Roadblocks and Progress in Playwright’s WebDriver BiDi Adoption

[yury-s]

Over the past few weeks, we've been working on integrating BiDi support into Playwright and running our test suite using the new protocol. We're pleased to see significant progress in BiDi implementation for both Firefox and Chromium, with many Playwright tests now passing. However, we still encounter a considerable number of issues preventing Playwright from fully adopting BiDi as the browser automation protocol. Given the volume of bugs we've identified, we're consolidating some of the most critical issues in a single report instead of filing them individually: microsoft/playwright#32577

We invite BiDi collaborators to review and help resolve these challenges. You can find instructions for running the tests here. The BiDi-specific implementation is located at this path. If anyone is interested in taking an active role in development of this component to drive quicker iterations, we are open to arranging that.

[whimboo]

@yury-s thanks a lot for the extensive list of issues over on the Puppeteer tracker. To have a better way to modify content and add further comments and references I've created a Google document that we all can collaboratively work on.

I've started to go through the issue list and will add discussion items for the upcoming meetings at TPAC of the browser testing and tools working group at: https://www.w3.org/wiki/WebDriver/2024-09-TPAC#Agenda

[whimboo]

@yury-s could you maybe set priorities (even roughly) to each of the items? It would be good to know from your perspective what's the most and least important items. I already added the [P?] prefix for each spec item in the document, so maybe you can fill out? Just request access for you and your colleagues and I'll grant edit permissions to each of you. Thanks!

[whimboo]

@yury-s during TPAC we added new items to our BiDi roadmap document. If you can find the time to check column G and add even a rough prioritization for Playwright that would be great. Thanks!

[yury-s]

Hi Henrik, thank you for your interest in this issue.

It seems we might have been looking at things from slightly different perspectives when sharing our thoughts earlier. Our intention was to provide a holistic view of the BiDi spec and highlight some of the larger gaps that would need to be addressed for potential Playwright adoption. For example, we can't even think about using BiDi unless the entirely of the emulation is addressed on the user context level, all the items in the list are critical blockers for running our tests.

So, rather than seeing this list as "once these items are fixed, Playwright can adopt BiDi," it's more of a high-level guide on where we think attention is needed to ensure BiDi becomes fully capable for modern web testing.

Having said that, we welcome you to the Playwright repo to experiment and proceed with the BiDi port implementation as BiDi progresses to unblock the issues and will provide any help necessary to keep the collaboration efficient!

[AutomatedTester]

@yury-s you and the rest of the Playwright team are always welcome to come join W3C Browser Testing and Tools WG so that we can improve web testing for all end users.

We met last week at the W3C TPAC you can see the minutes in

• Day 1
• Day 2

Our next meeting in on 9 October and we will be tracking the agenda on the W3C Wiki.

A rising tide lifts all ships and this is why we are making sure that everyone can be part of the process.