pageText (article detection and content capture) not working on Chrome

[rhelmer]

pageText tries to pass filters to browser.tabs.onUpdated.addListener which is Firefox-only, filtering needs to be done in the callback instead.

[rhelmer]

OK, so Firefox has an isArticle filter that this is using specifically:

https://developer.mozilla.org/en-US/docs/Mozilla/Add-ons/WebExtensions/API/tabs/onUpdated#isarticle

isArticle Optional
boolean. True if the tab is an article and is therefore eligible for display in Reader Mode.

I think we'll need to instead handle this in the callback, and use the Readability library (which is a standalone version of the code that powers Firefox's reader mode, and we already ship it in beyond-the-paywall. There's an npm package)

[rhelmer]

OK, so Firefox has an isArticle filter that this is using specifically:

https://developer.mozilla.org/en-US/docs/Mozilla/Add-ons/WebExtensions/API/tabs/onUpdated#isarticle

isArticle Optional
boolean. True if the tab is an article and is therefore eligible for display in Reader Mode.

I think we'll need to instead handle this in the callback, and use the Readability library (which is a standalone version of the code that powers Firefox's reader mode, and we already ship it in beyond-the-paywall. There's an npm package)

Specifically the isProbablyReaderable() function.

[rhelmer]

Small correction - we can't directly get the page contents from the listener (which is in the background script/worker), so instead I put the probablyIsReaderable() call in the content script, which does have direct access to the DOM of the web page. It does an early-return and does not send a message if this call is false.