment #17. This issue is observed from M-45(45.0.2410.0) build. Attached is the screen cast for reference. Hence this is a Non-Regression issue and marking this as Untriaged for further updates from Dev. Removing 'Needs-Bisect' label as this issue is observed from M-45 builds. Please feel free to add if needed. Thanks..
,
Sep 3
,
Sep 4
Here is a minimized reproduction: http://jsfiddle.net/int32_t/oztck57e/5/ The key point is doc2.open() in a srcdoc document context. The current Chrome behavior conforms to the HTML specification, and other browsers don't. However IMO we should change the specification. I'll file a specification issue.
,
Sep 4
|
|||||||||||||||
►
Sign in to add a comment |
|||||||||||||||
Comment 1 by susan.boorgula@chromium.org
, Aug 19