1637922 - Consider disabling dom.netinfo.enabled on mobile

Status: RESOLVED FIXED

(GeckoView :: General, defect, P3)

Description

[Anne (:annevk)]

This is not a feature we ship on Firefox desktop and we don't plan to as we think it exposes too much user information. It would be good if we did the same on mobile for the same reason.

See bug 1057169.

A follow-up bug could then be used to remove the code once some time has passed.

Comment 1

[[:fabrice] Fabrice Desré]

Note that we need to keep it on b2g, so disabling is fine but not removing the code.

Comment 2

[Karl Dubost💡 :karlcow]

So this probably creates a webcompat issue on Firefox Android for CDN shopify theming.
https://webcompat.com/issues/67436

Comment 3

[Karl Dubost💡 :karlcow]

If deactivating it creates more webcompat issues than solving this one, the webcompat team could still create a site intervention targeting the CDN of shopify.

Comment 4

[Karl Dubost💡 :karlcow]

Attachment: Bug 1637922. Disables dom.netinfo.enabled everywhere. r=annevk

Comment 5

[Karl Dubost💡 :karlcow]

Intent to unship.
https://groups.google.com/a/mozilla.org/g/dev-platform/c/u1QiOGUIUfk

Comment 6

[Karl Dubost💡 :karlcow]

Attachment: Bug 1637922. Disables dom.netinfo.enabled everywhere. r=annevk

Comment 7

[Karl Dubost💡 :karlcow]

Attachment: Bug 1637922. Switches to disable everywhere for netinfo. r=annevk

Depends on D126431

Comment 8

[Karl Dubost💡 :karlcow]

Attachment: Bug 1637922. Disables netinfo in some tests. r=annevk

Depends on D126442

Comment 9

[BugBot [:suhaib / :marco/ :calixte]]

There are some r+ patches which didn't land and no activity in this bug for 2 weeks.
:karlcow, could you have a look please?
For more information, please visit auto_nag documentation.

Comment 10

[Karl Dubost💡 :karlcow]

Hmmm Thomas do you think this is ready to land?

Comment 11

[Thomas Wisniewski [:twisniewski]]

@karlcow, based on what I'm seeing here, I don't think there's anything left to do, so it seems like we could land it.

Comment 12

[Karl Dubost💡 :karlcow]

You have the landing power. :)

Comment 13

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/e9ffc2ed8f4d
Disables dom.netinfo.enabled everywhere. r=annevk
https://hg.mozilla.org/integration/autoland/rev/469353b7c340
Switches to disable everywhere for netinfo. r=annevk,webidl,peterv
https://hg.mozilla.org/integration/autoland/rev/6676088e5fdf
Disables netinfo in some tests. r=annevk,webidl,peterv

Comment 14

[Cristina Cozmuta (:CrissCozmuta)]

Backed out for causing failures on test_worker_interfaces.html

Backout link: https://hg.mozilla.org/integration/autoland/rev/407cf6c951c48b80f7851134504b6e6ede964ed3

Push with failures : https://treeherder.mozilla.org/jobs?repo=autoland&duplicate_jobs=visible&resultStatus=testfailed%2Cbusted%2Cexception%2Crunnable&revision=6676088e5fdf7b2714e7b218b3f6a8078c0b06a2&selectedTaskRun=JxI2F6ztSq6mXxASL8gDwg.0

Link to failure log : https://treeherder.mozilla.org/logviewer?job_id=367078055&repo=autoland&lineNumber=2967

Comment 15

[Karl Dubost💡 :karlcow]

I will be working on it tomorrow. Thanks james for trying to land it.

Comment 16

[Thomas Wisniewski [:twisniewski]]

The failures are:

[task 2022-02-08T15:47:53.447Z] 15:47:53  WARNING -  TEST-UNEXPECTED-FAIL | dom/workers/test/test_worker_interfaces.html | false: NetworkInformation should  be defined on self
[task 2022-02-08T15:47:53.447Z] 15:47:53     INFO -      SimpleTest.ok@SimpleTest/SimpleTest.js:417:16
[task 2022-02-08T15:47:53.447Z] 15:47:53     INFO -      workerTestExec/worker.onmessage@dom/workers/test/worker_driver.js:35:9


TEST-UNEXPECTED-FAIL | dom/workers/test/test_worker_interfaces.html | 1 === 0: The following interface(s) are not enumerated: NetworkInformation
[task 2022-02-08T15:47:53.448Z] 15:47:53     INFO -      SimpleTest.ok@SimpleTest/SimpleTest.js:417:16
[task 2022-02-08T15:47:53.448Z] 15:47:53     INFO -      workerTestExec/worker.onmessage@dom/workers/test/worker_driver.js:35:9

So hopefully we should just need to change the test expectation here to "disabled": https://searchfox.org/mozilla-central/source/dom/workers/test/test_worker_interfaces.js#228

Comment 17

[Karl Dubost💡 :karlcow]

Attachment: Bug 1637922. Disables for serviceworker tests. r=twisniewski

Depends on D126447

Comment 18

[Karl Dubost💡 :karlcow]

Thomas, could you review this and try to land again?
https://phabricator.services.mozilla.com/D138398

Thanks.

Comment 19

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/85a5290f71e4
Disables dom.netinfo.enabled everywhere. r=annevk
https://hg.mozilla.org/integration/autoland/rev/cda9709db315
Switches to disable everywhere for netinfo. r=annevk,webidl,peterv
https://hg.mozilla.org/integration/autoland/rev/38737ed32dc1
Disables netinfo in some tests. r=annevk,webidl,peterv
https://hg.mozilla.org/integration/autoland/rev/4a3025a77f16
Disables for serviceworker tests. r=twisniewski

Comment 20

[Iulian Moraru]

https://hg.mozilla.org/mozilla-central/rev/85a5290f71e4
https://hg.mozilla.org/mozilla-central/rev/cda9709db315
https://hg.mozilla.org/mozilla-central/rev/38737ed32dc1
https://hg.mozilla.org/mozilla-central/rev/4a3025a77f16

Comment 21

[Hamish Willee]

FF99 docs work for this can be tracked in https://github.com/mdn/content/issues/13840#issuecomment-1066255533

For my information, has the Network Information API been formally deprecated/removed from the standards track? It looks to me like this is not the case - though the API is being revisited in specification to address issues like the ones that required the feature be disabled.

Comment 22

[Anne (:annevk)]

It's not standards track at the moment, is it? It's just a WICG document, not a deliverable of a W3C WG, WHATWG, or IETF WG.

Comment 23

[Hamish Willee]

Thanks Anne - you're right - the compat data says it is on "standards track" but the spec is indeed WICG. I'll query with the compat people as to why.

Comment 24

[Hamish Willee]

Hi @Anne,
FYI only, the comment from the browser compat team was that:

WICG is a gray area in terms of whether it's considered standard track or not, depending on who you ask. In BCD, we have treated WICG as if it is a standard track, and WICG spec URLs are included in w3c/browser-specs

That is all I personally needed.

Comment 25

[Anne (:annevk)]

To be clear, Mozilla does not consider WICG standards track.

Comment 26

[Hamish Willee]

Yes - for anyone else coming to this discussion the short version is that the standard_track key is in the data is a misleading name. All it means is that there is a specification associated with the feature.
We're hoping to get rid of that key altogether: https://github.com/mdn/browser-compat-data/issues/12685

Comment 27

[Karl Dubost💡 :karlcow]

For https://github.com/webcompat/web-bugs/issues/95799
deactivating it solved an issue and made the website working.