**Is your feature request related to a problem?
We've just lost time to understand why our tests could not execute on some agents. It appears that node on those agents was still old 12 version, while playwright-fluent needs node 14 or more.

Describe the solution you'd like
Add an engine constraints in package.json to get a relevant message when an old node version is used.
engines check description

Is your feature request related to a problem? Please describe.
Yes to check dialog or alert I have to switch on pure Playwright.
Describe the solution you'd like
a few more api functions to expect dialog or alert with expected text and a few functions to accept and dismiss them.
Describe alternatives you've considered
Currently I have to call currentPage() and use page.on('dialog', ...) but with TS it make me handle optional chaining) due to return type of currentPage().
Additional context
I am trying to wright tests without pure Playwright because it allows me to create very readable code plus Playwright-Fluent acts more like real user. I really like use this wrapper.

Is your feature request related to a problem? Please describe.
I found that next assert: await playwrightFluent.expectThatSelector(SELECTOR).isVisible();
Will fail if element is out of viewport. But in most cases I need to check is element is visible any were and then I can scroll to it,
also this behaviour with viewport is not mentioned in method description.

Describe the solution you'd like
I would like to have two methods: isVisible() - checks if element visible in general and isVisibleInViewport() - which works as existing one.
Describe alternatives you've considered
isVisible() - works as it is. isDisplayed() - checks if element visible in general.

Discussed in #27

Is your feature request related to a problem? Please describe.
I would like to select an item by value and not by label

Describe the solution you'd like
Having a selectByValue that works like select

Is your feature request related to a problem? Please describe.
Currently to turn on tracing feature I need to do this via pure playwright. My code:
const story = playwrightFluent
.withBrowser(props.browser)
.withOptions({headless: props.isHeadless, slowMo: 50, args: getBrowserArgs(props.browser)})
.withDefaultWaitOptions({stabilityInMilliseconds: 50, timeoutInMilliseconds: PLAYWRIGHT_TIMEOUT})
.withDefaultAssertOptions({stabilityInMilliseconds: 100, timeoutInMilliseconds: 3000})
.recordNetworkActivity({path: props.harPath})
.WithDialogs();
await getPage(story).context().tracing.start({screenshots: true, snapshots: true});

Describe the solution you'd like
Please add .withTrace() and stopTracing() methods to api and let them use original parameters.

Is your feature request related to a problem? Please describe.
I use raw playwright to test service with several iframes, I would like to use Playwright-fluent but when I get frame content I can not continue use fluent api.

Describe the solution you'd like

1. Methods to switch to frame and back.
2. PlaywrightFluent constructor can accept Page | Frame.

Is your feature request related to a problem? Please describe.
Starting a browser for each test takes time. We could make a test campaigns faster by reusing single browser instance.

Describe the solution you'd like
I would like to use one and only one browser instance and to create a context for each test.
However this could have pitfalls for those who try to use parallelism. We should have the choice between using a single instance or not.

BR

Franck

add Double Click Event

Describe the bug
In case when expectThatDialog fail due to assert then dialog window blocks browser close
To Reproduce
Steps to reproduce the behavior:

Actual script:
await playwright
.waitForDialog()
.expectThatDialog()
.hasMessage(error) // This line throw assert as expected
.acceptDialog();

// first hook after scenario
After(async function (this: ScenarioContext) {
if(this.playwright.isDialogOpened()){
await this.playwright.cancelDialog(); // This line close dialog
}
})

// second hook
After(async function (this: ScenarioContext, test: ITestCaseHookParameter) {
await this.attach(await scenarioContext.playwright.takeFullPageScreenshotAsBase64(), 'base64:image/png'); // This line throws "page.screenshot: Timeout 20000ms exceeded. at Object.takeFullPageScreenshotAsBase64"
await this.playwright.close();
})

Expected behavior
This should work)

Desktop (please complete the following information):

• OS: OSX
• Browser chromium

Thank you in advance for your efforts.

Is your feature request related to a problem? Please describe.
Playwright allows to record video and network logs (HAR), those files save when context closes.
Describe the solution you'd like
It would be great to have api which allows get path to saved video and HAR files after context is closed.

Describe the bug
withBrowser('chrome') do not work in docker.
To Reproduce
Environment - Dockerfile:
FROM mcr.microsoft.com/playwright:focal

WORKDIR tests

COPY package.json /tests

RUN npm install
RUN npx playwright install chrome --with-deps

COPY cucumber.ts /tests
COPY tsconfig.json /tests
COPY src /tests/src

ENTRYPOINT ["npm", "run", "test:pipeline"]

Script:
playwright
.withBrowser('chrome')
.withOptions({headless: props.isHeadless, slowMo: 50})
.withDefaultWaitOptions({stabilityInMilliseconds: 50, timeoutInMilliseconds: PLAYWRIGHT_TIMEOUT})
.withDefaultAssertOptions({stabilityInMilliseconds: 100, timeoutInMilliseconds: 3000})
.WithDialogs();

ERROR: browserType.launch: Failed to launch chromium because executable doesn't exist at google-chrome-stable

But it works with next script:
async function main() {
const browser = await playwright['chromium'].launch({
channel: 'chrome'
});
const page = await browser.newPage();
await page.goto('https://google.com');
await browser.close();
console.log("DONE!")
};
main();

Expected behavior
Both scripts should work with chrome

Additional context
Both scripts works as expected on my mac book pro with Big Sur.
Also I found that I can not add 'channel' property to withOptions(...).

Hi! This looks like a great project. I'm curious to know why the package is called playwright-controller. I'm new to the Fluent pattern, and I'm probably missing something about the role of "controller"