Look for class="article" for content

Quite a few sites use this, but fail the current
heuristic.
This commit is contained in:
Martin Richtarsky
2024-03-03 21:52:14 +09:00
committed by josc146
parent 0d7d7354d8
commit 08a8920661
+10 -5
View File
@@ -43,23 +43,28 @@ function findLargestElement(e) {
}
export function getCoreContentText() {
function getTextFrom(e) {
return e.innerText || e.textContent
}
for (const [siteName, selectors] of Object.entries(adapters)) {
if (location.hostname.includes(siteName)) {
const element = getPossibleElementByQuerySelector(selectors)
if (element) return element.innerText || element.textContent
if (element) return getTextFrom(element)
break
}
}
const element = document.querySelector('article')
if (element) {
return getTextFrom(element)
}
const largestElement = findLargestElement(document.body)
const secondLargestElement = findLargestElement(largestElement)
console.log(largestElement)
console.log(secondLargestElement)
function getTextFrom(e) {
return e.innerText || e.textContent
}
let ret
if (!largestElement) {
ret = getTextFrom(document.body)