Files
parcer/mx-interpreter/browserSide/scraper.js
2024-07-16 00:25:47 +05:30

53 lines
1.6 KiB
JavaScript

/* eslint-disable @typescript-eslint/no-unused-vars */
const area = (element) => element.offsetHeight * element.offsetWidth;
function getBiggestElement(selector) {
const elements = Array.from(document.querySelectorAll(selector));
const biggest = elements.reduce(
(max, elem) => (
area(elem) > area(max) ? elem : max),
{ offsetHeight: 0, offsetWidth: 0 },
);
return biggest;
}
/**
* Generates structural selector (describing element by its DOM tree location).
*
* **The generated selector is not guaranteed to be unique!** (In fact, this is
* the desired behaviour in here.)
* @param {HTMLElement} element Element being described.
* @returns {string} CSS-compliant selector describing the element's location in the DOM tree.
*/
function GetSelectorStructural(element) {
// Base conditions for the recursive approach.
if (element.tagName === 'BODY') {
return 'BODY';
}
const selector = element.tagName;
if (element.parentElement) {
return `${GetSelectorStructural(element.parentElement)} > ${selector}`;
}
return selector;
}
/**
* Heuristic method to find collections of "interesting" items on the page.
* @returns {Array<HTMLElement>} A collection of interesting DOM nodes
* (online store products, plane tickets, list items... and many more?)
*/
function scrapableHeuristics(maxCountPerPage = 50, minArea = 20000, scrolls = 3, metricType = 'size_deviation') {
const restoreScroll = (() => {
const { scrollX, scrollY } = window;
return () => {
window.scrollTo(scrollX, scrollY);
};
})();
/**
* @typedef {Array<{x: number, y: number}>} Grid
*/