Skip to main content

​
Overview

Scraper Studio IDE uses two types of code to build web scrapers:
  1. Interaction Code - Controls browser automation and navigation
  2. Parser Code - Extracts and structures data from HTML
Each type uses different technologies and serves different purposes.

​
IDE Interaction code

​
What It Does

Interaction code controls a real browser session to:
  • Navigate to URLs
  • Wait for elements to load
  • Click buttons and links
  • Fill out forms
  • Scroll pages
  • Handle dynamic content (JavaScript-heavy sites)
You can use jQuery-like expressions in interaction code—CSS selector patterns and methods used to find and manipulate HTML elements. Here you can find all available functions you can use in the IDE: input - Global object available to the interaction code. Provided by trigger input or next_stage() calls 
navigate(input.url);
navigate - Navigate the browser session to a URL
  • url: A URL to navigate to
navigate([url]);
navigate(input.url);
navigate('https://example.com');
navigate options 
// waits until DOM content loaded event is fired in the browser
navigate([url], {wait_until: 'domcontentloaded'});

// adds a referer to the navigation
navigate([url], {referer: [url]});

// the number of milliseconds to wait for. Default is 30000 ms
navigate([url], {timeout: 45000});

// add headers to the navigation
navigate([url], {header : 'accept: text/html'});

// specify browser width/height
navigate([url], {fingerprint: {screen: {width: 400, height: 400}}});
parse - Parse the page data 
let page_data = parse();
collect - Adds a line of data to the dataset created by the crawler
  • data_line: A object with the fields you want to collect
  • validate_fn: Optional function to validate that the line data is valid
collect([data_line] [validate_fn]);
collect({ title: page_data.title price: page_data.price });
collect({ price: data.price });

collect(line, l=>!l && throw new Error('Empty line'));
next_stage - Run the next stage of the crawler with the specified input
  • input: Input object to pass to the next browser session
next_stage({url: 'http://example.com', page: 1});
rerun_stage - Run this stage of the crawler again with new input
  • input: Input object to pass to the next browser session
rerun_stage({url: 'http://example.com/other-page'});
run_stage - Run a specific stage of the crawler with a new browser session
  • input: Input object to pass to the next browser session
  • stage: Which stage to run (1 is first stage)
run_stage(2, {url: 'http://example.com', page: 1});
country - Configure your crawl to run from a specific country
  • code: 2-character ISO country code
country(<code>);
wait - Wait for an element to appear on the page
  • selector: Element selector
  • opt: wait options (see examples)
wait (<selector>);
wait('#welcome-splash');
wait('.search-results .product');
wait('[href^='/product']');
wait(<selector>, {timeout: 5000});
wait(<selector>, {hidden: true});
wait_for_text - Wait for an element on the page to include some text
  • selector: Element selector
  • text: The text to wait for
wait_for_text(<selector>, <text>);
wait_for_text('.location', 'New York');
click - Click on an element (will wait for the element to appear before clicking on it)
  • selector: Element selector
click(<selector>);
click('#show-more');
type - Enter text into an input (will wait for the input to appear before typing)
  • selector: Element selector
  • text: The text to wait for
type(<selector>, <text>);
type('#location', 'New York');
type(<selector>, ['Enter']);
type(<selector>, ['Backspace']);
select - Pick a value from a select element
  • selector: Element selector
select(<select>, <value>);
select('#country', 'Canada');
URL - URL class from NodeJS standard “url” module
  • url: URL string
let u = new URL('https://example.com');
location - Object with info about current location. Available fields: href
  • url: URL string
navigate('https://example.com');
location.href;
tag_response - Save the response data from a browser request
  • name: The name of the tagged field
  • pattern: The URL pattern to match
tag_response(<field>, <pattern>);
tag_response('teams', /\/api\/teams/);
navigate('https://example.com/sports');

let teams = parse().teams;
for (let team of teams) collect(team);
response_header - Returns the response headers of the last page load 
let headers = response_headers(); 
console.log('content-type', headers['content-type']);
console - Log messages from the interaction code 
console.log(1, 'luminati', [1, 2], {key: value});
load_more - Scroll to the bottom of a list to trigger loading more items. Useful for lazy-loaded infinite-scroll sites
  • selector: Element selector
load_more(<selector>);
load_more('.search-results');
scroll_to - Scroll the page so that an element is visible 
scroll_to(<selector>);
scroll_to('.author-profile');
$ - Helper for jQuery-like expressions
  • selector: Element selector
$(<selector>);
wait($('.store-card'))

​
IDE Parser code

​
Overview

Parser code is responsible for extracting and structuring data from HTML content. Scraper Studio Parser code uses the pre-installed Cheerio library—a library that provides jQuery-like syntax for parsing HTML documents. $ - An instance of cheerio 
$('#example').text()
$('$example').attr('href')
Find more information on the cheerio website.
location - A global variable available to the parser code. An object containing information about the current location. 
let current_url = location.href;