Closed
Bug 1394841
Opened 7 years ago
Closed 7 years ago
WebExtensions - Messaging is not working in responsive mode
Categories
(DevTools :: Responsive Design Mode, defect, P2)
Tracking
(firefox57 fix-optional, firefox58 fixed)
RESOLVED
FIXED
Firefox 58
Tracking | Status | |
---|---|---|
firefox57 | --- | fix-optional |
firefox58 | --- | fixed |
People
(Reporter: ido.y, Assigned: jryans)
References
Details
(Whiteboard: [designer-tools])
Attachments
(4 files)
User Agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3112.101 Safari/537.36
Steps to reproduce:
In responsive mode,
Listen to message in content script
And send a message from background
Happens also in nightly version
Actual results:
Nothing
Expected results:
Should handle message
Updated•7 years ago
|
Component: Untriaged → WebExtensions: General
Product: Firefox → Toolkit
Comment 1•7 years ago
|
||
This is because the page loaded into a tab in responsive design mode is a developer tools page (which the extension is not allowed to access), with the actual web content loaded into a remote subframe.
I'm not really sure whether we should try to handle this corner case. It affects a lot of things other than messaging.
Component: WebExtensions: General → WebExtensions: Developer Tools
Hi Kris,
We rely on this functionality, and we have cross-browser solution which is working on Chrome
Is there a simple way to solve this issue?
Assignee | ||
Comment 3•7 years ago
|
||
Similar bugs have also been filed for legacy-style add-ons back in the day, such as bug 1320159.
I think it would be nice to have it working for a consistent experience, but not a major issue in the scheme of things.
I think I'll move this to RDM component, because RDM is really the cause and potentially where a fix might be made.
Assignee | ||
Updated•7 years ago
|
Component: WebExtensions: Developer Tools → Developer Tools: Responsive Design Mode
Product: Toolkit → Firefox
Assignee | ||
Updated•7 years ago
|
Assignee | ||
Updated•7 years ago
|
Assignee | ||
Comment 4•7 years ago
|
||
Seems to also be an issue for DevTools extensions like React:
https://github.com/facebook/react-devtools/issues/904
I think I'll investigate a bit to see how much work would be needed here.
Flags: needinfo?(jryans)
Assignee | ||
Comment 7•7 years ago
|
||
It looks like only a small number of changes are needed to support this, so I'll give it a try.
Assignee: nobody → jryans
Status: NEW → ASSIGNED
Flags: needinfo?(jryans)
Priority: P3 → P2
Whiteboard: [designer-tools]
Comment hidden (mozreview-request) |
Comment hidden (mozreview-request) |
Comment hidden (mozreview-request) |
Assignee | ||
Comment 11•7 years ago
|
||
Comment hidden (mozreview-request) |
Assignee | ||
Comment 13•7 years ago
|
||
Comment 14•7 years ago
|
||
mozreview-review |
Comment on attachment 8924341 [details]
Bug 1394841 - Add WebExt messaging to RDM tunnel.
https://reviewboard.mozilla.org/r/195584/#review201798
Thanks for looking into that!
Attachment #8924341 -
Flags: review?(poirot.alex) → review+
Comment 15•7 years ago
|
||
mozreview-review |
Comment on attachment 8924342 [details]
Bug 1394841 - Test WebExt messaging with RDM.
https://reviewboard.mozilla.org/r/195586/#review201796
::: devtools/client/responsive.html/test/browser/browser_ext_messaging.js:100
(Diff revision 2)
> + let { id } = sender;
> + respond({ id });
> + });
> +
> + await new Promise(resolve => {
> + browser.test.onMessage.addListener(resolve);
Are you waiting for "resume" message?
It may be helpful to comment about that here.
::: devtools/client/responsive.html/test/browser/browser_ext_messaging.js:120
(Diff revision 2)
> + extension.awaitMessage("content-script-ready"),
> + extension.awaitMessage("background-script-ready"),
> + ]);
> + extension.sendMessage("resume");
> +
> + await extension.awaitFinish("rdm-messaging");
nit: for both extension.startup() and extension.sendMessage(resume) we do the action first and listen right after. This is the typical pattern to get intermittent/races.
I imagine this is unlikely here.
But for example it would make the wait for content-script-ready and background-script-ready easier.
If you wait for them before startup you can then do:
await extension.startup()
await onContentScriptReady;
await onBackgroundScriptReady;
Attachment #8924342 -
Flags: review?(poirot.alex) → review+
Comment 16•7 years ago
|
||
mozreview-review |
Comment on attachment 8924342 [details]
Bug 1394841 - Test WebExt messaging with RDM.
https://reviewboard.mozilla.org/r/195586/#review202584
Attachment #8924342 -
Flags: review?(kmaglione+bmo) → review+
Comment 17•7 years ago
|
||
mozreview-review |
Comment on attachment 8924340 [details]
Bug 1394841 - Check iframes for related tabs in WebExt.
https://reviewboard.mozilla.org/r/195582/#review202588
::: browser/components/extensions/ext-browser.js:38
(Diff revision 1)
> - } else if (target instanceof Ci.nsIDOMXULElement) {
> + } else if (target instanceof Ci.nsIDOMXULElement ||
> + ExtensionUtils.instanceOf(target, "HTMLIFrameElement")) {
How does this work? getBrowserData only knows how to handle <browser>s that are a child of <tabbrowser> in a browser window, or belong to about:addons.
Does the RDM UI have a stub gBrowser global, or something like that?
Assignee | ||
Comment 18•7 years ago
|
||
mozreview-review-reply |
Comment on attachment 8924340 [details]
Bug 1394841 - Check iframes for related tabs in WebExt.
https://reviewboard.mozilla.org/r/195582/#review202588
> How does this work? getBrowserData only knows how to handle <browser>s that are a child of <tabbrowser> in a browser window, or belong to about:addons.
>
> Does the RDM UI have a stub gBrowser global, or something like that?
Yes, the RDM UI defines a usable `gBrowser` for call sites like this that want to look up the tab for the browser frame, so that's what allows this to work here as well.
Comment 19•7 years ago
|
||
mozreview-review |
Comment on attachment 8924340 [details]
Bug 1394841 - Check iframes for related tabs in WebExt.
https://reviewboard.mozilla.org/r/195582/#review203018
Attachment #8924340 -
Flags: review?(kmaglione+bmo) → review+
Assignee | ||
Comment 20•7 years ago
|
||
mozreview-review-reply |
Comment on attachment 8924342 [details]
Bug 1394841 - Test WebExt messaging with RDM.
https://reviewboard.mozilla.org/r/195586/#review201796
> Are you waiting for "resume" message?
> It may be helpful to comment about that here.
Ah right, I'll add a comment, thanks!
> nit: for both extension.startup() and extension.sendMessage(resume) we do the action first and listen right after. This is the typical pattern to get intermittent/races.
> I imagine this is unlikely here.
> But for example it would make the wait for content-script-ready and background-script-ready easier.
> If you wait for them before startup you can then do:
> await extension.startup()
> await onContentScriptReady;
> await onBackgroundScriptReady;
The extension isn't installed until `startup`, so that one should be safe. Anyway, I agree the other listeners should be moved earlier. (I am usually watching out for this anyway, but I guess I missed it here.) Thanks for catching this.
Comment hidden (mozreview-request) |
Comment hidden (mozreview-request) |
Comment hidden (mozreview-request) |
Assignee | ||
Comment 24•7 years ago
|
||
Comment 25•7 years ago
|
||
Pushed by jryans@gmail.com:
https://hg.mozilla.org/integration/autoland/rev/4aa421a32de9
Check iframes for related tabs in WebExt. r=kmag
https://hg.mozilla.org/integration/autoland/rev/134b46ce0aa5
Add WebExt messaging to RDM tunnel. r=ochameau
https://hg.mozilla.org/integration/autoland/rev/bbf3fb8a107c
Test WebExt messaging with RDM. r=kmag,ochameau
Comment 26•7 years ago
|
||
bugherder |
https://hg.mozilla.org/mozilla-central/rev/4aa421a32de9
https://hg.mozilla.org/mozilla-central/rev/134b46ce0aa5
https://hg.mozilla.org/mozilla-central/rev/bbf3fb8a107c
Status: ASSIGNED → RESOLVED
Closed: 7 years ago
status-firefox58:
--- → fixed
Resolution: --- → FIXED
Target Milestone: --- → Firefox 58
Updated•6 years ago
|
Product: Firefox → DevTools
You need to log in
before you can comment on or make changes to this bug.
Description
•