Emulation: Support geolocation override

[jimevans]

One request often heard from users is that they want to simulate a browser at a specific geolocation for their sites. WebDriver should provide a mechanism to facilitate this testing.

[whimboo]

Thank you @jimevans for filing!

I want to add that clients should be able to set this location override per tab.

[css-meeting-bot]

The Browser Testing and Tools Working Group just discussed Geolocation emulation.

The full IRC log of that discussion

<jgraham> Topic: Geolocation emulation
<jgraham> github: https://github.com//issues/343
<orkon> q+
<jgraham> jgraham: We know there are requests for this and Puppeteer has it, so this should be the next priority for emulation and overrides
<jgraham> ack next
<jgraham> orkon: We support this in puppeteer. The blocker for using this in BiDi is that we need to be able to provide permissions. It might be possible to use a preload script to override things, but I'm not sure.
<jgraham> q?
<jgraham> jgraham: I assumed we'd just provide mock data, does that require implementing permissions?
<jgraham> orkon: To request geolocation data you first have to give permissions. We could say that providing the data atuo-grants the permissions, but it might make sense to do permissions first, if that's needed cross-browser.
<orkon> q+
<JimEvans> q+
<jgraham> jgraham: Makes sense. We can do the work in parallel though; browsers might have some out of band way to disable the permissions prompt so you could still test that the right geolocation data is returned without supporting the permissions API.
<jgraham> ack next
<jgraham> orkon: Perhaps there is also a relation to the user prompt handing changes; permissions are like user prompts and we could have similar auto-accept functionality.
<jgraham> ack next
<jgraham> JimEvans: Does the permissions spec have a BiDi implementation?
<jgraham> orkon: We have it in Chrome, but I don't know about other implementations. It only covers overrides.
<jgraham> jgraham: We have the permissions spec implemented in classic, but not yet for BiDi (need to check which parts are actually shipping already)
<jgraham> RRSAgent: make minutes
<RRSAgent> I have made the request to generate https://www.w3.org/2024/04/10-webdriver-minutes.html jgraham
<jgraham> zakim, bye
<Zakim> leaving. As of this point the attendees have been AutomatedTester_, orkon, jdescottes, whimboo, jgraham, sasha, MaksimSadym, JimEvans, lightning00blade
<sasha> orkon: `browsingContext.setViewport` command already has a step to check if a specified context is a top-level context, and throw an error if it's not, so we don't need to do anything for iframes
<jgraham> RRSAgent: bye
<RRSAgent> I see no action items

[OrKoN]

Should we propose it as part of the https://w3c.github.io/geolocation-api/ spec?

[OrKoN]

Chrome currently supports latitude, longitude, accuracy overrides which works for the top-level context and iframes. It does not affect the permission and it does not allow to use geolocation on insecure contexts.

[christian-bromann]

WebdriverIO currently already supports this by overwriting the navigator.geolocation interface in the browser via the preload script:

const patchedFn = options instanceof Error
    ? `cbError(new Error(${JSON.stringify(options.message)}))`
    : `cbSuccess({
        coords: ${JSON.stringify(options)},
        timestamp: Date.now()
    })`
await this.scriptAddPreloadScript({
    functionDeclaration: /*js*/`() => {
        Object.defineProperty(navigator.geolocation, 'getCurrentPosition', {
            value: (cbSuccess, cbError) => ${patchedFn}
        })
    }`
})

[OrKoN]

@christian-bromann Thanks for sharing! Do you support changing it after the page is loaded/during the test? Also, what about Geolocation.watchPosition()?

I think the override on the browser side would be more robust but preload scripts offer the ability to implement something already.

[christian-bromann]

I agree. The current emulation features we have can't be changed after the page has loaded. It would be helpful, especially for component tests to modify these capabilities without having to do a page load.

[whimboo]

Note that there is an open PR for the geolocation spec for WebDriver integration: w3c/geolocation#170

[whimboo]

And here are the minutes from yesterday's joined meeting with the WebApplication group.

[css-meeting-bot]

The Browser Testing and Tools Working Group just discussed emulation - geolocation.

The full IRC log of that discussion

<AutomatedTester> topic: emulation - geolocation
<jgraham> q+
<jgraham> https://pr-preview.s3.amazonaws.com/w3c/geolocation/pull/170.html#automation
<AutomatedTester> github https://github.com//issues/343
<AutomatedTester> github: https://github.com//issues/343
<AutomatedTester> ack jgraham
<AutomatedTester> jgraham: geolocation is a special case. THe webapps group have created an extension for this command. THere are a few issues but we can take their PR and put it directly into our spec
<sadym> q+
<AutomatedTester> ... and since we're putting emulaton directly into our spec that we take theirs and merge it and have that as a starting point for our discussions
<AutomatedTester> ack sadym
<jgraham> s/and merge it/and create a PR/
<AutomatedTester> sadym: when you say putting these into our spec are we going talking about the earlier topic or something else
<orkon> q+
<sadym> q+
<AutomatedTester> jgraham: like viewport is an emulation but I see what you mean
<AutomatedTester> ack orkon
<simonstewart> q+
<jgraham> q+
<AutomatedTester> orkon: I wonder if this would be better in the geolocation spec so that it maintained as a whole rather than in this spec?
<AutomatedTester> ack sadym
<AutomatedTester> sadym: if we keep the geolocation in the geolocation spec where should the other emulatons live?
<AutomatedTester> gsnedders: did the old JSON wire protocol handle geolocations?
<AutomatedTester> simonstewart:
<AutomatedTester> ... no it didn't... yay past ys
<AutomatedTester> ack simonstewart
<AutomatedTester> simonstewart: if geolocation are happy of this being an extension we should encourage to them to keep it since they have a better understnading
<gsnedders> s/what people want and the implementations might not work at user context/how people are actually using this; developers might not actually run into the differences here because most testing situations only have a single test running at a time, so maybe it doesn't super matter how we define this?/
<AutomatedTester> ... and to sadym question we should keep things where possible as close to the owners of that area
<gsnedders> https://www.selenium.dev/documentation/legacy/json_wire_protocol/#sessionsessionidlocation sure makes it seem like there was some command in the legacy protocol to change geolocation, though idk if there were implementations; it seems like safaridriver never did?
<whimboo> q+
<AutomatedTester> ... if we could keep webdriver super simple that would be awesome
<AutomatedTester> ack jgraham
<AutomatedTester> jgraham: it is essentially true that we could have this across all different specs
<AutomatedTester> ... but other specs have issues around our "stability"
<AutomatedTester> ... I just feel like it being a core feature for us it should be in our spec
<AutomatedTester> ... if we know about automation frameworks want this then it should be our spec
<AutomatedTester> simonstewart: how do you figure out where this would work in ours or there?
<jimevans> q+
<AutomatedTester> jgraham: I don't think it will change how things are implemented
<AutomatedTester> ack whimboo
<orkon> https://w3c.github.io/webdriver-bidi/#external-specifications
<AutomatedTester> whimboo: we already have an issue open to list external specs that are extending webdriver
<gsnedders> q+
<AutomatedTester> ... and we don't have anything for pulling in and generating the cddl specs
<sadym> q
<sadym> q+
<AutomatedTester> ... so we ould need to implement that
<AutomatedTester> ack jimevans
<AutomatedTester> jimevans: to echo what whimboo is saying. As a client side implementor it can be difficult to figure out where things have been implemented
<AutomatedTester> ... this is a problem in the classic and bidi spec. we try maintain the list but this is not always viable
<AutomatedTester> ... I am more inclined to keep things in 1 place to help create better client bindings
<AutomatedTester> q?
<AutomatedTester> ack gsnedders
<AutomatedTester> gsnedders: there is a project called webref that contains things that it knows about the idl/cddl/properties
<AutomatedTester> ... we could speak to them to get this implemented for us
<gsnedders> s|idl/cddl/properties|WebIDL/CSS Properties|
<AutomatedTester> q?
<whimboo> i just filed https://github.com//issues/791 to get our cddl file generation to care about external definitions
<gsnedders> s/get this implemented/get extracting CDDL implemented/
<AutomatedTester> ack sadym
<jgraham> q+
<AutomatedTester> sadym: where is the line for things should live is I think it should be "Is this in classic?" if it is not then it should not live in bidi since it is a new way of doing things and other things should be in their respective homes
<AutomatedTester> ack jgraham
<orkon> q+
<AutomatedTester> jgraham: I think if that is the line then we are already breaking that with network interception and a few others. Practically following that will move it hard for us to complete things... e.g. getting Fetch to do the network intercept stuff
<AutomatedTester> ... our spec has been cases of are they willing to handle the potential extra work load
<AutomatedTester> ... we also have not been good enough of reviewing things. e.g. geolocation have followed permissions who were doing the wrong things with using webidl
<AutomatedTester> ... centralisation has a lot of advantages
<AutomatedTester> ... I don't mind it being in geolocation, it just felt a little easier
<AutomatedTester> q?
<AutomatedTester> ack orkon
<AutomatedTester> orkon: I think would be great if the experts on their spec contribute to the testing part too
<AutomatedTester> ... but I don't have a very strong opinion here... also the tooling problem can be solved so not concerned by that
<AutomatedTester> q?
<orkon> present-
<gsnedders> Filed https://github.com/w3c/webref/issues/1353 for webref, FYI.
<gsnedders> jimevans is mentioning https://github.com/w3c/webauthn/issues/2072 in the room, about the response from nsatragno on the issue, and who should do the work to migrate the WebAuthn stuff to BiDi.
<gsnedders> RRSAgent, make minutes
<RRSAgent> I have made the request to generate https://www.w3.org/2024/09/26-webdriver-minutes.html gsnedders

[css-meeting-bot]

The Browser Testing and Tools Working Group just discussed Geolocation emulation, and agreed to the following:

• ACTION: Spec to be updated to throw an error for display:none [1]
• ACTION: whimboo to write a new issue for display:contents with test case and bring it back to the WG [2]

The full IRC log of that discussion

<tidoust> Topic: Geolocation emulation
<tidoust> github: https://github.com//issues/343
<tidoust> sacha: We would need to pick it up to work on it. Some discussion at TPAC on whether to move things here or in Geolocation spec. It would be good to clarify so we can work on it.
<jgraham> q+
<tidoust> ack jgraham
<tidoust> jgraham: I think we should consider pulling this stuff in our spec.
<tidoust> ... Extending WebDriver BiDi is good, but it has trade-offs that makes it harder to refactor things afterwards.
<orkon> q+
<tidoust> ... We know that emulation is a good feature that we want to have in BiDi and that it's a good idea to have alignment on a model for that.
<tidoust> ... I think there's a demand for such features, easiest way to make progress for me is to work on them here.
<tidoust> ... Of course, if Geolocation folks want to continue to work on this on their own, that's fine.
<tidoust> ack orkon
<jgraham> q+
<tidoust> orkon: A middle-ground here would be useful. Perhaps it makes sense if Geolocation spec provides some mechanism but then CDDL and wiring remains on our side.
<tidoust> ... I didn't review the draft. If it's already good and Geolocation editors want to maintain it, that's good.
<tidoust> ack jgraham
<tidoust> jgraham: We want the actual hooks and algorithms to exist in the spec, I agree.
<tidoust> ... We should talk to Geolocation people, ping that PR. If they can land the hook, we can work on the underlying bits.
<tidoust> ... We don't want to maintain the details of how Geolocation works.
<sasha> q+
<tidoust> ack sasha
<tidoust> sasha: That sounds good to me. I think we have agreement.
<jgraham> q+
<tidoust> ... Follow-up question is where to put it in our modules. It could be in browsingContext but maybe something more modular. Thoughts on this?
<tidoust> ack jgraham
<tidoust> jgraham: I think we should start the emulation module when we add the ability to add to userContext. browsingContext.setViewport. It's just about organizing things nicely.
<orkon> q+
<tidoust> ... There's a bunch of things that we believe that we will have here, e.g., locale override.
<tidoust> ... There's a fairly consistent API that we could end up with.
<tidoust> ... Pretending that you are a mobile device, in a different timezone, etc.
<jimevans> +1 to what jgraham said. "emulation" module for the win.
<tidoust> ack orkon
<jgraham> q+
<tidoust> orkon: An emulation module sounds good. In CDP, there's a problem of the domain. For BiDi, it makes sense to be able to define an alias for commands so that we don't have to duplicate things.
<tidoust> ack jgraham
<tidoust> jgraham: I think we can do that to a very large extent without anything new.
<tidoust> ... "This command uses the parameters from this other place".
<tidoust> ... I think we can make a very small wrapper.
<jgraham> RRSAgent: make minutes
<RRSAgent> I have made the request to generate https://www.w3.org/2025/02/12-webdriver-minutes.html jgraham
<jgraham> s/unambiguously script closable/unambiguously not script closable/
<jgraham> s/html isn't doing it for not script closable/html isn't expecting the script closable flag to apply to user created top-level traversables/
<jgraham> RRSAgent: make minutes
<RRSAgent> I have made the request to generate https://www.w3.org/2025/02/12-webdriver-minutes.html jgraham
<jgraham> RRSAgent: bye
<RRSAgent> I see 2 open action items saved in https://www.w3.org/2025/02/12-webdriver-actions.rdf :
<RRSAgent> ACTION: Spec to be updated to throw an error for display:none [1]
<RRSAgent> recorded in https://www.w3.org/2025/02/12-webdriver-irc#T17-11-43
<RRSAgent> ACTION: whimboo to write a new issue for display:contents with test case and bring it back to the WG [2]
<RRSAgent> recorded in https://www.w3.org/2025/02/12-webdriver-irc#T17-12-03
<jgraham> zakim, bye
<Zakim> leaving. As of this point the attendees have been simons, AutomatedTester, jimevans, orkon, tidoust, sadym, jgraham, chrmod