15:58:41 <RRSAgent> RRSAgent has joined #webdriver
15:58:41 <RRSAgent> logging to https://www.w3.org/2021/04/14-webdriver-irc
16:00:02 <jgraham> Meeting: WebDriver-BiDi
16:00:03 <brwalder> brwalder has joined #webdriver
16:00:28 <AutomatedTester> Chair: AutomatedTester
16:00:30 <jgraham> Agenda: https://www.w3.org/wiki/WebDriver/2021-04-BiDi#Agenda
16:00:37 <AutomatedTester> Scribe: David Burns
16:00:43 <jgraham> Scribe: AutomatedTester
16:00:45 <AutomatedTester> ScribeNick: AutomatedTester
16:01:00 <Honza> Honza has joined #webdriver
16:01:02 <AutomatedTester> present+
16:01:06 <jgraham> present+
16:01:09 <brwalder> present+
16:01:11 <jgraham> RRSAgent: make logs public
16:01:16 <jgraham> RRSAgent: make minutes v2
16:01:16 <RRSAgent> I have made the request to generate https://www.w3.org/2021/04/14-webdriver-minutes.html jgraham
16:01:22 <Honza> present+
16:01:23 <shengfa> shengfa has joined #webdriver
16:01:29 <shengfa> present+
16:01:32 <sadym> present+
16:01:58 <cb> present+
16:02:16 <foolip> present+
16:03:59 <jgraham> RRSAgent: make minutes v2
16:03:59 <RRSAgent> I have made the request to generate https://www.w3.org/2021/04/14-webdriver-minutes.html jgraham
16:04:19 <JohnChen> JohnChen has joined #webdriver
16:04:28 <simonstewart> simonstewart has joined #webdriver
16:04:40 <AutomatedTester> https://www.w3.org/wiki/WebDriver/2021-04-BiDi#Agenda
16:04:40 <simonstewart> present+
16:04:43 <JohnChen> present+
16:06:14 <sadym> https://github.com/w3c/webdriver-bidi/issues/86
16:06:24 <AutomatedTester> topic: Serial Max depth
16:06:30 <AutomatedTester> github https://github.com/w3c/webdriver-bidi/issues/86
16:06:42 <AutomatedTester> github: https://github.com/w3c/webdriver-bidi/issues/86
16:07:05 <AutomatedTester> sadym: I was prototyping serialisation a month ago
16:07:31 <jgraham> q+
16:07:42 <AutomatedTester> ... and I came across this problem. I have wondered if we want a depth of 1
16:08:07 <AutomatedTester> ... jgraham has asked questions before
16:08:12 <AutomatedTester> ack jgraham
16:08:34 <AutomatedTester> jgraham: I think the context here is the remote value serialisation
16:08:50 <AutomatedTester> ... one thing here is that we are doing things differently to CDP
16:09:14 <AutomatedTester> ... in Bidi we serialise everything and return it
16:09:35 <AutomatedTester> ... in cdp you have to do more journeys when you have things like arrays
16:10:02 <AutomatedTester> ... and we want to optimize for the use case where you are not local to the browser
16:10:34 <AutomatedTester> ... and we have cases where there is a large object/array and you only want specific parts of it not the entire thing
16:10:51 <AutomatedTester> ... so the question is how do we address this?
16:11:35 <AutomatedTester> ... the spec has a depth property so that you can specify what we are doing
16:12:17 <AutomatedTester> ... but this can allow people to send too much data... e.g. if it has 2^32-1 properties its a lot to send over the wire
16:12:43 <AutomatedTester> ... and there are ways we can solve this
16:14:20 <AutomatedTester> ... The main thing is that we want to have a choice that is more choice than what CDP has. We want to have a way to control the verbosity
16:14:21 <foolip> q+
16:14:23 <AutomatedTester> q?
16:14:30 <sadym> q+
16:14:35 <AutomatedTester> ack foolip
16:14:46 <AutomatedTester> q?
16:14:52 <AutomatedTester> ack sadym
16:15:08 <AutomatedTester> sadym: There are definite pros for the bidi
16:15:56 <AutomatedTester> ... there are times that we need to do the simplest way
16:16:13 <jgraham> q+
16:16:42 <AutomatedTester> ... in cdp there are ways that we can use it for the next call and optimizes for that
16:16:48 <AutomatedTester> ack jgraham
16:17:23 <AutomatedTester> jgraham: to the idea of using `maxDepth=1` is what makes CDP very chatty
16:17:53 <AutomatedTester> ... and if you have N elements in the array you have to do N calls over the wire to populate it
16:19:05 <AutomatedTester> ... if there are cases that if you know that know you need we return cases where that we return a small subset of concreate items
16:19:37 <foolip> q+
16:19:39 <AutomatedTester> ... and then for events we have specific serialisation algorithms rather than trying to solve this at a global level
16:19:46 <AutomatedTester> ack foolip
16:20:08 <brwalder> q+
16:20:28 <AutomatedTester> foolip: part of the unstated goal here is that it can be racy when on top  of CDP.
16:20:49 <AutomatedTester> ... I guess in the future we may need to have this as a goal in CDP in the future
16:21:10 <brwalder> q-
16:21:11 <jgraham> Doesn't this cause a problem for CDP based clients; you're just pushing the race onto them
16:21:50 <brwalder> q+
16:22:06 <sadym> q+
16:22:11 <jgraham> q+
16:22:29 <AutomatedTester> foolip: I presume this is a problem for them too already. We are not wanting to see cdp change right now but I guess we need to solve that at some point
16:22:34 <AutomatedTester> q?
16:22:45 <AutomatedTester> ack brwalder
16:23:11 <AutomatedTester> brwalder: I don't want this to be an implementation chat but I dont think this is a problem in CDP
16:23:32 <foolip> good point, I keep forgetting about that option, brwalder
16:23:59 <AutomatedTester> ... if we keep a weakmap of things then we could probably get around it
16:24:01 <AutomatedTester> q?
16:24:07 <AutomatedTester> ack sadym
16:24:52 <AutomatedTester> sadym: Most of the time in devtools we send back the ids to the devtools
16:25:05 <AutomatedTester> ... and that's how we avoid that race condition
16:25:06 <AutomatedTester> q?
16:25:11 <AutomatedTester> ack jgraham
16:25:39 <AutomatedTester> jgraham: what's in the spec currently that does serialisation is just events
16:26:03 <AutomatedTester> ... and we can have limits for it
16:26:25 <AutomatedTester> ... so maxDepth=1 is fine but we will need to revisit this during script execution
16:26:30 <AutomatedTester> q?
16:26:44 <AutomatedTester> github: end topic
16:26:56 <AutomatedTester> github-bot: end topic
16:27:16 <AutomatedTester> Topic: Navigation
16:27:21 <AutomatedTester> github: https://github.com/w3c/webdriver-bidi/pull/93
16:27:37 <AutomatedTester> jgraham: This is mostly a status update topic
16:27:56 <drousso> drousso has joined #webdriver
16:28:16 <AutomatedTester> ... the Navigation part of the spec is there that uses Navigable and the events
16:28:19 <drousso> whoops just remembered
16:28:21 <drousso> present+
16:28:22 <drousso> :P
16:28:35 <AutomatedTester> ... so could people please read the PR
16:29:25 <simonstewart> q+
16:29:32 <AutomatedTester> q?
16:30:02 <AutomatedTester> ... it does things that WebDriver and CDP doesnt do things which I think is nicer for end users
16:30:06 <AutomatedTester> ack simonstewart
16:30:10 <foolip> I will review, jgraham. Also, I can only join half of the call today, and will leave now.
16:30:28 <jgraham> s/does things/resolves relative urls/
16:31:31 <AutomatedTester> simonstewart: the reason it's not done in navigate is that I am not clever enough because it's non-trivial
16:31:51 <AutomatedTester> ... most users just want to do one navigation and then user interactions to move around the rest of the site
16:32:05 <AutomatedTester> ... cb may have more anecdotal data on this
16:32:09 <AutomatedTester> q?
16:32:54 <AutomatedTester> github-bot: end topic
16:33:16 <AutomatedTester> topic: Traversing History
16:33:46 <AutomatedTester> jgraham: the next thing after navigation is traversing the history
16:34:04 <AutomatedTester> ... in existing clients there is a page forward/back methods
16:34:37 <AutomatedTester> ... however CDP doesnt have what normal clients have. It has the ability to move to specific IDs
16:34:51 <AutomatedTester> ... so do we want to standardize on the clients or what on CDP
16:35:17 <AutomatedTester> ... I think that CDP is a lot more flexible compared to our clients currently
16:35:29 <AutomatedTester> q?
16:37:13 <AutomatedTester> AutomatedTester: while I think CDP sounds more flexible but in my experience people rarely want to move more than 1 or 2 pages(max) and 2 calls to `back` should be fine
16:37:44 <simonstewart> q+
16:38:25 <AutomatedTester> jgraham: I can image a case where the back button could populate the history when clicked and it creates an infinite loop
16:38:45 <AutomatedTester> ... so do we want to standardize on what all the clients do it
16:39:01 <AutomatedTester> .. or allow more nuanced APIs
16:39:36 <sadym> q+
16:39:43 <AutomatedTester> ... I am in favor of doing what clients do right now as that is the easier option
16:39:49 <AutomatedTester> ack simonstewart
16:40:08 <AutomatedTester> simonstewart: I support the current clients as it's simplest way
16:40:41 <AutomatedTester> ... in the original spec we focused on audiences
16:40:56 <AutomatedTester> ... the simpler model works well for testers
16:41:03 <AutomatedTester> ... but what about spec authors
16:41:36 <AutomatedTester> jgraham: I think that is valid but it would be good to get the history spec to work in the webdriver way
16:41:39 <AutomatedTester> q?
16:41:47 <AutomatedTester> ack sadym
16:42:27 <AutomatedTester> sadym: I can think of one scenario that needs the complex version but we can add that later if we really need it
16:43:05 <AutomatedTester> topic: Bidi only sessions
16:43:08 <AutomatedTester> github: https://github.com/w3c/webdriver-bidi/pull/99
16:43:34 <AutomatedTester> jgraham: we have been standing up the initial implementation in Gecko
16:44:11 <AutomatedTester> ... one of the questions that came up was "should we allow people to start a session without going via HTTP session"
16:44:39 <AutomatedTester> ... its for clients that never care about the http commands that webdriver has
16:44:58 <AutomatedTester> ... and then there are existing tools like puppeteer/playwright
16:45:11 <brwalder> q+
16:45:23 <AutomatedTester> ... from Mozilla we don't really want to ship a http server in Firefox
16:45:56 <AutomatedTester> ... or we can rewrite the webdriver http spec to show how this maps across to webdriver bidi
16:46:19 <AutomatedTester> ... and I have created a PR on how this could work
16:46:55 <AutomatedTester> ... [ describes what is in the PR ]
16:47:30 <AutomatedTester> q?
16:47:34 <AutomatedTester> ack brwalder
16:48:30 <simonstewart> q+
16:49:03 <drousso> q+
16:49:16 <jgraham> q+
16:50:19 <AutomatedTester> brwalder: in chromium we ahve a link between webdriver and browser processes that is 1:1. This is done when things are started up and we go from there. I am not against connecting to an existing browser just mentioning it
16:50:35 <AutomatedTester> q?
16:50:39 <AutomatedTester> ack simonstewart
16:51:27 <brwalder> q+
16:51:45 <AutomatedTester> simonstewart: I don't mind, from selenium, that we connect to a websocket but we need a sessionID so we can make sure that intermediary nodes route things to the right place
16:52:14 <AutomatedTester> q?
16:52:30 <AutomatedTester> ack drousso
16:53:55 <AutomatedTester> drousso: I don't know if we want to expose the websocket because of security/privacy concerns
16:54:00 <AutomatedTester> q?
16:54:05 <AutomatedTester> ack jgraham
16:54:57 <AutomatedTester> jgraham: I don't think we want people to have start up the standard browser and there is websocket open as that is a security nightmare
16:55:25 <AutomatedTester> ... e.g you would need to tell the browser to start up with a websocket via a driver
16:56:39 <AutomatedTester> ... my question is mostly around do we need to do a start a session by calling HTTP POST /SESSION first or connect to a websocket first
16:57:19 <AutomatedTester> ... and to brwalder comment I think I have already handled the use case you've mentioned
16:57:37 <AutomatedTester> ... [describes a use case for 2 tools running]
16:58:13 <AutomatedTester> ... which won't work in the original webdriver because of the blocking commands there
16:59:42 <brwalder> q-
17:00:31 <AutomatedTester> github-bot: end topic
17:01:05 <jgraham> RRSAgent: make minutes v2
17:01:05 <RRSAgent> I have made the request to generate https://www.w3.org/2021/04/14-webdriver-minutes.html jgraham
17:01:36 <AutomatedTester> https://github.com/web-platform-tests/wpt/pull/28381
20:17:43 <zcorpan> zcorpan has joined #webdriver
20:18:42 <zcorpan_> zcorpan_ has joined #webdriver
20:20:33 <zcorpan> zcorpan has joined #webdriver
20:22:22 <zcorpan_> zcorpan_ has joined #webdriver
20:23:30 <zcorpan> zcorpan has joined #webdriver
20:23:43 <Zakim> Zakim has left #webdriver
20:24:50 <jgraham> RRSAgent: bye
20:24:50 <RRSAgent> I see no action items